WO2025251246A1

WO2025251246A1 - Access point configuration

Info

Publication number: WO2025251246A1
Application number: PCT/CN2024/097713
Authority: WO
Inventors: Quan Gan; Xiaoying Li; Fei ZHAO; Weiqiang Jiang
Original assignee: Telefonaktiebolaget LM Ericsson AB
Current assignee: Telefonaktiebolaget LM Ericsson AB
Priority date: 2024-06-06
Filing date: 2024-06-06
Publication date: 2025-12-11
Anticipated expiration: 2026-12-06

Abstract

There is provided a computer-implemented method for configuring at least one first access point (AP) device of a telecommunications network. The at least one first AP device is comprised in a first environment of the network. The orientation of the at least one first AP device is remotely configurable. The method comprises obtaining (102), from at least one first wireless device of the network, first information indicative of a network measurement associated with the at least one first wireless device. The method also comprises obtaining (104), from the at least one first AP device, second information indicative of a status of the at least one first AP device. The method also comprises determining (106), using a machine learning (ML) model, an orientation configuration of the at least one first AP device based on the first information, the second information, third information, and fourth information. The third information is indicative of a target state of the network, and the fourth information is indicative of one or more characteristics of the first environment.

Description

ACCESS POINT CONFIGURATION

TECHNICAL FIELD

The present disclosure relates to methods for configuring at least one first access point device of a telecommunications effort, methods for training a machine learning model to configure the at least one first access point, entities configured to operate in accordance with those methods, and a docking apparatus.

BACKGROUND

In the era of mobile internet, wireless network coverage is crucial for both ordinary consumers and business users. The fundamental requirement for all consumers and users is to have consistent network connectivity whether at home, on the move, or at work. This is especially true in scenarios which involve wireless terminal devices on production lines in enterprises, which need to ensure round-the-clock network connectivity within a designated area. Such wireless signal coverage is primarily achieved through various types of antenna devices.

In fourth generation (4G) and fifth generation (5G) networks, commonly seen antenna devices include, but are not limited to, macro station antennas, micro station antennas, and small cell antennas. These antennas have different coverage ranges and characteristics, suitable for various environments and scenarios. For instance, macro station antennas are typically installed at higher positions for wide-area coverage, while small cell antennas are typically used in densely populated urban areas to provide smaller-scale, higher-density coverage.

For indoor environments, indoor antennas are designed to address the unique challenges posed by indoor settings. These indoor antennas, often smaller in size and less obtrusive in design than most other antennas, are tailored to provide effective coverage within buildings where external signals may be weak or blocked. Indoor antennas are adept at navigating around physical obstructions like walls, furniture, or production line machines, ensuring that a broadcasted signal reaches every corner of the space. Such antennas are crucial in places like offices, shopping malls, and large indoor venues like factories, where maintaining a strong and reliable wireless connection is important for both operational efficiency and user satisfaction.

The types and features of outdoor antenna products are diverse, and their installation methods also vary according to the relevant scenario. For example, beamforming technology, which adjusts the phase of signals in an antenna array to focus the main lobe in a specific direction, enhances signal quality and coverage range. This technology is widely used in combination with large outdoor antennas, particularly in 5G networks, where beamforming can play a crucial role in increasing network throughput and reducing interference.

In contrast, indoor coverage antenna products typically employ omnidirectional antenna designs. Omnidirectional antennas emit and receive signals uniformly in all directions, which is highly effective for signal coverage in indoor environments. Due to the smaller size of indoor spaces, and the presence of various physical obstacles, omnidirectional antennas ensure coverage in all directions within a certain range. However, these antennas usually do not use beamforming technology, as the complexity and variability of indoor environments make it difficult to leverage the advantages of beamforming, and the cost of doing so is relatively high.

In the design of indoor coverage schemes, factors like the installation location and orientation (e.g. azimuth and elevation angle) of antennas are typically considered. For example, in indoor environments like factory workshops, a network designer will typically determine optimal antenna positions based on building layout, frequency of use, and user/device density. During construction, various measurements and tests may be conducted, such as signal strength testing and coverage range testing, to ensure that the wireless coverage meets the design requirements.

The physical form of indoor antennas is generally compact, allowing them to be mounted on walls or ceilings. The installation process is relatively simple but needs to consider the orientation of the antenna and interference with other devices. In wireless network construction, interference and weak coverage are common issues. Interference can come from other wireless devices, electronic equipment, and even the building itself. Mitigating such interference often involves adjusting the position of antennas, improving antenna design, or using more efficient antenna technology. Weak coverage is typically addressed by increasing the number of antennas or changing the layout of antennas.

Thus, antenna technology is vital for wireless network coverage. Different scenarios require different types of antenna products. Outdoor environments usually utilise high-performance antenna technologies like beamforming, while indoor environments favour omnidirectional antennas to achieve balanced coverage. The different designs of indoor and outdoor antennas reflect their adaptation to various physical environments. With proper design and installation, these antennas can provide stable and efficient wireless network coverage in a variety of settings.

However, as 5G networks increasingly penetrate more indoor scenarios, traditional wireless network engineering methods, and the coverage characteristics of omnidirectional antennas, face new challenges in meeting the complex demands of these environments.

SUMMARY

As mentioned above, there are certain challenges associated with existing techniques for providing high-quality continuous network coverage. Indeed, the rise of both 5G and Industry 4.0 has significantly increased the demands placed on network coverage, especially in industrial settings. When integrated into production environments, 5G networks must deliver on their core strengths: low latency and high reliability. However, the constant evolution of the internet and ongoing industrial revolution require businesses to adapt to market demands with greater agility. Gone are the days of static production lines operating for years unchanged. Modern enterprises need to be nimble, adjusting production lines and capacity on the fly to meet customer needs. This flexibility presents a challenge to traditional fixed-area network coverage, demanding continuous network optimisation to accommodate these dynamically changing requirements.

Currently, network adjustments rely heavily on manual intervention, particularly for intricate communication networks that pose technical hurdles for most businesses. At their best, network designers can engage in limited "pre-planning" . When these measures prove to be lacking, they must resort to external network installation or optimisation specialists. Unfortunately, this approach often involves lengthy on-site adjustments and configuration by dedicated personnel, taking days or even weeks to complete. These delays significantly hinder production efficiency and progress, while introducing additional security risks for companies, especially for "dark factories" with strict access controls.

It is an object of the disclosure to obviate or eliminate at least some of these challenges associated with existing techniques.

Therefore, according to an aspect of the disclosure, there is provided a first method for configuring at least one first access point (AP) device of a telecommunications network. The first method is computer-implemented. The at least one first AP device is comprised in a first environment of the network. The orientation of the at least one first AP device is remotely configurable. The first method comprises obtaining, from at least one first wireless device of the network, first information indicative of a network measurement associated with the at least one first wireless device. The first method also comprises obtaining, from the at least one first AP device, second information indicative of a status of the at least one first AP device. The first method further comprises determining, using a machine learning (ML) model, an orientation configuration of the at least one first AP device based on the first information, the second information, third information, and fourth information. The third information is indicative of a target state of the network. The fourth information is indicative of one or more characteristics of the first environment.

In some examples, the first method may comprise initiating transmission of fifth information indicative of the orientation configuration towards the at least one first AP device. In some examples, the fifth information may comprise a request for the at least one first AP device to adjust an orientation of the at least one first AP device based on the fifth information.

In some examples, the orientation configuration may comprise an angle of rotation for the at least one first AP device, and/or an angle of tilt for the at least one first AP device.

In some examples, the first information may comprise one or more key performance indicators (KPIs) associated with the at least one first wireless device. In some examples, the measurement associated with the at least one first wireless device can be associated with one or more applications running on the at least one first wireless device.

In some examples, the network measurement associated with the at least one first wireless device may comprise one or more of a reference signal received power (RSRP) value, a serving cell RSRP (SS-RSRP) value, a reference signal received quality (RSRQ) value, a serving cell RSRQ (SS-RSRQ) value, a signal to interference and noise ratio (SINR) value, an uplink rate, a downlink rate, a latency value, and a jitter value. In some examples, the second information may comprise one or more KPIs associated with the network.

In some examples, determining the orientation configuration of the at least one first AP device may comprise generating, using the ML model, a simulation of one or more possible orientation configurations of the at least one first AP device, and selecting the orientation configuration from the one or more possible orientation configurations based on the target state of the network. In some examples, the orientation configuration is selected from the one or more possible orientation configurations if the orientation configuration is determined to satisfy the target state of the network. In some examples, the generated simulation can be a digital twin model of the first environment. In some examples, generating the simulation is based at least in part on the fourth information. In some examples, the first method may comprise generating a visual representation of the generated simulation.

In some examples, the fourth information may comprise information indicative of a physical layout of the first environment, a location of the at least one first AP device in the first environment, a location of the at least one first wireless device in the first environment, and/or a location of one or more objects in the first environment.

In some examples, the second information may comprise information indicative of one or more of: an orientation of the at least one first AP device, an angle of rotation of the at least one first AP device, an angle of tilt of the at least one first AP device, an identifier of the at least one first AP device, sensor data associated with the at least one first AP device, an action performed by the at least one first AP device, a power status of the at least one first AP device, and a result of a previously executed configuration of the at least one first AP device.

In some examples, the target state of the network may comprise one or more of a target network coverage for at least part of the first environment, and a target network signal strength for the at least one first wireless device.

In some examples, the first environment is an indoor environment. In some examples, the first environment may comprise an industrial facility, and/or an enterprise facility.

In some examples, the at least one first AP device may comprise an indoor antenna.

In some examples, the ML model may comprise a deep Q-network (DQN) . In some examples, the ML model can be a trained ML model. In some examples, the ML model is trained using the second method as described herein.

According to another aspect of the disclosure, there is provided a second method for training a ML model for configuring at least one first AP device of a telecommunications network. The second method is computer-implemented. The at least one first AP device is comprised in a first environment of the network. The orientation of the at least one first AP device is remotely configurable. The second method comprises training a ML model to determine an orientation configuration of the at least one first AP device based on first information, second information, third information, and fourth information. The first information is indicative of a network measurement associated with the at least one first wireless device. The second information is indicative of a status of the at least one first AP device. The third information is indicative of a target state of the network. The fourth information is indicative of one or more characteristics of the first environment.

In some examples, the ML model can be trained using a first training dataset. The first training dataset may comprise information obtained from a test environment. Alternatively, or in addition, the ML model can be trained using a second training dataset. The second training dataset may comprise information obtained from the first environment. In some examples, the test environment and the first environment are different.

In some examples, the first training dataset may comprise information indicative of a location of at least one second AP device, an orientation configuration of the at least one second AP device, a plurality of network measurements associated with at least one second wireless device, sensor data associated with the test environment, and/or a physical layout of the test environment. The at least one second AP device can be comprised in the test environment. The at least one second wireless device can be comprised in the test environment.

In some examples, the plurality of network measurements associated with the at least one second wireless device may be performed at a plurality of locations in the test environment. In some examples, the plurality of network measurements associated with the at least one second wireless device may comprise one or more key performance indicators (KPIs) associated with the at least one second wireless device. In some examples, the plurality of network measurements associated with the at least one second wireless device can be associated with one or more applications running on the at least one second wireless device.

In some examples, the plurality of network measurements associated with the at least one second wireless device comprise one or more of a reference signal received power (RSRP) value, a serving cell RSRP (SS-RSRP) value, a reference signal received quality (RSRQ) value, a serving cell RSRQ (SS-RSRQ) value, a signal to interference and noise ratio (SINR) value, an uplink rate, a downlink rate, a latency value, and a jitter value.

In some examples, the orientation configuration of the at least one second AP device may comprise an angle of rotation for the at least one second AP device, and/or an angle of tilt for the at least one second AP device.

In some examples, the second training dataset may comprise information indicative of a location of at the least one first AP device, an orientation configuration of the at least one first AP device, a plurality of network measurements associated with the at least one first wireless device, sensor data associated with the first environment, and/or a physical layout of the first environment.

In some examples, the plurality of network measurements associated with the at least one first wireless device may comprise one or more KPIs associated with the at least one first wireless device. In some examples, the plurality of network measurements associated with the at least one first wireless device can be associated with one or more applications running on the at least one first wireless device.

In some examples, the plurality of network measurements associated with the at least one first wireless device comprises one or more of a reference signal received power (RSRP) value, a serving cell RSRP (SS-RSRP) value, a reference signal received quality (RSRQ) value, a serving cell RSRQ (SS-RSRQ) value, a signal to interference and noise ratio (SINR) value, an uplink rate, a downlink rate, a latency value, and a jitter value.

In some examples, the orientation configuration of the at least one first AP device may comprise an angle of rotation for the at least one first AP device, and/or an angle of tilt for the at least one first AP device. In some examples, the at least one first AP device and the at least one second AP device may comprise the same AP device. In some examples, the at least one first AP device and the at least one second AP device may comprise at least one indoor antenna. In some examples, the ML model may comprise a deep Q-network, DQN.

According to another aspect of the disclosure, there is provided a docking apparatus for an indoor antenna. The apparatus comprises a first portion, a second portion, a third portion, and a controller. The first portion comprises first coupling means configured to couple the apparatus to a surface. The third portion comprises second coupling means configured to couple the apparatus to the indoor antenna. The second portion is coupled to the first portion and the third portion. The controller is remotely configurable to rotate the second portion and the third portion relative to the first portion about a first axis of rotation, and tilt the third portion relative to the second portion.

