CN109803285A - A kind of cell processing method, device and the network equipment - Google Patents

Abstract

The present invention provides a cell processing method, device and network equipment, wherein the cell processing method includes: predicting the traffic volume when the user is located in a target cell according to the current location of the user and the state of the user's current use of services; The traffic volume when the target cell is located, and the target cell is switched to an energy-saving state. The solution of the present invention can perform cell energy saving processing based on user analysis, so that when performing energy saving processing on the target cell, the influence of the user entering, staying and leaving the target cell on the traffic volume of the cell under the current actual situation is considered, so as to understand more accurately The future network load of the target cell improves the accuracy of cell energy saving, reduces the probability of deviation, and improves network performance and user experience.

Description

Cell processing method, device and network device

technical field

The present invention relates to the field of communication technologies, and in particular, to a cell processing method, device and network device.

Background technique

Reducing energy consumption and reducing carbon dioxide emissions is one of the sustainable development goals of modern society. A large number of base stations are usually deployed in a wireless communication network, and these base stations usually work 24 hours a day. Therefore, the base stations of the mobile communication network consume a lot of energy every day. In order to reduce energy consumption and realize a green communication network, an energy-saving mechanism has been introduced in LTE (Long Term Evolution) and LTE-A (LTE-Advanced, Advanced Long Term Evolution) systems. The cells of some base stations use other base station cells to provide services to achieve the purpose of energy saving.

Specifically, a cell energy saving scheme that is often used at present is, based on the prediction of the cell's own traffic volume, if the cell's own traffic volume is predicted to be small, the cell is switched to enter the energy-saving state; and if the cell's own traffic volume is predicted to be large, it is guaranteed The cell is in normal working state.

However, since the traffic volume of the cell itself is usually predicted based on the historical data of the cell, and the influence of the current actual situation on the traffic volume of the cell itself is not considered, the existing energy saving schemes of the cell often deviate, thus affecting the network performance and user experience.

SUMMARY OF THE INVENTION

Embodiments of the present invention provide a cell processing method, apparatus, and network equipment, so as to solve the problem that deviations often occur in existing cell energy saving solutions, which affect network performance and user experience.

In a first aspect, an embodiment of the present invention provides a cell processing method, including:

According to the current location of the user and the state of the user's current use of services, predict the traffic volume when the user is located in the target cell;

The target cell is switched to an energy-saving state according to the traffic when the user is located in the target cell.

In a second aspect, an embodiment of the present invention further provides a cell processing apparatus, including:

The prediction module is used to predict the traffic volume when the user is located in the target cell according to the current location of the user and the state of the user's current use of services;

The processing module is configured to switch the target cell to an energy-saving state according to the traffic when the user is located in the target cell.

In a third aspect, an embodiment of the present invention further provides a network device, including a memory, a processor, and a computer program stored on the memory and running on the processor, wherein the computer program is The processor implements the steps of the above cell processing method when executed.

In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, wherein when the computer program is executed by a processor, the steps of the above cell processing method are implemented.

In the embodiment of the present invention, by predicting the traffic volume of the user when the user is located in the target cell according to the current location of the user and the current state of the user's use of services, and switching the target cell to the energy-saving state according to the traffic volume when the user is located in the target cell, it is possible to Perform cell energy saving processing based on user analysis, so that when performing energy saving processing on the target cell, the impact of users entering, staying and leaving the target cell on the traffic volume of the cell under the current actual situation is considered, so as to more accurately understand the future network load of the target cell. , improve the accuracy of cell energy saving, reduce the probability of deviation, and improve network performance and user experience.

Description of drawings

In order to illustrate the technical solutions of the embodiments of the present invention more clearly, the following briefly introduces the accompanying drawings that need to be used in the embodiments of the present invention. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without any creative effort.

1 shows a schematic diagram of a grid map according to an embodiment of the present invention;

2 shows a schematic diagram of the use of a trajectory prediction model according to an embodiment of the present invention;

FIG. 3 shows a schematic diagram of the use of a service prediction model according to an embodiment of the present invention;

FIG. 4 shows a flowchart of a cell processing method according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of a resource utilization mapping model according to an embodiment of the present invention;

FIG. 6 shows a flowchart of a cell processing process according to an embodiment of the present invention;

FIG. 7 shows one of the schematic structural diagrams of the cell processing apparatus according to the embodiment of the present invention;

FIG. 8 shows the second schematic structural diagram of a cell processing apparatus according to an embodiment of the present invention;

FIG. 9 is a schematic structural diagram of a network device according to an embodiment of the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

In order to facilitate the understanding of the embodiments of the present invention, the following contents involved in the embodiments of the present invention are first explained.

Target cell: a processing object in the embodiment of the present invention, a cell that predicts whether to wake up or shut down through user analysis.

User presence status information: indicates the relationship between the user and the target cell, that is, the user is located in the target cell, or the user is not located in the target cell. For example, if the number of users involved in this embodiment of the present invention is m, the user presence status information can be described by using matrix E: For example, e ₁ represents an existence state relationship between m users and the target cell. For another example, if m is equal to 4, e ₁ can be expressed as: It states that users 1, 2, and 3 are located in the target cell, and user 4 is not located (or left) in the target cell; alternatively, e ₁ can be expressed as: It states that users 1 and 3 are located in the target cell, users 2 and 4 are not located (or left) in the target cell; and so on.