In some examples, the second portion may comprise a second surface and a third surface. In some examples, the second surface and the third surface can be planar surfaces. In some examples, the third surface may be inclined relative to the second surface. In some examples, a first angle between the second surface and the third surface may be 130° to 150°. In some examples, the first angle can be 140°.

In some examples, the controller may be remotely configurable to tilt the third portion from a first tilt configuration to a second tilt configuration. Tilting the third portion from the first tilt configuration to the second tilt configuration may comprise moving a first section of the third portion towards the third surface, and moving a second section of the third portion away from the second surface. In some examples, in the first tilt configuration, a first surface of the third portion can be substantially parallel to the second surface. In some examples, in the second tilt configuration, the first surface can be substantially parallel to the third surface. In some examples, the third portion may be in contact with the second portion in the first tilt configuration and the second tilt configuration.

In some examples, the first portion may comprise a first opening, the second portion may comprise a second opening, and the third portion may comprise a third opening. In some examples, the first opening and the second opening may be positioned on the first axis.

In some examples, the controller may be configured to communicate with a network. In some examples, the docking apparatus may comprise a subscriber identity module (SIM) card port. In some of these examples, the controller may be configured to communicate with the network using a SIM card.

In some examples, the controller may be configured to perform a rotation operation and/or a tilting operation based on orientation configuration information received via the network. In some examples, the network can be a telecommunications network. In some examples, the second portion may be coupled to the third portion by one or more hinges. In some examples, the second portion may be coupled between the first portion and the third portion.

According to another aspect of the disclosure, there is provided an AP device. The AP device comprises an indoor antenna, and the docking apparatus as described herein.

According to another aspect of the disclosure, there is provided a first entity comprising processing circuitry configured to operate in accordance with the first method referred to herein, and/or the second method referred to herein. In some embodiments, the first entity may comprise at least one memory for storing instructions which, when executed by the processing circuitry, cause the first entity to operate in accordance with the first method referred to herein, and/or the second method referred to herein.

According to another aspect of the disclosure, there is provided a system. The system comprises at least one first entity, as described earlier, and at least one AP device, as described earlier.

According to another aspect of the disclosure, there is provided a computer program comprising instructions which, when executed by processing circuitry, cause the processing circuitry to perform the first method referred to herein and/or the second method referred to herein.

According to another aspect of the disclosure, there is provided a computer program product, embodied on a non-transitory machine-readable medium, comprising instructions which are executable by processing circuitry to cause the processing circuitry to perform the first method referred to herein and/or the second method referred to herein.

Thus, in the manner described above, improved techniques and apparatus for configuring an AP device of a telecommunications network are provided. Advantageously, the techniques described herein provide several benefits in the realm of telecommunications network optimisation, especially when applied to indoor environments. Specifically, the techniques and apparatus described herein address the critical need for adaptable, efficient, and economically viable indoor wireless network optimisation. Moreover, the techniques and apparatus disclosed herein set a new standard for managing and optimising 5G networks (e.g. in enterprise environments) .

BRIEF DESCRIPTION OF THE DRAWINGS

For a better understanding of the techniques, and to show how they may be put into effect, reference will now be made, by way of example, to the accompanying drawings, in which:

Figure 1 is a block diagram illustrating a first entity according to an embodiment;

Figures 2A and 2B are a block diagrams illustrating a method performed according to some embodiments;

Figures 3A, 3B and 4 are a schematic illustrations of a system according to some embodiments;

Figure 5 is a signalling diagram illustrating a method performed according to an embodiment;

Figure 6 is a block diagram illustrating a method performed according to an embodiment;

Figure 7 illustrates an example neural network;

Figures 8-13 illustrate examples of a user interface according to some embodiments; and

Figures 14-18 are schematic illustrations of a docking apparatus according to some embodiments.

DETAILED DESCRIPTION

Generally, all terms used herein are to be interpreted according to their ordinary meaning in the relevant technical field, unless a different meaning is clearly given and/or is implied from the context in which it is used. All references to a/an/the element, apparatus, component, means, step, etc. are to be interpreted openly as referring to at least one instance of the element, apparatus, component, means, step, etc., unless explicitly stated otherwise. The steps of any methods disclosed herein do not have to be performed in the exact order disclosed, unless a step is explicitly described as following or preceding another step and/or where it is implicit that a step must follow or precede another step. Any feature of any of the embodiments disclosed herein may be applied to any other embodiment, wherever appropriate. Likewise, any advantage of any of the embodiments may apply to any other embodiments, and vice versa. Other objectives, features and advantages of the enclosed embodiments will be apparent from the following description.

Some of the embodiments contemplated herein will now be described more fully with reference to the accompanying drawings. Other embodiments, however, are contained within the scope of the subject-matter disclosed herein, the disclosed subject-matter should not be construed as limited to only the embodiments set forth herein; rather, these embodiments are provided by way of example to convey the scope of the subject-matter to those skilled in the art.

In some instances, detailed descriptions of well-known methods, entities, interfaces, circuits, and devices are omitted so as not obscure the description with unnecessary detail. Those skilled in the art will appreciate that the functions described may be implemented in one or more entities using hardware circuitry (e.g., analogue and/or discrete logic gates interconnected to perform a specialised function, ASICs, PLAs, etc. ) and/or using software programs and data in conjunction with one or more digital microprocessors or general purpose computers. Entities that communicate using the air interface also have suitable radio communications circuitry. Moreover, where appropriate the technology can additionally be considered to be embodied entirely within any form of computer-readable memory, such as solid-state memory, magnetic disk, or optical disk containing an appropriate set of computer instructions that would cause a processor to carry out the techniques described herein.

Some of the techniques described herein involve an ML model. For the purposes of the present disclosure, the term “ML model” can encompass, within its scope, an ML algorithm, comprising processes or instructions through which data may be used in a training process to generate a model artefact for performing a given task, or for representing a real world process or system. Alternatively, or in addition, the term “ML model” may encompass the model artefact that is created by such a training process, and which comprises the computational architecture that performs the task. Alternatively, or in addition, the term “ML model” may encompass the process performed by the model artefact in order to complete the task. References to “ML model” , “model” , “model parameters” , “model information” , etc., may thus be understood as relating to any one or more of the above concepts encompassed within the scope of “ML model” .

Some of the techniques described herein involve an environment. An environment, as referred to herein, can be an enclosed environment and/or an interior environment. For example, the environment referred to herein may be comprise a building and/or a structure. In some examples, the environment can correspond to an internal volume of an artificial (e.g. man-made) and/or natural structure (e.g. a cave) . As such, in some examples, the environment referred to herein can be an indoor environment. For example, an environment, as referred to herein, may comprise an enterprise facility, and/or an industrial facility. The environment referred to herein may comprise a warehouse, an exhibition center, a mining environment, an office space, a shopping mall, a hotel, an airport, etc. Although some examples have been provided for the type of environment referred to herein, it will be understood that the environment referred to herein can be any other type of environment.

The telecommunications network referred to herein can be any type of telecommunications network. In some embodiments, the telecommunications network can be a mobile network, such as a fifth generation (5G) mobile network or any other generation mobile network (e.g. 6G) . For example, the telecommunications network can be a 5G core (5GC) network. In some embodiments, the telecommunications network can be a radio access network (RAN) . Although some examples have been provided for the type of telecommunications network referred to herein, it will be understood that the telecommunications network referred to herein can be any other type of telecommunications network. The telecommunications network can be referred to herein as the “network” .

Some of the techniques described herein involve an access point (AP) device of the telecommunications network. An AP device (e.g. the at least one first AP device referred to herein, and/or the at least one second AP device referred to herein) may comprise hardware configured to enable a wireless device to access a telecommunications network, as referred to herein. For example, the AP device may comprise an antenna. In some examples, the AP device may comprise an indoor antenna. For example, the AP device may comprise a radio dot, such as an Ericsson radio dot. The AP device may comprise a docking apparatus (e.g. configured to attach the AP device to a structure, such as a wall) . In some examples, the AP device can comprise an antenna and a docking apparatus for the antenna. For example, the AP device can comprise an indoor antenna and a docking apparatus for the indoor antenna.

Some of the techniques described herein involve a wireless device. As used herein, a wireless device (e.g. the at least one first wireless device referred to herein, and/or the at least one second wireless device referred to herein) may refer to a device capable, configured, arranged and/or operable to communicate wirelessly with network nodes and/or other wireless devices (e.g. user equipment (UE) ) . Examples of a wireless device include, but are not limited to, a smart phone, mobile phone, cell phone, voice over IP (VoIP) phone, wireless local loop phone, desktop computer, personal digital assistant (PDA) , wireless camera, gaming console or device, music storage device, playback appliance, wearable terminal device, wireless endpoint, mobile station, tablet, laptop, laptop-embedded equipment (LEE) , laptop-mounted equipment (LME) , smart device, wireless customer-premise equipment (CPE) , vehicle, vehicle-mounted or vehicle embedded/integrated wireless device, etc. Other examples include any wireless device identified by the 3GPP.

As mentioned above, there exist challenges in meeting the complex demands of indoor environments. The challenges can be categorised into two main areas: characteristics of omnidirectional antenna signals, and engineering and maintenance of indoor wireless coverage.

In more detail, with regards to omnidirectional antenna signals, in (e.g. vertical industry) environments demanding high-quality continuous coverage, the omnidirectional coverage capability of antennas reveals inconsistencies. Even without obstructions or interference, the signal strength and quality received by devices are not necessarily linear with distance from the antenna. Practical tests show that while omnidirectional antennas emit waves that appear spherical, irregularities in signal strength occur within certain horizontal planes of their coverage. A test can be set up in which a grid-marked space is covered by an omnidirectional antenna, and a terminal moves across each grid to collect network speeds and signal strengths. The findings from this test, repeated over three rounds, reveal non-linear and non-uniform signal distributions, conflicting with the anticipated omnidirectional pattern. Additionally, parties using 5G indoor coverage report regular signal instability in moving automated guided vehicles (AGVs) on production lines. These findings suggest that while omnidirectional antennas have stable coverage characteristics, their signal distribution within the intended range is not uniformly linear in real-world settings.

For engineering and maintenance of indoor wireless coverage, the installation characteristics of devices like omnidirectional antennas pose challenges in large spaces where improper placement or changes in spatial usage necessitate wireless signal adjustments. Traditional fixed installations typically require on-site professional work or transmission power adjustments (e.g. reducing power to mitigate inter-cell interference) , which are problematic in dynamic (e.g. industrial) settings. Moreover, parties with specific business needs require dynamic flexibility in their wireless coverage, ideally supported by network capabilities or operator services. However, achieving this is not economically viable under current operator service models and engineering capabilities.

Indeed, existing technologies which merely use software for configuration adjustment, lack the flexibility to dynamically adjust coverage and performance in response to real-time conditions and demands within indoor environments. Current network optimisation methods are no longer sufficient to meet customer demands. Current network optimisation relies heavily on manual labour. Engineers typically carry heavy hardware equipment to conduct on-site detection, scanning, and optimisation. This approach is both time-consuming and labour-intensive, and it cannot achieve timely and on-demand optimisation results.

Moreover, with the advent of Industry 4.0, enterprises are increasingly concerned about data and privacy security. They are reluctant to allow professional network optimisation personnel to frequently enter their production or office environments, which poses a greater challenge to network optimisation.

These observations highlight the need for more adaptable, efficient, and economically viable wireless networking solutions, particularly in the face of evolving 5G technology and its integration into complex indoor environments.

Artificial intelligence (AI) presents a promising tool to automate network adjustments and optimisation. AI can be configured to collect and analyse network data, making intelligent decisions about network adjustments. This not only improves efficiency, but also enhances the security and reliability of (e.g. 5G) networks deployed in industrial settings.

Figure 1 illustrates a first entity 10 in accordance with an embodiment. The first entity 10 is for configuring at least one first access point (AP) device of a telecommunications network. Alternatively, or in addition, the first entity 10 is for training a machine learning (ML) model for configuring at least one first AP device of a telecommunications network. In some embodiments, the first entity 10 referred to herein can refer to equipment capable, configured, arranged and/or operable to communicate directly or indirectly with the at least one first AP device referred to herein, the at least one first wireless device referred to herein, the at least one second AP device referred to herein, the at least one second wireless device referred to herein, and/or with other nodes or equipment to enable and/or to perform the functionality described herein. In some embodiments, the first entity 10 referred to herein can, for example, be a physical node (e.g. a physical machine or server) or a virtual node (e.g. a virtual machine, VM) .

As illustrated in Figure 1, the first entity 10 comprises processing circuitry (or logic) 12. The processing circuitry 12 controls the operation of the first entity 10 and can implement the method described herein in respect of the first entity 10. The processing circuitry 12 can be configured or programmed to control the first entity 10 in the manner described herein. The processing circuitry 12 can comprise one or more hardware components, such as one or more processors, one or more processing units, one or more multi-core processors and/or one or more modules. In particular implementations, each of the one or more hardware components can be configured to perform, or is for performing, individual or multiple steps of the method described herein in respect of the first entity 10. In some embodiments, the processing circuitry 12 can be configured to run software to perform the method described herein in respect of the first entity 10. The software may be containerised according to some embodiments. Thus, in some embodiments, the processing circuitry 12 may be configured to run a container to perform the method described herein in respect of the first entity 10.

Briefly, the processing circuitry 12 of the first entity 10 is configured to obtain, from at least one first wireless device of the network, first information indicative of a network measurement associated with the at least one first wireless device. The processing circuitry 12 of the first entity 10 is configured to obtain, from at least one first AP device, second information indicative of a status of the at least one first AP device. The at least one first AP device is comprised in a first environment of the network, and the orientation of the at least one first AP device is remotely configurable. The processing circuitry 12 of the first entity 10 is configured to determine, using a ML model, an orientation configuration of the at least one first AP device based on the first information, the second information, third information, and fourth information. The third information is indicative of a target state of the network. The fourth information is indicative of one or more characteristics of the first environment.

Alternatively, or in addition, the processing circuitry 12 of the first entity 10 is configured to train a ML model to determine an orientation configuration of at least one first AP device based on first information, as defined herein, second information, as defined herein, third information, as defined herein, and fourth information, as defined herein.

As illustrated in Figure 1, in some embodiments, the first entity 10 may optionally comprise a memory 14. The memory 14 of the first entity 10 can comprise a volatile memory or a non-volatile memory. In some embodiments, the memory 14 of the first entity 10 may comprise a non-transitory media. Examples of the memory 14 of the first entity 10 include, but are not limited to, a random access memory (RAM) , a read only memory (ROM) , a mass storage media such as a hard disk, a removable storage media such as a compact disk (CD) or a digital versatile disk (DVD) , and/or any other memory.

The processing circuitry 12 of the first entity 10 can be communicatively coupled (e.g. connected) to the memory 14 of the first entity 10. In some embodiments, the memory 14 of the first entity 10 may be for storing program code or instructions which, when executed by the processing circuitry 12 of the first entity 10, cause the first entity 10 to operate in the manner described herein in respect of the first entity 10. For example, in some embodiments, the memory 14 of the first entity 10 may be configured to store program code or instructions that can be executed by the processing circuitry 12 of the first entity 10 to cause the first entity 10 to operate in accordance with the method described herein in respect of the first entity 10. Alternatively or in addition, the memory 14 of the first entity 10 can be configured to store any information, data, messages, requests, responses, indications, notifications, signals, or similar, that are described herein. The processing circuitry 12 of the first entity 10 may be configured to control the memory 14 of the first entity 10 to store any of the information, data, messages, requests, responses, indications, notifications, signals, or similar, that are described herein.

In some embodiments, as illustrated in Figure 1, the first entity 10 may optionally comprise a communications interface 16. The communications interface 16 of the first entity 10 can be communicatively coupled (e.g. connected) to the processing circuitry 12 of the first entity 10 and/or the memory 14 of the first entity 10. The communications interface 16 of the first entity 10 may be operable to allow the processing circuitry 12 of the first entity 10 to communicate with the memory 14 of the first entity 10 and/or vice versa. Similarly, the communications interface 16 of the first entity 10 may be operable to allow the processing circuitry 12 of the first entity 10 to communicate with any one or more nodes (e.g. the at least one first AP device, the at least one first wireless device, the at least one second AP device, and/or the at least one second wireless device) referred to herein and/or any other node. The communications interface 16 of the first entity 10 can be configured to transmit and/or receive any of the information, data, messages, requests, responses, indications, notifications, signals, or similar, that are described herein. In some embodiments, the processing circuitry 12 of the first entity 10 may be configured to control the communications interface 16 of the first entity 10 to transmit and/or receive any of the information, data, messages, requests, responses, indications, notifications, signals, or similar, that are described herein.

Although the first entity 10 is illustrated in Figure 1 as comprising a single memory 14, it will be appreciated that the first entity 10 may comprise at least one memory (i.e. a single memory or a plurality of memories) 14 that operate in the manner described herein. Similarly, although the first entity 10 is illustrated in Figure 1 as comprising a single communications interface 16, it will be appreciated that the first entity 10 may comprise at least one communications interface (i.e. a single communications interface or a plurality of communications interfaces) 16 that operate in the manner described herein. It will also be appreciated that Figure 1 only shows the components required to illustrate an embodiment of the first entity 10 and, in practical implementations, the first entity 10 may comprise additional or alternative components to those shown.

Figure 2A illustrates a computer-implemented method performed in accordance with an embodiment. The method is for configuring at least one first AP device of a telecommunications network. The at least one first AP device is comprised in a first environment of the network, and the orientation of the at least one first AP device is remotely configurable. The first entity 10 described earlier with reference to Figure 1 can be configured to operate in accordance with the method of Figure 2A. For example, the method can be performed by or under the control of processing circuitry 12 of the first entity 10.

With reference to Figure 2A, at block 102, first information is obtained from at least one first wireless device of the network. The first entity 10 (e.g. the processing circuitry 12 of the first entity 10) can be configured to obtain (e.g. receive) the first information (e.g. via the communications interface 16 of the first entity 10) . The first information is indicative of a network measurement associated with the at least one first wireless device. Therefore, wireless device (e.g. UE) data can be leveraged to inform network optimisation strategy.

The first information can comprise any information that is indicative of a performance of the network. In some examples, the first information may comprise one or more key performance indicators (KPIs) associated with the at least one first wireless device. In some examples, the measurement associated with the at least one first wireless device can be associated with one or more applications (apps) running on the at least one first wireless device. For example, the at least one wireless device can be equipped with wireless signal measurement, and/or traffic measurement apps. The network measurement comprised in the first information may be performed by the one or more apps running on the at least one first wireless device. In some examples, the network measurement associated with the at least one first wireless device may comprise one or more of a reference signal received power (RSRP) value, a serving cell RSRP (SS-RSRP) value, a reference signal received quality (RSRQ) value, a serving cell RSRQ (SS-RSRQ) value, a signal to interference and noise ratio (SINR) value, an uplink rate, a downlink rate, a latency value, and a jitter value. It will be understood that the network measurement associated with the at least one first wireless device can comprise any relevant performance parameter measurement of the network.

As illustrated at block 104 of Figure 2A, second information is obtained from the at least one first AP device. The first entity 10 (e.g. the processing circuitry 12 of the first entity 10) can be configured to obtain (e.g. receive) the second information (e.g. via the communications interface 16 of the first entity 10) . The second information is indicative of a status of the at least one first AP device. Obtaining the second information may comprise receiving, from the at least one first AP device, a (e.g. current) status report comprising the second information. The second information may comprise one or more KPIs associated with the network.

In some examples, the second information can comprise information indicative of an orientation of the at least one first AP device (e.g. within the first environment) . For example, the second information can comprise location information of the at least one first AP device within the first environment. In some examples, the second information may comprise information indicative of an angle of rotation of the at least one first AP device. The angle of rotation of the at least one first AP device may be relative to a starting point and/or a fixed axis of the at least one first AP device. In some examples, the second information may comprise information indicative of an angle of tilt of the at least one first AP device. The angle of tilt of the at least one first AP device may be relative to a starting point and/or a fixed axis of the at least one first AP device. In some examples, the second information may comprise information indicative of an identifier (ID) of the at least one first AP device. The identifier of the at least one first AP device may be referred to herein as a “SmartDock ID” . In some examples, the second information may comprise information indicative of sensor data associated with the at least one first AP device. The sensor data may be data associated with the first environment (e.g. obtained by the at least one first AP device) . The sensor data may comprise, for example, temperature data, humidity data, and/or light data. In some examples, the second information may comprise information indicative of an action performed by the at least one first AP device. The action performed by the at least one first AP device may comprise a list of instructions that the at least one first AP device is (e.g. currently) executing. In some examples, the action performed by the at least one first AP device may comprise a work status of the at least one first AP device. In some examples, the second information may comprise information indicative of a power status of the at least one first AP device. The power status may comprise a battery status of the at least one first AP device. In some examples, the second information may comprise information indicative of a result of a previously executed configuration (e.g. executed by the at least one first AP device) . The result may indicate whether a (e.g. previously received command) is running, successful, and/or failed.

As illustrated at block 106 of Figure 2A, an orientation configuration of the at least one first AP device is determined, using a ML model, based on the first information, the second information, third information, and fourth information. The first entity 10 (e.g. the processing circuitry 12 of the first entity 10) can be configured to perform the determination. The third information is indicative of a target state of the network, and the fourth information is indicative of one or more characteristics of the first environment. The orientation configuration can comprise an angle of rotation for the at least one first AP device. Alternatively, it in addition, the orientation configuration can comprise an angle of tilt for the at least one first AP device.

In some examples, the target state of the network (as indicated by the third information) may comprise a target network coverage for at least part of the first environment. For example, the target state of the network may comprise a particular level of network coverage for a particular volume and/or area of the first environment. Alternatively, or in addition, the target state of the network may comprise a target network signal strength for the at least one first wireless device. For example, the target state of the network may correspond to a target network performance for the at least one first wireless device.

In some examples, the fourth information may comprise information indicative of a physical layout of the first environment. The physical layout of the first environment may correspond to a geography of the first environment, and/or an architecture of the first environment. For example, the information indicative of the physical layout of the first environment may comprise information indicative of one or more objects and/or partitions (e.g. walls) present in the first environment. In a particular example, the information indicative of the physical layout of the first environment may comprise information indicative of an arrangement of pieces of equipment and/or walls in a production area of a manufacturing facility. In some examples, the fourth information may comprise information indicative of a location of the at least one first AP device in the first environment. The at least one first AP device may be located on a wall, roof, support, etc. of the first environment. In some examples, the fourth information may comprise information indicative of a location of the at least one first wireless device in the first environment. In some examples, the fourth information may comprise information indicative of a location of one or more objects in the first environment. Information indicative of a location in the first environment may comprise co-ordinate information corresponding to the first environment.

Although not illustrated in Figure 2A, in some examples, determining the orientation configuration of the at least one first AP device may comprise generating, using the ML model, a simulation of one or more possible orientation configurations of the at least one first AP device, and selecting the orientation configuration from the one or more possible orientation configurations based on the target state of the network. In some examples, the orientation configuration can be selected from the one or more possible orientation configurations if the orientation configuration is determined to satisfy the target state of the network. For example, selection of the orientation configuration from the one or more possible orientation configurations can be based on satisfying the target state of the network. As such, a plurality of possible orientation configurations of the at least one first AP device can be generated and the most optimal orientation configuration can be selected based on whether the simulation of said orientation configuration exhibits the relevant characteristics required to satisfy the target state of the network.

In some examples, the generated simulation is a digital twin model of the first environment. Thus, the generated simulation may correspond to a complete digital twin of the entire first environment (e.g. an indoor 5G network environment) . The digital twin model can correspond to an accurate virtual representation of the physical and/or operational characteristics of the at least one first AP device in the first environment. The digital twin model can be defined as a model which accurately reflects the physical network, including its layout, components, and/or configurations. The twin model can be used to perform automatic simulation and analysis of optimisation of adjustment plans within the digital twin model. As such, the digital twin model can be a resource for evaluating potential changes before they are implemented in the real network, minimising risks and disruptions.

In some examples, generating the simulation is based at least in part on the first, second, third, and/or fourth information, as defined herein. As such, the generated simulation (e.g. the digital twin model) can encompass (e.g. a simulated representation of) the at least one first AP device and the at least one first wireless device. The generated simulation can simulate the wireless network generated (e.g. at least in part) by the at least one first AP device in the first environment. Therefore, the first entity 10 (e.g. the processing circuitry 12 of the first entity 10) can process real-time coverage data, run simulations for coverage adjustment, and generate feasible solutions for optimising network performance. In some examples, updated information can be obtained automatically such that the digital twin model can be automatically updated (e.g. using network data) . This ensures that the digital twin model remains accurate and up-to-date, reflecting the latest changes in the physical network. In some examples, a visual representation of the generated simulation can be generated. As such, a holistic view of the network’s performance can be provided.

The generation of a digital twin model, as defined herein, can provide the ability to simulate and analyse changes in the network, and/or the first environment, and thus allows for rapid optimisation and adjustment. This enables network operators to quickly respond to changing demands and ensure optimal network performance. Moreover, the digital twin can provide a real-time representation of the first environment, allowing operators to monitor and identify potential issues before they impact performance. This enables proactive network management and prevents disruptions to user experience. Furthermore, the digital twin can be used to simulate and analyse security threats, helping operators to identify and implement appropriate security measures. In addition, the digital twin can be used to plan and simulate future network changes, helping operators to make informed decisions about network expansion and upgrades.

Although not illustrated in Figure 2A, in some examples, transmission of fifth information may be initiated towards the at least one first AP device. The fifth information can be indicative of the determined orientation configuration, as defined herein. In some examples, the fifth information may comprise a request for the at least one first AP device to adjust an orientation of the at least one first AP device based on the fifth information. Therefore, in some examples, the at least one first AP device can be instructed to adjust their configuration based on the fifth information.

In some examples, the ML model referred to herein may comprise a deep Q-network (DQN) . In some examples, the ML model is a trained ML model. The ML model can be trained according to the method as described herein (e.g. with reference to Figure 2B below) .

Thus, in the manner described with reference to the method of Figure 2A, AI can be leveraged to offer a flexible and scalable approach to (e.g. indoor) wireless telecommunications network maintenance. The method can be implemented to transform the manner in which networks are optimised.

Figure 2B illustrates a computer-implemented method performed in accordance with an embodiment. The method is for training a ML model for configuring at least one first AP device of a telecommunications network. The at least one first AP device is comprised in a first environment of the network, and the orientation of the at least one first AP device is remotely configurable. The first entity 10 described earlier with reference to Figure 1 can be configured to operate in accordance with the method of Figure 2B. For example, the method can be performed by or under the control of processing circuitry 12 of the first entity 10.

As illustrated with reference to Figure 2B, at block 202, a ML model is trained to determine an orientation configuration of the at least one first AP device based on first information, second information, third information, and fourth information. The first entity 10 (e.g. the processing circuitry of the first entity 10) can be configured to perform this training. The first, second, third, and fourth information is as defined herein (e.g. with reference to Figure 2A) .

In some examples, the ML model may be trained using a first training dataset. The first training dataset may comprise information obtained from a test environment. Alternatively, or in addition, in some examples, the ML model may be trained using a second training dataset. The second training dataset can comprise information obtained from the first environment, as defined herein. In some examples, the test environment and the first environment are different. The test environment can be a testing environment for collecting information for the first training dataset. For example, the test environment may comprise a laboratory environment. The test environment can be a physical environment. In some examples, the test environment may be an indoor environment.

In some examples, the first training dataset may comprise information indicative of a location of at least one second AP device. The at least one second AP device is comprised in the test environment. The information indicative of the location of the at least one at least one second AP device can comprise (e.g. three-dimensional) co-ordinate information of the at least one second AP device within the test environment (e.g. relative to an origin point of the test environment) . In some examples, the first training dataset may comprise information indicative of an orientation configuration, as defined herein, of the at least one second AP device. For example, the information indicative of the orientation configuration of the at least one second AP device can comprise information indicative of a rotation and/or a tilt of the at least one second AP device. In some examples, the information indicative of the orientation configuration of the at least one second AP device can comprise information indicative of a plurality of network measurements, as defined herein, associated with the at least one second wireless device. The at least one second wireless device is comprised in the test environment. In some examples, the first training dataset may comprise information indicative of sensor data, as defined herein, associated with the test environment. In some examples, the first training dataset may comprise information indicative of a physical layout, as defined herein, of the test environment.

In some examples, the plurality of network measurements associated with the at least one second wireless device may be performed at a plurality of locations in the test environment. In these examples, the first training dataset may comprise information indicative of a location of the at least one second wireless device at which each network measurement was performed. The information indicative of the location of the at least one second wireless device can comprise (e.g. three-dimensional) co-ordinate information of the at least one second wireless device within the test environment (e.g. relative to an origin point of the test environment) . In some examples, obtaining the plurality of network measurements associated with the at least one second wireless device may comprise attaching the at least one second wireless device to a robot and configuring the robot to move through the test environment according to a set path. Information indicative of the set path may be comprised in the first training dataset according to some examples.

In some examples, the plurality of network measurements associated with the at least one second wireless device may comprise one or more key performance indicators (KPIs) associated with the at least one second wireless device. In some examples, the plurality of network measurements associated with the at least one second wireless device can be associated with one or more applications running on the at least one second wireless device.

In some examples, the second training dataset may comprise information indicative of a location of at the least one first AP device, as defined herein, an orientation configuration of the at least one first AP device, as defined herein, a plurality of network measurements associated with the at least one first wireless device, as defined herein, sensor data associated with the first environment, as defined herein, and/or a physical layout of the first environment, as defined herein.

In some examples, the at least one first AP device and the at least one second AP device may comprise the same AP device. For example, the at least one first AP device and the at least one second AP device may be of the same make and model. In a particular example, each first AP device and each second AP device may be the same type of AP device (e.g. an Ericsson dot antenna device) .

Figures 3A and 3B show a first environment illustrating the performance of a method according to an embodiment. As illustrated in Figures 3A and 3b, the first environment comprises at least one first AP device 302, 304. In the example illustrated in Figures 3A and 3B, the at least one first AP device 302, 304 comprises a plurality of AP devices. However, it will be understood that the at least one first AP device can comprise any type and/or number of AP device according to other examples. As also illustrated in Figures 3A and 3B, the first environment comprises at least one first wireless device 306, 308, 310. In the example illustrated in Figure 3B, the at least one first wireless device comprises a first UE 306, a second UE 308, and a third UE 310. However, it will be understood that the at least one first wireless device can comprise any type and/or number of wireless device according to other examples.

As illustrated in Figure 3A, the at least one first AP device 302, 304 of the first environment are in a first orientation configuration. As illustrated by the signal bars of Figure 3A, in the first orientation configuration, the signal strength experienced by the first UE 306 and the third UE 310 is high. However, as also illustrated in Figure 3A, the signal strength experienced by the second UE 308 is relatively low.

Although not explicitly illustrated in Figures 3A and 3B, a second orientation configuration of the at least one first AP device 302, 304 can be determined as described with reference to Figure 2A. In this example, the target state of the network, as defined herein, can be that each of the at least one first wireless device 306, 308, 310 experience the same signal strength. The second orientation configuration can then be determined based at least in part on the target state of the network.

As illustrated in Figure 3B, the at least one first AP device 302, 306 can reconfigure their orientation based on the determined second orientation configuration. As illustrated by the signal bars of Figure 3B, the target state of the network is achieved via reconfiguration based on the determined second orientation configuration.

Figure 4 is a block diagram illustrating a system according to an embodiment. As illustrated in Figure 4, the system can comprise a first entity 10 ( “WiseWave server” ) , as defined herein, a “WiseWave Front-End” , and a “SmartDock” . In some examples, the first entity 10 referred to herein can be configured to perform the functionality of the “WiseWave server” and the “WiseWave Front-End” . In some examples, the at least one first AP device referred to herein may comprise the “SmartDock” . In some examples, the “SmartDock” may be a docking apparatus as defined herein. As illustrated in Figure 4, the system can be integrated into an existing (e.g. 5G) telecommunications network infrastructure.

The first entity 10 can be understood as the intelligent control center of the system. As illustrated in Figure 4, in some examples, the first entity 10 can be situated within a user plane ( “N6/Gi” ) of the (e.g. 5G) telecommunications network. As described herein, in some examples, the first entity 10 can generate a simulation. In some examples, the generated simulation can be a digital twin model of the first environment (e.g. an indoor wireless environment covered by the telecommunications network) .

In some examples, the first entity 10 can obtain first information, as defined herein, from the WiseWave Front-End. In some examples, the first information and/or the second information, as defined herein, may be collected by the WiseWave Front-End. The first information can include (e.g. current) wireless coverage data associated with at least one first wireless device as described herein.

As described herein, in some examples, the first entity 10 can generate a visual representation of the generated simulation referred to herein. As such, the first entity 10 can generate visual outputs for a human-machine interface. In some examples, the first entity 10 can execute simulation calculations within the generated simulation (e.g. digital twin model) based on manual inputs for network optimisation and/or coverage adjustments. The calculations may be performed as part of the determination of an orientation configuration of the at least one first AP device as described herein. In some examples, determination of the orientation configuration can comprise assessing feasibility of possible orientation configurations and/or formulating adjustment strategies for the at least one first AP device (e.g. SmartDock) . As described herein, in some examples, the first entity 10 can initiate transmission of fifth information towards the at least one first AP device. In some examples, initiating transmission of the fifth information may comprise sending one or more commands to the at least one first AP device (e.g. SmartDock) via the WiseWave Front-End. The one or more commands may be to adjust 5G wireless signal coverage in the first environment.

As illustrated in Figure 4, in some examples, the first entity 10 can comprise a plurality of units (elements) . As illustrated in the examples of Figure 4, the first entity 10 can comprise a “Data collection &data management” unit, an “AI Model processing” unit, a “Decision handling” unit, a “User management” unit, and a “System log &monitoring” unit.

The “Data collection &management” unit may be configured to collect data and/or information (e.g. first information, as defined herein) from the WiseWave Front-End. As illustrated in Figure 4, the data and/or information collected from the Wise-Wave Front-End can be collected via interface 1 ( “i/f 1” ) . In some examples, the “Data collection &management” unit may be configured to collect data and/or information (e.g. second information, as defined herein, and/or network management data) from a 5G Network Management System ( “EMS” ) . The i/f 1 may be a direct communication interface between the first entity 10 and the WiseWave Front-End. The i/f 1 interface may, in some examples, undertake all messages and data transmission between the first entity 10 and the WiseWave Front-End. The i/f 1 interface may be a RESTful interface.

As illustrated in Figure 4, the data and/or information collected from the 5G Network Management System can be collected via interface 2 ( “i/f 2” ) . The i/f 2 interface may be between the first entity 10 and the EMS. The i/f 2 interface may be used to obtain network element side state information from EMS to assist the first entity 10 in determining orientation configurations for the at least one first AP device. Messages and/or data handled by the i/f 2 interface can include standard network fault data, configuration data, accounting data, performance data, security (FCAPS) index reporting data, and/or configuration management (CM) operation instructions for (e.g. RAN) network elements.

In some examples, the “Data collection &management” unit may be configured to collect data and/or information (e.g. real-time click-through rate (CTR) data) directly from the network (e.g. radio access network (RAN) ) . As illustrated in Figure 4, the data and/or information collected from the network can be collected via interface 3 ( “i/f 3” ) . The i/f 3 may be an interface configured to provide a direct connection between the first entity 10 and the (e.g. radio access) network. The i/f 3 can be used to collect real-time CTR data of the network for decision support.

As illustrated in Figure 4, an interface 4 ( “i/f 4” ) may be provided between the WiseWave Front-End and the at least one first AP device. Communication between the at least one first AP device and other system components can be realised through the i/f 4. In some examples, messages and/or data handled by the i/f 4 can include status data, sensor data, execution result feedback messages, etc. from the at least one first AP device. In some examples, messages and/or data handled by the i/f 4 can include adjustment commands, operation commands, etc. transmitted to the at least one first AP device. The i/f 4 can be a RESTful interface that is capable of exchanging information, such as commands, from external systems.

The “AI Model processing” unit may be configured to generate the simulation of one or more possible orientation configurations, as described herein. For example, “AI Model processing” unit may be configured to generate a digital twin model tailored to the first environment, enabling data simulation and analysis. In some examples, the “AI Model processing” unit may process simulation tasks and produce decision recommendations. In some examples, the “AI Model processing” unit may manage the lifecycle of the digital twin model. For example, the “AI Model processing” unit may perform iterative updating of the digital twin model through retraining (e.g. with real-world data collected by the WiseWave Front-End) , and/or a complete offline upgrade (e.g. if significant deviations or changes in the first environment occur) .

The ” Decision Handling” unit may be configured to perform an interactive process with the WiseWave Front-End to validate and confirm adjustment commands for the at least one first AP device (e.g. SmartDock) . Operators may have the flexibility to manually intervene in simulation results, choosing to abandon, trigger a recalculation, and/or continue with the dispatch of adjustment commands to the at least one first AP device. This capability ensures historical actions are recorded for reference, allowing for the retracing or repetition of specific adjustments when necessary.

The “User Management” unit may be configured to secure access to the system. As such, the “User Management” unit may ensure that only authorised personnel can safely execute operations within the system.

The “System Log &Monitoring” unit may be configured to track all operations and maintain the health status of the system. In some examples, the system illustrated in Figure 4 can exchange decision outputs with other systems, such as enterprise management systems, facilitating a broader collaboration for network management and optimisation.

The integration of some or all of the components of the system provides a powerful and flexible platform for both enterprise self-management and remote service provision by telecom operators. This adaptability is advantageous in addressing the varied challenges faced in indoor wireless network deployments, particularly in complex and dynamic enterprise environments.

The modular design of the system illustrated in Figure 4 allows for seamless integration with other network optimisation systems. An open architecture ensures that the system can evolve alongside changing technological landscapes and enterprise needs.

Figure 5 is a signalling diagram illustrating an exchange of signals in a system (e.g. telecommunications network) according to an embodiment. As illustrated in Figure 5, the system can comprise a first entity 10, as referred to herein, at least one first wireless device 504 ( “UE” ) , as referred to herein, and at least one first AP device 506 ( “SmartDock” ) , as referred to herein. As also illustrated in Figure 5, in some examples, the system may comprise a front-end 502 ( “Front-end APP” ) , an iPerf Server 522, an EMS node 524, and an other system 526. The front-end 502 may correspond to the WiseWave Front-End as described with reference to Figure 4 above.

As illustrated in Figure 5, the first entity 10 can comprise one or more units 510, 512, 514, 516, 518, 520. The one or more units 510, 512, 514, 516, 518, 520 may comprise physical hardware units and/or software (e.g. virtualised) units. As illustrated in Figure 5, in some examples, the first entity 10 can comprise a Data collection &data management unit 512, as described herein, an AI Model processing unit 514, as described herein, a Decision handling unit 516, as described herein, a User management unit 518, as described herein, and a System log &monitoring unit 520, as described herein. As illustrated by arrow 528 of Figure 5, each of the Data collection &data management unit 512, the AI Model processing unit 514, the Decision handling unit 516, and the User management unit 518 can communicate with (e.g. send information to) the System log &monitoring unit 520. As such, the first entity 10 can be equipped with system logging capabilities. For example, the first entity 10 (e.g. the system log &monitoring unit 520 of the first entity 10) can generate log records for events and/or operations occurring within the system.

In some examples, the first entity 10 can comprise a front-end adapter unit 510. The front-end adapter unit 510 may be comprised in the communications interface 16 of the first entity 10. The front-end adapter unit 510 may be configured to communicate with the front-end 502. In some examples, the front-end adapter unit 510 may be configured to transform data and/or information generated by the first entity 10 into a form that is suitable for (e.g. interpretation by) the front-end 502. Alternatively, or in addition, the front-end adapter unit 510 may transform data and/or information received from the front-end 502 into a form that is suitable for (e.g. interpretation by) the first entity 10. As illustrated in Figure 5, the front-end adapter unit 510 can communicate with each of the Data collection &data management unit 512, the AI Model processing unit 514, and the Decision handling unit 516, and the System log &monitoring unit 520.

The first entity 10 may obtain information and/or data in one or more different scenarios. The information obtained can include the first, second, third, and/or fourth information, referred to herein. For example, data and/or information collection (e.g. for actual on-site wireless coverage quality) may be required in one of the following scenarios:

· During initial deployment of the system (e.g. first entity 10) , real data may be collected on-site to train the ML model referred to herein. The collected data may also be used to generate a simulation (e.g. create a digital twin model) , as described herein.

· For routine model maintenance, data sampling can be conducted to verify the accuracy of the ML model.

· When modifications or updates to an existing ML model are desired.

As illustrated by arrow 530 of Figure 5, in some examples, a (e.g. internet protocol) connection can be established between the at least one first wireless device and the iPerf Server 522.

As illustrated by arrow 532 of Figure 5, the first entity 10 can obtain (e.g. first) information from the at least one first wireless device 504 and/or from the front-end 502 in coordination with (e.g. via) the at least one first wireless device 504. Thus, in some examples, data and/or information collection can be carried out using the front-end 502 in coordination with the at least one first wireless device 504 (e.g. a UE) . In some examples, the at least one first wireless device 504 can comprise any commercially available (e.g. Android) smartphone. The at least one first wireless device 504 may be equipped with wireless signal measurement and/or traffic measurement apps. In some examples, the front-end 502 may connect to the at least one first wireless device 504 via WiFi to acquire data measured by the apps running on the at least one first wireless device 504.

First information, as defined herein, is indicative of a network measurement associated with the at least one first wireless device 504. For example, the first information can include information indicative of a wireless coverage quality (e.g. experienced by the at least one first wireless device 504) . In some examples, the front-end 502 can obtain (e.g. collect) the first information from the at least one first wireless device 504. As illustrated by arrow 532 of Figure 5, the first entity 10 (e.g. the data collection &management unit 512 of the first entity 10) can receive the first information from the at least one first wireless device 504 via the front-end 502. Thus, the front-end 502 can transmit the first information towards the first entity 10. As also illustrated by arrow 532 of Figure 5, the network measurement associated with the at least one first wireless device, as defined herein, can comprise a RSRP value, a SINR value, an uplink rate (e.g. uplink data throughput) , and/or a downlink rate (e.g. downlink data throughput) .

Second information, as defined herein, is indicative of a status of the at least one first AP device 506. As illustrated by arrow 536 of Figure 5, in some examples, the front-end 502 can obtain (e.g. collect) the second information from the at least one first AP device 506. As illustrated by arrow 538 of Figure 5, the first entity 10 (e.g. the data collection &management unit 512 of the first entity 10) can obtain (e.g. receive) the second information from the at least one first wireless device 504 via the front-end 502. Thus, the front-end 502 can transmit the second information towards the first entity 10.

In some examples, the front-end 502 can transmit the first information and/or the second information to the first entity 10 (e.g. the data collection &management unit 512 of the first entity 10) in a text format. In some examples, the data collection &management unit 512 of the first entity 10 can process the first information and/or the second information (e.g. using one or more Extract, Transform, Load (ETL) operations) . The first entity 10 (e.g. the data collection &management unit 512 of the first entity 10) may store the (e.g. processed) first information and/or second information (e.g. in the memory 14 of the first entity 10) .

As illustrated by arrows 534 and 540 of Figure 5, in some examples, the data collection &management unit 512 may transfer the first information and second information, respectively, to the AI model processing unit 514. In some examples, the AI model processing unit 514 can use the first information and/or the second information to update the ML model referred to herein. In some examples, the AI model processing unit 514 may use the first information and/or the second information to generate and/or correct the generated simulation (e.g. digital twin model of the first environment) , as referred to herein.

As illustrated by arrow 542 of Figure 5, the first entity 10 (e.g. the AI model processing unit 514 of the first entity 10) can obtain third information, as defined herein. The third information is indicative of a target state of the network. As illustrated in Figure 5, in some examples, the first entity 10 can obtain the third information from the front-end 502.

The third information may comprise network optimisation and/or wireless coverage adjustment intentions (e.g. input by a user of the system, such as a network admin) . In some examples, the front-end 502 may comprise a graphical user interface (GUI) through which at least some of the third information can be obtained (e.g. from a user of the GUI) . The target state of the network can comprise an alteration to signal coverage strength in a specific wireless coverage area (e.g. by increasing or decreasing RSRP by a specific percentage) .

The first entity 10 (e.g. the AI model processing unit 514 of the first entity 10) determines an orientation configuration of the at least one first AP device based on the first, second, third and fourth information, as defined herein, using the ML model, as defined herein. As described herein, determining the orientation configuration can comprise generating a simulation. In some examples, generating the simulation may comprise (re) generating and/or altering a previously generate simulation (e.g. a previously created digital twin model) . Determining the orientation configuration can comprise using the ML model to conduct simulation calculations based on the third information (e.g. set targets/intentions) .

The fourth information, as defined herein, can be specific to the first environment. For example, the fourth information can comprise information indicative of a measurement plan of the first environment. In some examples, data acquisition by the first entity 10 can be specific (e.g. custom-designed) to on-site conditions of the first environment. The on-site conditions of the first environment can include the physical on-site environment, the information technology (IT) environment, and/or the actual deployment of the (e.g. 5G) telecommunications network.

As illustrated by arrow 544 of Figure 5, in some examples, the orientation configuration can be output by the first entity 10 (e.g. the AI model processing unit 514 of the first entity 10) . Alternatively, or in addition, the generated simulation can be output by the first entity 10 (e.g. the AI model processing unit 514 of the first entity 10) . In some examples, the first entity 10 may generate a visual representation of the generated simulation. As illustrated by arrow 546 of Figure 5, in some examples, the orientation configuration and/or the generated simulation may be processed by the front-end adapter unit 510 (e.g. to transform the determined orientation configuration and/or the generated simulation into a form that can be sent to the front-end 502) . As illustrated by arrow 548 of Figure 5, in some examples, the first entity 10 (e.g. the decision handling unit 516 of the first entity 10) can initiate transmission of the determined orientation configuration and/or the generated simulation towards the front-end 502. In some examples, the visual representation of the simulation can be displayed via the GUI of the front-end 502. As such, the front-end 502 can serve as a user interface for users (e.g. enterprise customers and/or telecoms operators) . The front-end 502 can be configured to allow for visualisation of the generated simulation (e.g. the digital twin model referred to herein) , allow for setting coverage adjustment goals, and/or executing simulations to evaluate feasibility of orientation configuration adjustments.

In some examples, some or all of the steps associated with arrows 542, 544, 546 and 548 of Figure 5 can be performed iteratively. For example, the first entity 10 may perform these steps iteratively until it is determined that the determined orientation configuration satisfies the target state of the network (e.g. as indicated by the generated simulation) .

In some examples, a user (e.g. of the front-end 502) can modify the target state of the network (e.g. intended objectives) and rerun the determination of the orientation configuration (e.g. simulation calculations) . In some examples, as illustrated by arrows 550 and 552 of Figure 5, the first entity 10 (e.g. the decision handling unit 516 of the first entity 10) may transmit (e.g. forward) the determined orientation configuration (e.g. strategy suggestions that emerge from simulation results) to external systems, such as the EMS node 524 and the other system 526. In some examples, the first entity 10 (e.g. the decision handling unit 516 of the first entity 10) can request EMS collaboration for configuration changes in (e.g. RAN) elements of the telecommunications network.

As illustrated by arrows 554 and 556 of Figure 5, in some examples, the first entity 10 (e.g. the decision handling unit of the first entity 10) can initiate transmission of fifth information indicative of the determined orientation configuration towards the at least one first AP device 506. As illustrated by arrows 554 and 556 of Figure 5, the transmission of the fifth information can be performed via the front-end 502. As described herein, the fifth information can comprise a request for the at least one first AP device 506 to adjust an orientation of the at least one first AP device 506 based on the fifth information. For example, the request can be a command for the at least one first AP device 506 to adjust its configuration based on the determined orientation configuration. In some examples, transmission of the information may only be initiated if the first entity 10 determines that the determined orientation configuration satisfies the target state of the network (as comprised in the third information) . In some scenarios, a user of the front-end 502 may approve the fifth information before it is dispatched to the at least one first AP device 506 actual implementation.

The AI-driven system illustrated in Figure 5 allows for real-time adaptation to changing indoor environments. This dynamic optimisation ensures consistent and high-quality network coverage, even in complex and variable indoor settings.

Moreover, by automating network optimisation and eliminating the need for manual intervention, the solution significantly reduces the need for manual labour associated with traditional network management.

Figure 6 is a block diagram illustrating a method performed according to an embodiment. The method illustrated in Figure 6 can be performed by the first entity 10 as referred to herein. The steps of Figure 6 illustrate example ways in which methods as described with reference to Figures 2A and 2B may be implemented and supplemented to achieve the above discussed and additional functionality.

As illustrated at block 602 of Figure 6, the process may begin by collecting initial information ( “Data measurement” ) . The initial information can comprise, for example, the first training dataset, as referred to herein, and/or the second training dataset, as referred to herein. In some examples, the initial information can comprise information indicative of a target network, spatial maps of the coverage area of the network, existing deployment locations of at least one first AP device, deployment heights of the at least one first AP device, power consumption of the at least one first AP device, AP device type, tilt angles of the at least one first AP device, rotate angles of the at least one first AP device, and/or current (e.g. 5G indoor radio) network parameters in operation.

In some examples, based on the initial information, a real testing environment may be established (e.g. in a laboratory) to gather data. Therefore, the process can comprise creating a real-world testing environment setup. The volume of data collected may be determined by feedback (e.g. from a model implementation stage) . The collected data can include heights of at least one second AP device, AP device type, tilt angles of the at least one second AP device, rotate angles of the at least one second AP device, operational parameters of the network supported by the at least one second AP device, signal strength (e.g. RSRP) values, uplink throughput, downlink throughput, and/or latency values. One or more of the operational parameters of the network, the signal strength (e.g. RSRP) values, the uplink throughput, the downlink throughput, and the latency values may be obtained from the at least one second wireless device, as referred to herein.

As illustrated at block 604 of Figure 6, an ML model may be generated based on the collected data (e.g. as described with reference to block 60 of Figure 6) . In some examples, generating the ML model may involve training the ML model, as described herein. Training of the ML model may be based on the initial information, product characteristics of the at least one first AP device, and data collected from the test environment. Training the ML model may comprise running the ML model to use ML techniques (e.g. neural network techniques) to predict and generate values for various regions under different conditions. In some examples, training the ML model may comprise generating an initial simulation of the first environment, as defined herein. For example, generating the initial simulation may comprise generating an initial digital twin model representing the first environment.

As illustrated at block 606 of Figure 6, a calibration step may be performed. In some examples, calibration can comprise calibrating the ML model based on the first environment, as defined herein. In some examples, information can be collected from the first environment to perform the calibration. The information collected can comprise the first, second, third and/or fourth information referred to herein. As such, additional data can be collected from the actual network to refine the ML model. The amount (volume) of information and/or data collected can depend on acceptable error margins.

In some examples, the volume of information and/or data collected may be determined by an error value “E” and a predefined acceptable constant “ (c) ” . The constraint on the error value can be understood using Equation 1 below. In Equation 1, “M” is the model output, “T” is the target value (e.g. of the network) , and “c” is a constant. Equation 1 can be utilised to ensure that the difference between the model output “M” and the target value “T” (i.e., the error “E” ) remains within a certain range defined by the constant “c” . This type of constraint is commonly used in optimisation problems to ensure that the model output closely matches the target value, while controlling the magnitude of the error. Suggested values for Equation 1 may be provided after testing of the model in a test environment (e.g. laboratory) as defined herein.

E= (M-T) ≤ (c)

Equation 1

In some examples, calibration can involve homomorphic transfer learning. Collected information and/or data can be used to adjust the ML model through transfer learning, enhancing its accuracy in predicting network performance under various conditions, and leading to a more precise digital twin model generation.

Transfer learning can be utilised to train the model. For example, real target network measurement data can be used as input for homomorphic transfer learning. In transfer learning, it is often assumed that there is some degree of correlation between source domain data and target domain data, even if the data comes from different domains or have different feature distributions. When data distributions of the source and target domains are similar, it is possible to leverage knowledge from the source domain to improve learning performance on the target domain, thus achieving the goal of transfer learning. When Equation 2 (shown below) is true, source domain data ( “D_s” ) and target domain data ( “D_t” ) have the same, or very similar, distributions.

D_s=D_t

Equation 2

As illustrated at block 608 of Figure 6, the ML model can be implemented. In some examples, the ML model can obtain input indicative of areas of the first environment, as defined herein, which require optimisation. In determining an orientation configuration for at least one first AP device of the first environment, a simulation of the first environment may be generated. For example, using a refined digital twin model of the first environment, simulations may be run to determine optimal angles of the at least one first AP device and/or other parameters for the specified areas. The ML model can output feasibility of optimisation for specific orientation configurations and, if applicable, the adjustments needed, such as angles and/or power settings for the at least one first AP device.

As illustrated at block 610 of Figure 6, in some examples, the ML model may be updated. In some examples, the model can be manually prompted to recalibrate (e.g. restarting the process of refining the digital twin model) . Alternatively, the ML model may automatically recalibrate at regular intervals or when changes in measured data (e.g. Call Trace Report data) meets a predefined criterion. As such, it can be ensured that the digital twin model remains accurate and up to date.

Figure 7 is a schematic illustration of an example neural network model. The ML model referred to here can comprise a neural network model according to some examples.

A neural network model can be constructed based on input information associated with a test environment, as referred to herein, and/or a first environment, as referred to herein. For example, the input to the neural network model can comprise characteristics of an AP device (e.g. spectrum, model) , deployment heights of an AP device, angles of an AP device, power consumption of an AP device, AP device groups, and/or target area maps (e.g. of the first environment referred to herein) . The output of the neural network model can comprise predicted values such as signal strength (e.g. RSRP, SS-RSRP, SINR and/or SS-SINR) , uplink speed, downlink speed, jitter, and/or latency.

The ML model can be trained using collected data (e.g. the first training data set, referred to herein, and/or the second training dataset, referred to herein) , and applying a backpropagation algorithm to adjust parameters of the ML model based on the relationship between input and output (e.g. data) . As such, in some examples, the training of the ML model can involve forward propagation of input data through the network, error calculation by comparing predicted and actual values, backward propagation of the error to adjust weights and biases, and iteratively refining the model to minimise error:

The relationship between input “x” , output “y” , and forward propagation function “f_w” is illustrated by Equation 3 below:

y = f_w (x)

Equation 3

Error calculation can be performed by comparing predicted value ( “f_w (x) ” ) and actual values ( “y” ) using Equation 4 below:

Backward propagation of the error can be performed to adjust weights and biases, and the model iteratively refined to minimise error, as illustrated by Equation 5 below:

Once trained, the model can be tested and evaluated using new data to assess its predictive accuracy and reliability. Matrix 1 below can be used (e.g. updated) to obtain a more accurate expression and understanding of the behaviour of the model. Matrix 1 can be considered to correspond to a matrical representation of the example neural network model illustrated in Figure 7. In Matrix 1, input variables are denoted by “X” elements. In this example, there are a range of input variables given by the range “a” to “z” . For example, the output “y_1” is based on the inputs “X_1a” to “X_1z” . Assuming there are 1-n variable sets from “a” to “z” , 1-n output results (i.e. “y_1” to “y_n” ) can be obtained, as illustrated in Matrix 1.

Traditional network optimisation methods mainly focus on macro cell services, utilising network configuration parameters and network statistical reports such as call success rate, drop rate, and roaming success rate to ensure maximum network availability. As such, traditional methods neglect specific terminal performance indicators such as signal strength, SINR, uplink and downlink rates, latency, and jitter.

The ML model referred to herein can be configured to utilise SwarmTune AI. SwarmTune AI stands out by not only being able to incorporate 5G network parameters (e.g. power consumption, spectrum, etc. ) but also terminal configuration parameters (e.g. device group, performance parameters, such as RSRP, SINR, DL, UL, Jitter, Latency, etc. ) , and actual physical deployment parameters (e.g. angle, power consumption, height, relative distance between AP devices) as inputs for network optimisation calculations and evaluations. This comprehensive parameter set allows SwarmTune AI to achieve unparalleled accuracy in network status evaluation and prediction, empowering it to meet increasingly stringent customer requirements.

Some of the advantages of SwarmTune AI are as follows:

· Comprehensive Parameters: Provides more accurate and comprehensive network status evaluation and prediction.

· High Precision: Delivers more precise solutions to meet customer requirements.

· Strong Scalability: Supports various types of 5G enterprise networks.

· Integration with Smart Dock Hardware: SwarmTune AI can seamlessly integrate with Smart Dock hardware to adjust physical deployment parameters (horizontal and vertical angles, power, etc. ) , maximising the utilisation of existing network resources and achieving optimal network coverage and performance enhancements.

SwarmTune AI represents a groundbreaking 5G network optimisation technique specifically tailored to the unique requirements of indoor networks. By enabling more accurate network status evaluation and prediction, and maximising the utilisation of network resources, SwarmTune AI empowers users with unparalleled network experiences.

Figures 8 to 13 are examples of a visual representation of a generated simulation, as referred to herein. In the examples illustrated in Figures 8 to 13, the generated simulation comprises a digital twin model of the first environment, as defined herein.

This visual representation of the generated simulation can be part of a systemised indoor wireless coverage maintenance suite. The suite can be locally deployed (e.g. on-premises managed by a customer) , or cloud-based (e.g. managed by an operator) . The suite is capable of sensing changes in wireless network coverage quality, performing coverage adjustment simulations, and optimising and/or adjusting coverage through controllable and enhanced omnidirectional antennas.

As described with reference to Figure 5 above, in some examples, generating the visual representation of the generated simulation can comprise displaying the visual representation via a GUI of the front-end 502. As such, the front-end 502 can serve as a human-machine interactive interface for the system, as described with reference to Figure 5.

As illustrated in Figure 8, a visualised, three-dimensional representation of the digital twin model of the (e.g. 5G) network coverage area in the first environment can be generated. This visual representation of the digital twin model allows for a comprehensive view of wireless coverage parameters (e.g. RSRP, SINR and/or network bandwidth estimations) at specific locations of the first environment. A sampling granularity can be set in order to divide the area and/or volume of the first environment into distinct units, as illustrated in Figure 8.

As illustrated in Figure 9, a user may select (e.g. via the interface of the front-end 502) a target area ( “Target area digitalization” ) of the first environment within the digital twin model. In some examples, the user can set objectives for a wireless coverage adjustment for the target area. The objectives can be comprised within the third information indicative of the target state of the network, as described herein. As such, the interface of the front-end 502 can be interactive and facilitate precise planning for coverage optimisation. As illustrated in Figure 10, in some examples, the user may request that the wireless coverage for the target area is improved by 10%. It will be understood that this is merely an example, and that the target state of the network could apply to any part of the first environment, and/or any performance metric of the network.

As illustrated in Figure 11, simulation calculation (s) can be performed to assess the feasibility of the intended coverage adjustments. The GUI can be configured to present (e.g. display) the results of the calculation (s) . If the feasibility is confirmed, the system may suggest one or more orientation configurations for the at least one first AP device, as defined herein. The one or more orientation configurations may comprise, for example, one or more adjustment strategies. These strategies can be dispatched to at least one first AP device for actual implementation. As illustrated in Figure 12, the at least one first AP device can be commanded to adjust a configuration of the at least one first AP device based on information indicative of a determined orientation configuration. The digital twin model can be (re) generated to track the progress of the adjustment of the at least one first AP device.

As illustrated in Figure 13, following the execution of the adjustment of the at least one first AP device, the digital twin model can confirm the status of the at least one first AP device, verifying the completion of command (s) . The effectiveness of the adjustment can be verified through a separate sampling and measurement process, ensuring the accuracy of the network coverage improvements.

Figure 14 illustrates a schematic view of a docking apparatus according to an embodiment. In particular, Figure 14 illustrates a cross-sectional view of the docking apparatus. The docking apparatus is for an indoor antenna. Figure 14 illustrates the docking apparatus in a first configuration and a second configuration. The docking apparatus may be referred to herein as a SmartDock. The docking apparatus may be comprised in an AP device, as referred to herein. For example, the docking apparatus may be comprised in the at least one first AP device, referred to herein, and/or the at least one second AP device, referred to herein.

As illustrated in Figure 14, the docking apparatus comprises a first portion, 1402, a second portion 1404, and a third portion 1406. Although not illustrated in Figure 14, the first portion 1402 comprises first coupling means configured to couple the apparatus to a surface. The surface may be, for example, a wall and/or a ceiling. As such, the docking apparatus may be configured to be ceiling mounted according to some examples. In some examples, the first coupling means may comprise one or more screw holes. Although not illustrated in Figure 14, the third portion 1406 comprises second coupling means configured to couple the apparatus to an indoor antenna. As such, the second coupling means can be configured to mount an indoor antenna to the docking apparatus. In some examples, the indoor antenna may comprise a dot antenna, such as an Ericsson radio dot antenna. The second coupling means may comprise one or more screw holes. The one or more screw holes of the second coupling means may be arranged to match mounting holes and/or positions of the indoor antenna. In this way, the docking apparatus can integrate seamlessly into existing installations, for example, replacing existing dot antenna mounts.

As also illustrated in Figure 14, the second portion 1404 is (e.g. directly) coupled to the first portion 1402 and the third portion 1406. As illustrated in the example of Figure 14, the second portion 1404 can be positioned between the first portion 1402 and the third portion 1406. Although not explicitly illustrated in Figure 14, the docking apparatus comprises a controller. The controller is remotely configurable to rotate the second portion 1404 and the third portion 1406 relative to the first portion 1402 about a first axis 1418 of rotation. The controller can be configured to rotate the second portion 1404 and the third portion 1406 together, relative to the first portion 1402. The range of rotational movement can be in the range 0° to 360°. As such, the controller can be configured to rotate the second portion 1404 and the third portion 1406 through one full rotation, relative to the first portion 1402. The angle of rotation of the docking apparatus may be referred to herein as the azimuth of the docking apparatus. The controller is also remotely configurable to tilt the third portion 1406 relative to the second portion 1404.

As such, the docking apparatus (SmartDock) is a specialised hardware component designed to enhance traditional fixed (e.g. 5G) indoor antennas, such as the Ericsson Radio Dot System (RDS) , by adding the capability for mechanical adjustments in orientation (i.e. tilt and rotation) . The docking apparatus can be integrated into the system described herein (e.g. with respect to Figures 4 and 5) as at least a part of the AP device referred to herein. In some examples, an AP device, as referred to herein, may comprise the docking apparatus and an indoor antenna. In some examples, the indoor antenna may be an omnidirectional indoor antenna.

As illustrated in Figure 14, in some examples, the second portion 1404 can comprise a second surface 1410 and a third surface 1412. In some examples, the second surface and the third surface can be planar surfaces. As also illustrated in Figure 14, in some examples, the third surface 1412 may be inclined relative to the second surface 1410. A first angle 1416 may exist between the second surface 1410 and the third surface 1412. The first angle 1416 may be a smallest (e.g. possible) angle between the second surface 1410 and the third surface 1412. In some examples, the first angle 1416 may be in the range 130° to 150°. In some examples, the first angle 1416 may be 140°.

In the example illustrated in Figure 14, the first configuration of the docking apparatus corresponds to a first tilt configuration 1400a, and the second configuration of the docking apparatus corresponds to a second tilt configuration 1400b. In some examples, the controller is remotely configurable to tilt the third portion 1406 from the first tilt configuration 1400a to the second tilt configuration 1400b. It will be understood that the controller can be configured to tilt the third portion 1406 through a range of angles from the first tilt configuration 1400a to the second tilt configuration 1400b, and vice-versa. In some examples, the third portion 1406 can be tilted through a range of 40° from the first tilt configuration 1400a to the second tilt configuration 1400b, and vice-versa.

As illustrated in Figure 14, in some examples, tilting the third portion 1406 from the first tilt configuration 1400a to the second tilt configuration 1400b may comprise moving a first section of the third portion 1406 towards the third surface, and moving a second section of the third portion 1406 away from the second surface. As also illustrated in Figure 14, in some examples, in the first tilt configuration 1400a, a first surface 1408 of the third portion 1406 can be substantially parallel to the second surface 1410, and, in the second tilt configuration 1400b, the first surface 1408 can be substantially parallel to the third surface 1412. As further illustrated in Figure 14, in some examples, the third portion 1406 can be in contact with the second portion 1404 in the first tilt configuration 1400a and the second tilt configuration 1400b.

Figure 15 is a schematic illustration of the docking apparatus according to an embodiment. In the examples illustrated in Figure 15, the apparatus may be said to be in the first tilt configuration.

As illustrated in Figure 15, in some examples, the first coupling means 1508 of the first portion 1402 can comprise one or more screw holes.

As also illustrated in Figure 15, in some examples, the first portion 1402 can comprise a first opening 1504, the second portion 1404 can comprise a second opening and the third portion 1406 can comprise a third opening 1502. In some examples, the first opening 1504, the second opening, and the third opening 1502 can be circular. In some examples, the first opening 1504, the second opening, and the third opening 1502 may lie on the same axis. For example, the first opening 1504, the second opening, and the third opening 1502 may be positioned centrally within the first portion 1402, the second portion 1404, and the third portion 1406, respectively. As illustrated in Figure 15, in some examples, the first opening 1504 and the second opening may be positioned on the first axis 1418, as described with reference to Figure 14 above. The first opening 1504, the second opening, and the third opening 1502 may be configured to allow a power cable (e.g. a power over ethernet (PoE) cable) to pass through the docking apparatus. In this way, a convenient route for supplying power to the indoor antenna is provided. In some examples, the docking apparatus and the indoor antenna may share the same power source, negating the need for separate power lines. In some examples, the docking apparatus may be powered via a 12 Volt direct current (DC) supply.

As illustrated in Figure 15, in some examples, the second portion 1404 can be coupled to the third portion 1406 by one or more hinges 1506. In the example illustrated in Figure 15, the one or more hinges comprise two hinges. However, it will be understood that the one or more hinges may comprise any number of hinges.

As illustrated in Figure 15, the third portion 1406 may comprise a cylindrical disc. In some examples, the first surface 1408, as defined herein, may correspond to a planar surface that is flush with the top of the third portion 1406. The top of the third portion 1406 may face the second portion 1404.

Figure 16 is a schematic illustration of an AP device according to an embodiment. In the example illustrated in Figure 16, the docking apparatus as described with reference to Figures 14 and 15 is coupled to an indoor antenna 1602. As illustrated in Figure 16, in some examples, the indoor antenna may comprise a radio dot antenna, as described herein.

The docking apparatus described herein can transform traditional fixed (e.g. 5G) indoor antennas into mechanically adjustable units. That is, the docking apparatus enables physical alteration of the indoor antenna’s tilt and azimuth angles, allowing for optimal signal coverage (e.g. tailored to the specific needs of the environment in which the indoor antenna is located) .

In particular, indoor coverage in (e.g. 5G) networks is typically ensured by fixed antenna units, such as Ericsson's Radio Dot System (RDS) . These indoor settings can span across office environments, public spaces, and production areas, with special considerations for large indoor spaces like exhibition halls. The applicability of antenna systems varies across two main dimensions: the horizontal open space and the vertical spatial extent. Office environments usually have smaller horizontal and vertical dimensions, whereas spaces like exhibition halls, or facilities for manufacturing large equipment, such as airplanes or ships, may feature vertical dimensions reaching 15 to 20 meters.

In such expansive areas, improper placement of indoor antennas (e.g. radio dots) , or changes within the environment, may necessitate wireless signal adjustments. Traditional fixed installations often require on-site engineering work for reconfiguration or power adjustments to the antenna system, such as reducing transmission power to mitigate inter-cell interference. However, in vertical industries, the entry of external engineers into enterprise production sites poses challenges, potentially leading to safety risks and disruptions in production, highlighting a significant pain point.

By utilising the docking apparatus described herein (e.g. to adjust the radiation angle of an indoor antenna 1602) , the above-mentioned challenges are overcome. For instance, in scenarios where new shelving is added in a factory, or when temporary adjustments to product lines are made, the docking apparatus can effectively resolve signal obstructions by re-angling the indoor antenna 1602.

Figure 17 is a schematic illustration of the docking apparatus according to an embodiment. In example illustrated in Figure 17, the docking apparatus may be said to be in the second tilt configuration, as defined herein.

Figure 17 illustrates a number of possible dimensions for the docking apparatus, according to a particular example. It will be understood that the dimensions illustrated in Figure 17 are merely exemplary, and that the docking apparatus may have other dimensions according to other examples.

As illustrated in Figure 17, in some examples, the first portion 1402, the second portion 1404, and the third portion 1406 may be cylindrical in shape. As also illustrated in Figure 17, in some examples, the radius of each of the first portion 1402, the second portion 1404, and the third portion 1406 may be equal.

Figure 18 is a schematic illustration of the docking apparatus in an exploded view according to an embodiment.

As illustrated in Figure 18, in some examples, the docking apparatus can comprise locking means 1802. The locking means may be configured to secure (lock) the docking apparatus to the surface to which it is coupled. As also illustrated in Figure 18, the docking apparatus can comprise a ring 1806. The ring 1806 may be configured to permit rotation of the second portion 1404 and the third portion 1406 relative to the first portion 1402.

As also illustrated in Figure 18, in some examples, the docking apparatus can comprise one or more motors 1808, 1812. The one or more motors 1808, 1812 can comprise a stepper motor according to some examples. In some examples, the controller referred to herein may be configured to actuate the one or more motors 1808, 1812 to perform rotation and/or tilting operations as described herein. As such, in some examples, the docking apparatus may include one or more stepper motors and one or more control circuits for mechanical adjustments of the docking apparatus. As illustrated in Figure 18, in some examples, the docking apparatus can comprise a screen 1810. The screen 1810 may be a light emitting diode (LED) screen. The screen 1810 may be configured to display information indicative of one or more parameters and/or characteristics of the docking apparatus (e.g. network connectivity status, error messages, etc. ) .

As illustrated in Figure 18, in some examples, the docking apparatus may comprise one or more gears 1814. The one or more gears 1814 can be configured to permit mechanical adjustment of the docking apparatus. As also illustrated in Figure 18, in some examples, the docking apparatus can comprise one or more hinges 1818. The one or more hinge 1818 can be configured to couple the third portion 1406 to the second portion 1404 (e.g. via a receiving bracket of the third portion 1406) . As illustrated in Figure 18, in some examples, the docking apparatus can comprise one or more legs 1816. The one or more legs 1816 can permit tilting operations of the docking apparatus (e.g. from the first tilt configuration to the second tilt configuration, and vice-versa) .

In some examples, the controller referred to herein may be configured to communicate with a network (e.g. the telecommunications network, as referred to herein) . Although not illustrated in Figure 18, in some examples, the docking apparatus may comprise a subscriber identify module (SIM) card port. In these examples, the controller may be configured to communicate with the network using a SIM card. In some examples, the controller may be configured to perform a rotation operation and/or a tilting operation based on orientation configuration information received via the network (e.g. as described with reference to Figures 4 and 5) . As such, the docking apparatus can adjust the physical tilt and/or azimuth angle of an indoor antenna based on (e.g. online) control commands. When the docking apparatus is comprised in an AP device in the system described herein (e.g. as described with reference to Figures 4 and 5) , the utilisation of signal characteristics learned through AI techniques can be maximised.

In some examples, the docking apparatus may comprise a communications interface. The communications interface of the docking apparatus may be configured to communicate with the first entity 10 referred to herein, and/or the front-end 502 referred to herein. In some examples, the communications interface of the docking apparatus may comprise a WiFi module and/or a 5G module.

In some examples, the docking apparatus may comprise one or more sensor modules. The one or more sensor modules may comprise a temperature sensor, a humidity sensor, and/or a light sensor. As such, the docking apparatus can be provided with enhanced environment sensing capabilities.

As described herein, the controller is remotely configurable. In some examples, the docking apparatus may have firmware (pre) installed which enables automated control of the controller. Updates and/or upgrades to the firmware may be performed online (e.g. via communication with the first entity 10 referred to herein) , or offline (e.g. via a USB interface of the docking apparatus) .

The docking apparatus (e.g. the communications interface of the docking apparatus) can receive one or more commands. The one or more commands can comprise a command to perform a tilt adjustment (e.g. +/-0-40°) . The one or more commands can comprise a command to perform a rotation adjustment (e.g. +/-0-360°) . The one or more commands can comprise a command to pause (stop) an adjustment. The one or more commands can comprise a command to terminate (e.g. future and/or planned) adjustments. The one or more commands can comprise a command to resume performing (e.g. previously terminated and/or paused) adjustments. The controller can be configured to execute the one or more commands.

There is also provided a computer program comprising instructions which, when executed by processing circuitry (such as the processing circuitry 12 of the first entity 10 described herein) , cause the processing circuitry to perform at least part of the method described herein. There is provided a computer program product, embodied on a non-transitory machine-readable medium, comprising instructions which are executable by processing circuitry (such as the processing circuitry 12 of the first entity 10 described herein) to cause the processing circuitry to perform at least part of the method described herein. There is provided a computer program product comprising a carrier containing instructions for causing processing circuitry (such as the processing circuitry 12 of the first entity 10 described herein) to perform at least part of the method described herein. In some embodiments, the carrier can be any one of an electronic signal, an optical signal, an electromagnetic signal, an electrical signal, a radio signal, a microwave signal, or a computer-readable storage medium.

In some embodiments, the first entity 10 functionality described herein can be performed by hardware. Thus, in some embodiments, the first entity 10 described herein can be a hardware entity. However, it will also be understood that optionally at least part or all of the first entity described herein can be virtualised. For example, the functions performed by the first entity 10 described herein can be implemented in software running on generic hardware that is configured to orchestrate the first entity functionality described herein. Thus, in some embodiments, the first entity 10 described herein can be a virtual entity. Virtualised instances of the first entity 10 can be rapidly deployed, scaled, and/or modified according to changing needs of a network environment. In some embodiments, at least part or all of the first entity functionality described herein may be performed in a network enabled cloud. Thus, the method described herein can be realised as a cloud implementation according to some embodiments. The first entity functionality described herein may be distributed, e.g. the first entity functionality may be performed by one or more different entities. For example, the first entity functionality (e.g. such as data processing, AI-driven analysis, and simulation calculations) can be distributed across multiple cloud servers. Such a distribution of functionality ensures robust performance, scalability, and redundancy.

Therefore, as described herein, there are provided improved techniques and apparatus for configuring at least one first AP of a telecommunications network. As 5G technology becomes increasingly integral across industries, technology users (e.g. enterprises) demand not only the enhanced connectivity it provides but also new levels of efficiency and security in indoor network maintenance and optimisation. The techniques and apparatus described herein answer this call with a versatile platform tailored to the diverse requirements of network tuning and optimisation for various contexts.

The techniques and apparatus described herein can form a system that facilitates dynamic network coverage optimisation. The system is designed to cater to self-management needs of users as well as offer capabilities for remote service delivery by telecom operators. The system offers intuitive interfaces and robust tools, enabling (e.g. enterprise IT) teams to monitor and modify network configurations proactively, enhancing control over infrastructure and maintaining data privacy.

Moreover, the system described herein accommodates traditional service models, allowing network service providers to offer bespoke optimisation services. Operators can utilise the system’s sophisticated technology suite for continuous network performance monitoring and adaptive optimisation, catering to complex network environments. The system’s versatility in management options not only grants adaptability, but also equips both customers and service providers with the tools to address the evolving challenges of 5G indoor network deployment and maintenance.

The techniques and apparatus described herein are suitable for a number of use cases, such as use cases involving indoor environments. Indoor environments, critical to the effectiveness of 5G network coverage, vary widely and present unique challenges. Indoor environments can range from office spaces, with limited horizontal and vertical expanses, to expansive areas such as public venues and production facilities. In particular, vertical industrial spaces, such as those used for manufacturing large machinery, vehicles, or aircraft, can have ceiling heights extending from 15 to 20 meters.

The horizontal openness and vertical dimensions significantly impact the planning and execution of wireless network installations. For instance, while offices might have minimal spatial complexity, large venues and production environments require a more nuanced approach due to their size and the potential for significant vertical reach. Traditional installations, like fixed antenna units (e.g. dots) , may be inefficient in these large spaces, particularly when spatial configurations or application scenarios change. Such environments traditionally necessitate on-site engineering work to reconfigure or adjust antenna power to address coverage needs, such as reducing power to mitigate inter-cell interference.

However, accessing these vertical industrial spaces poses logistical challenges and safety concerns, complicating the deployment and maintenance of network infrastructure. It is these pain points, particularly in vertical industries, that the techniques and apparatus described herein address by offering a system that dynamically adapts to both the horizontal and vertical dimensions of diverse indoor environments. The system described herein provides the ability to make precise, AI-driven adjustments to the network infrastructure without the need for on-site engineering. As such, the techniques and apparatus described herein revolutionise how indoor network optimisation is approached, offering a safer, more efficient, and more cost-effective route.

In addition, the implementation of AI-driven techniques in the manner described herein facilitates:

· Temporary coverage adjustments for new production areas or other needs.

· Compensation for misconfigured or malfunctioning indoor antennas (e.g. dots) , enhancing network availability.

· Adjustments to the coverage area to balance network load and prevent signal congestion.

· Power consumption reduction during low network utilisation, leading to cost savings and enhanced efficiency.

· Simulation of visual coverage effects through AI, which offers improved network maintainability and operability.

· Fault prediction and proactive maintenance through AI's real-time network monitoring, generating alerts or reports when coverage and performance deviate from preset standards.

These examples represent just a fraction of the potential applications of the techniques and apparatus described herein.

It should be noted that the above-mentioned embodiments illustrate rather than limit the idea, and that those skilled in the art will be able to design many alternative embodiments without departing from the scope of the appended claims. The word “comprising” does not exclude the presence of elements or steps other than those listed in a claim, “a” or “an” does not exclude a plurality, and a single processor or other unit may fulfil the functions of several units recited in the claims. Any reference signs in the claims shall not be construed so as to limit their scope.

Claims

A computer-implemented method for configuring at least one first access point, AP, device (302, 304, 506) of a telecommunications network, wherein the at least one first AP device (302, 304, 506) is comprised in a first environment of the network, wherein the orientation of the at least one first AP device (302, 304, 506) is remotely configurable, the method comprising:

obtaining (102, 532) , from at least one first wireless device (504) of the network, first information indicative of a network measurement associated with the at least one first wireless device (504) ;

obtaining (104, 538) , from the at least one first AP device (302, 304, 506) , second information indicative of a status of the at least one first AP device (302, 304, 506) ; and

determining (106) , using a machine learning, ML, model, an orientation configuration of the at least one first AP device (302, 304, 506) based on the first information, the second information, third information, and fourth information, wherein the third information is indicative of a target state of the network, and wherein the fourth information is indicative of one or more characteristics of the first environment.
The method as claimed in claim 1, the method comprising:

initiating (554) transmission of fifth information indicative of the orientation configuration towards the at least one first AP device (302, 304, 506) .
The method as claimed in claim 2, wherein:

the fifth information comprises a request for the at least one first AP device (302, 304, 506) to adjust an orientation of the at least one first AP device (302, 304, 506) based on the fifth information.
The method as claimed in any of the preceding claims, wherein the orientation configuration comprises:

an angle of rotation for the at least one first AP device (302, 304, 506) ; and/or

an angle of tilt for the at least one first AP device (302, 304, 506) .
The method as claimed in any of the preceding claims, wherein:

the first information comprises one or more key performance indicators, KPIs, associated with the at least one first wireless device (504) .
The method as claimed in any of the preceding claims, wherein:

the measurement associated with the at least one first wireless device (504) is associated with one or more applications running on the at least one first wireless device (504) .
The method as claimed in any of the preceding claims, wherein the network measurement associated with the at least one first wireless device (504) comprises one or more of:

a reference signal received power, RSRP, value;

a serving cell RSRP, SS-RSRP, value;

a reference signal received quality, RSRQ, value;

a serving cell RSRQ, SS-RSRQ, value;

a signal to interference and noise ratio, SINR, value;

an uplink rate;

a downlink rate;

a latency value; and

a jitter value.
The method as claimed in any of the preceding claims, wherein:

the second information comprises one or more KPIs associated with the network.
The method as claimed in any of the preceding claims, wherein:

determining (106) the orientation configuration of the at least one first AP device (302, 304, 506) comprises:

generating, using the ML model, a simulation of one or more possible orientation configurations of the at least one first AP device (302, 304, 506) ; and

selecting the orientation configuration from the one or more possible orientation configurations based on the target state of the network.
The method as claimed in claim 9, wherein the orientation configuration is selected from the one or more possible orientation configurations if the orientation configuration is determined to satisfy the target state of the network.
The method as claimed in claim 9 or 10, wherein:

the generated simulation is a digital twin model of the first environment.
The method as claimed in any of claims 9 to 11, wherein generating the simulation is based at least in part on the fourth information.
The method as claimed in any of claims 9 to 12, comprising:

generating a visual representation of the generated simulation.
The method as claimed in any of the preceding claims, wherein the fourth information comprises information indicative of:

a physical layout of the first environment;

a location of the at least one first AP device (302, 304, 506) in the first environment;

a location of the at least one first wireless device (504) in the first environment; and/or

a location of one or more objects in the first environment.
The method as claimed in any of the preceding claims, wherein the second information comprises information indicative of one or more of:

an orientation of the at least one first AP device (302, 304, 506) ;

an angle of rotation of the at least one first AP device (302, 304, 506) ;

an angle of tilt of the at least one first AP device (302, 304, 506) ;

an identifier of the at least one first AP device (302, 304, 506) ;

sensor data associated with the at least one first AP device (302, 304, 506) ;

an action performed by the at least one first AP device (302, 304, 506) ;

a power status of the at least one first AP device (302, 304, 506) ; and

a result of a previously executed configuration of the at least one first AP device (302, 304, 506) .
The method as claimed in any of the preceding claims, wherein the target state of the network comprises one or more of:

a target network coverage for at least part of the first environment; and

a target network signal strength for the at least one first wireless device (504) .
The method as claimed in any of the preceding claims, wherein:

the first environment is an indoor environment.
The method as claimed in any of the preceding claims, wherein the first environment comprises:

an industrial facility; and/or

an enterprise facility.
The method as claimed in any of the preceding claims, wherein the at least one first AP device (302, 304, 506) comprises an indoor antenna.
The method as claimed in any of the preceding claims, wherein the ML model comprises a deep Q-network, DQN.
The method as claimed in any of the preceding claims, wherein:

the ML model is a trained ML model.
The method as claimed in claim 21, wherein:

the ML model is trained using the method as claimed in any of claims 23 to 39.
A computer-implemented method for training a machine learning, ML, model for configuring at least one first access point, AP, device of a telecommunications network, wherein the at least one first AP device (302, 304, 506) is comprised in a first environment of the network, and wherein the orientation of the at least one first AP device (302, 304, 506) is remotely configurable, the method comprising:

training (202) a machine learning, ML, model to determine an orientation configuration of the at least one first AP device (302, 304, 506) based on first information, second information, third information, and fourth information;

wherein the first information is indicative of a network measurement associated with the at least one first wireless device (504) ;

wherein the second information is indicative of a status of the at least one first AP device (302, 304, 506) ;

wherein the third information is indicative of a target state of the network; and

wherein the fourth information is indicative of one or more characteristics of the first environment.
The method as claimed in claim 23, wherein the ML model is trained using:

a first training dataset, wherein the first training dataset comprises information obtained from a test environment; and/or

a second training dataset, wherein the second training dataset comprises information obtained from the first environment.
The method as claimed in claim 24, wherein:

the test environment and the first environment are different.
The method as claimed in claim 24 or 25, wherein the first training dataset comprises information indicative of:

a location of at least one second AP device, wherein the at least one second AP device is comprised in the test environment;

an orientation configuration of the at least one second AP device;

a plurality of network measurements associated with at least one second wireless device, wherein the at least one second wireless device is comprised in the test environment;

sensor data associated with the test environment; and/or

a physical layout of the test environment.
The method as claimed in claim 26, wherein:

the plurality of network measurements associated with the at least one second wireless device are performed at a plurality of locations in the test environment.
The method as claimed in claim 26 or 27, wherein the plurality of network measurements associated with the at least one second wireless device comprise:

one or more key performance indicators, KPIs, associated with the at least one second wireless device.
The method as claimed in any of claims 26 to 28, wherein:

the plurality of network measurements associated with the at least one second wireless device are associated with one or more applications running on the at least one second wireless device.
The method as claimed in any of claims 26 to 29, wherein the plurality of network measurements associated with the at least one second wireless device comprise one or more of:

a reference signal received power, RSRP, value;

a serving cell RSRP, SS-RSRP, value;

a reference signal received quality, RSRQ, value;

a serving cell RSRQ, SS-RSRQ, value;

a signal to interference and noise ratio, SINR, value;

an uplink rate;

a downlink rate;

a latency value; and

a jitter value.
The method as claimed in any of claims 26 to 30, wherein the orientation configuration of the at least one second AP device comprises:

an angle of rotation for the at least one second AP device; and/or

an angle of tilt for the at least one second AP device.
The method as claimed in any of claims 24 to 31, wherein the second training dataset comprises information indicative of:

a location of at the least one first AP device (302, 304, 506) ;

an orientation configuration of the at least one first AP device (302, 304, 506) ;

a plurality of network measurements associated with the at least one first wireless device (504) ;

sensor data associated with the first environment; and/or

a physical layout of the first environment.
The method as claimed in claim 32, wherein the plurality of network measurements associated with the at least one first wireless device (504) comprises:

one or more key performance indicators, KPIs, associated with the at least one first wireless device (504) .
The method as claimed in claim 32 or 33, wherein:

the plurality of network measurements associated with the at least one first wireless device (504) is associated with one or more applications running on the at least one first wireless device (504) .
The method as claimed in any of claims 32 to 34, wherein the plurality of network measurements associated with the at least one first wireless device (504) comprises one or more of:

a reference signal received power, RSRP, value;

a serving cell RSRP, SS-RSRP, value;

a reference signal received quality, RSRQ, value;

a serving cell RSRQ, SS-RSRQ, value;

a signal to interference and noise ratio, SINR, value;

an uplink rate;

a downlink rate;

a latency value; and

a jitter value.
The method as claimed in any of claims 32 to 35, wherein the orientation configuration of the at least one first AP device (302, 304, 506) comprises:

an angle of rotation for the at least one first AP device (302, 304, 506) ; and/or

an angle of tilt for the at least one first AP device (302, 304, 506) .
The method as claimed in any of claims 32 to 36, wherein:

the at least one first AP device (302, 304, 506) and the at least one second AP device comprise the same AP device.
The method as claimed in claim 37, wherein the at least one first AP device (302, 304, 506) and the at least one second AP device comprise at least one indoor antenna.
The method as claimed in any of claims 23 to 38, wherein the ML model comprises a deep Q-network, DQN.
A docking apparatus for an indoor antenna (1602) , the apparatus comprising:

a first portion (1402) , wherein the first portion (1402) comprises first coupling means (1508) configured to couple the apparatus to a surface;

a second portion (1404) ;

a third portion (1406) , wherein the third portion (1406) comprises second coupling means configured to couple the apparatus to the indoor antenna (1602) , and wherein the second portion (1404) is coupled to the first portion (1402) and the third portion (1406) ; and

a controller, wherein the controller is remotely configurable to:

rotate the second portion (1404) and the third portion (1406) relative to the first portion (1402) about a first axis (1418) of rotation; and

tilt the third portion (1406) relative to the second portion (1404) .
The docking apparatus as claimed in claim 40, wherein:

the second portion (1404) comprises a second surface (1410) and a third surface (1412) .
The docking apparatus as claimed in claim 41, wherein:

the second surface (1410) and the third surface (1412) are planar surfaces.
The docking apparatus as claimed in any of claim 41 or 42, wherein:

the third surface (1412) is inclined relative to the second surface (1410) .
The docking apparatus as claimed in any of claims 41 to 43, wherein:

a first angle (1416) between the second surface (1410) and the third surface (1412) is 130° to 150°.
The docking apparatus as claimed in claim 44, wherein:

the first angle (1416) is 140°.
The docking apparatus as claimed in any of claims 41 to 45, wherein the controller is remotely configurable to:

tilt the third portion (1406) from a first tilt configuration (1400a) to a second tilt configuration (1400b) .
The docking apparatus as claimed in claim 46, wherein tilting the third portion (1406) from the first tilt configuration (1400a) to the second tilt configuration (1400b) comprises:

moving a first section of the third portion (1406) towards the third surface (1412) ; and

moving a second section of the third portion (1406) away from the second surface (1410) .
The method as claimed in claim 46 or 47, wherein:

in the first tilt configuration (1400a) , a first surface of the third portion (1406) is substantially parallel to the second surface (1410) ; and

in the second tilt configuration (1400b) , the first surface is substantially parallel to the third surface (1412) .
The docking apparatus as claimed in any of claims 46 to 48, wherein:

the third portion (1406) is in contact with the second portion (1404) in the first tilt configuration (1400a) and the second tilt configuration (1400b) .
The docking apparatus as claimed in any of claims 40 to 49, wherein:

the first portion (1402) comprises a first opening (1504) ;

the second portion (1404) comprises a second opening; and

the third portion (1406) comprises a third opening (1502) .
The docking apparatus as claimed in claim 50, wherein:

the first opening (1504) and the second opening are positioned on the first axis (1418) .
The docking apparatus as claimed in in any of claims 40 to 51, wherein:

the controller is configured to communicate with a network.
The docking apparatus as claimed in claim 52, further comprising:

a subscriber identity module, SIM, card port, wherein the controller is configured to communicate with the network using a SIM card.
The docking apparatus as claimed in claim 52 or 53, wherein:

the controller is configured to perform a rotation operation and/or a tilting operation based on orientation configuration information received via the network.
The docking apparatus as claimed in any of claims 52 to 54, wherein:

the network is a telecommunications network.
The docking apparatus as claimed in any of claims 40 to 55, wherein:

the second portion (1404) is coupled to the third portion (1406) by one or more hinges (1506) .
The docking apparatus as claimed in any of claims 40 to 56, wherein:

the second portion (1404) is coupled between the first portion (1402) and the third portion (1406) .
An access point, AP, device comprising:

an indoor antenna (1602) ; and

the docking apparatus as claimed in any of claims 40 to 57.
A method performed by a system, the method comprising:

the method as claimed in any of claims 1 to 22; and/or

the method as claimed in any of claims 23 to 39.
A first entity (10) comprising:

processing circuitry (12) configured to operate in accordance with any of claims 1 to 22; and/or any of claims 23 to 39.
A first entity (10) as claimed in claim 60, wherein:

the first entity (10) comprises:

at least one memory (14) for storing instructions which, when executed by the processing circuitry (12) , cause the first entity (10) to operate in accordance with any of claims 1 to 22; and/or any of claims 23 to 39.
A computer program comprising instructions which, when executed by processing circuitry, cause the processing circuitry to perform the method according to any of claims 1 to 22; and/or any of claims 23 to 39.
A computer program product, embodied on a non-transitory machine-readable medium, comprising instructions which are executable by processing circuitry to cause the processing circuitry to perform the method according to any of claims 1 to 22;and/or any of claims 23 to 39.