Trajectory prediction model: It is pre-established according to the user's historical trajectory data (position data), and there is a preset time interval between adjacent trajectory data. In this embodiment of the present invention, the cell granularity is taken as the minimum grid unit, and all the location points included in the cell are regarded as a node (Node). The grid size and number of trajectories can be varied by setting the minimum granularity of the map (N cell area) used to collect trajectory data. When the map grid is larger, the area of each grid node is larger, and the difference between historical trajectories is smaller. For example, as shown in FIG. 1 , when the grid map includes cell nodes n ₁ to n ₇ , the trajectory T ₁ may be represented as {n ₁ , n ₂ , n ₃ , n ₆ } using the cell nodes.

In order to ensure the accuracy of the trajectory prediction model, when collecting trajectory data, the trajectories with less than or equal to 2 nodes in each trajectory can be deleted. Since the user can only move to several surrounding cells within one step, the missing values can be filled for the cells that cannot be reached in one step. The filling method is, for example, query the existing track. A reachable path between nodes in each cell, and the proportion of similar trajectories including this path is high, it will be filled with this path, including reverse trajectories.

Specifically, the trajectory prediction model can be based on a Bayesian inference framework. The basic idea is to use a Bayesian algorithm to predict the user's destination based on the user's current travel trajectory T _p , where the probability that the user's destination is n _j is:

where G represents the total number of destinations for the user. The destination can be identified by screening the area where the user dwell time exceeds a certain threshold (for example, 1h). The longer the dwell time, the more likely it is the destination, but a short stay may be missed.

P(d∈n _j ) represents the prior probability that the destination is n _j , and its value is obtained by counting the proportion of the number of historical trajectories of the user’s destination at n _j in a period of time, namely represents the total number of historical trajectories whose destination location is within n _j , and S _total represents the total number of trajectories of all historical data of the user. Only the prior probability of the destination that the user has reached is non-zero, that is to say, only the grid node that the user has reached can be predicted as the current destination of the user. It should be noted that the prior probability is related to time. When the amount of data is large, the user's trajectories on weekdays and weekends can be counted separately.

P(T _p |d∈n _j ) is the posterior probability, indicating the probability of passing the trajectory T _p when the destination is n _j , in other words, indicating that when the destination is the grid node n _j , the entire travel trajectory The probability that it can partially match the current trajectory T _p (that is, the travel trajectory passes through the current trajectory T _p ). Specifically, the posterior probability P(T _p |d∈n _j ) can be expressed as: in refers to the total number of trajectories whose destination is n _j contains the number of trajectories of trajectories T _p .

It should be pointed out that the destination predicted according to the trajectory prediction model is the destination with the highest probability among all the destinations of the user. After using the trajectory prediction model to predict the user's destination, the trajectory with the highest switching probability under the predicted destination can be selected, and the next location point of the trajectory is the predicted point (ie, the target cell).

Referring to Figure 2, the process of using the trajectory prediction model to obtain the probability of the user arriving at the predicted point can be as follows:

Step 21: Obtain the user's trajectory data, that is, a detailed list of location data;

Step 22: According to the user's trajectory data, use the trajectory prediction model to predict the user's destination;

Step 23: From the different trajectories from the current point of the user to the destination (the next position point of the current point in this trajectory is the predicted point), select the trajectory with the largest cell transition probability;

Step 24: Determine the probability that the user arrives at the predicted point based on the trajectory with the largest cell transition probability.

For example, assuming that the user is at the current point n _c , the destination is predicted to be n _j , and the user has different trajectories from the current point to the destination (the next position point of the current point n _c in this trajectory is the predicted point). Among the different trajectories from _c to n _j , the trajectory with the largest cell transition probability is: P(n _c →n _j )=P(n _c →n _c+1 )·P(n _c+1 →n _c+2 )·· ··P(n _j=1 →n _j ), then it can be determined that the probability that the user reaches the predicted point from the current point n _c is P(n _c →n _c+1 ).

For the transition probability between two adjacent cell nodes, the Markov transition probability model can be used to obtain it. For example, the transition probability of cell node n _a to n _b may be equal to the number of trajectories including cell node {n _a ,n _b } in the historical trajectory database divided by the number of all trajectories including cell node n _a . Therefore, for each pair of adjacent grids in the grid map, their transition probabilities can be pre-computed and stored in a two-dimensional transition matrix, one of which corresponds to the current state of the grid, The other dimension corresponds to the next state. If the symbol M _ij is used to represent the transition matrix, the corresponding transition matrix in Figure 1 is:

Among them, in _Mij , p ₁₂ represents the transition probability of cell nodes n ₁ to n ₂ , p ₅₄ represents the transition probability of cell nodes n ₅ to n ₄ , and so on.

Business prediction model: It is pre-established based on the user's historical business data.

It should be noted that the user's business type generally has the following two characteristics: sudden, the user's behavior of which APP to use at a certain time and place is relatively random, and it has a relatively random behavior in a longer granularity time (greater than 30min). Strong regularity, for example, users watch more video services at home every night, and watch more news in the subway, but they are not predictable in a short granularity time (1min to 5min), but with the user's current Depending on the state and mood, the type of business is more sudden and random; continuous, the length of time a user uses a certain APP can be long or short, from a few seconds to a few hours. The type of business used is related to the previous moment. For example, a user watches TV dramas in the subway every day when commuting, and sees it from the beginning to the end during the ride. Another example is a user who checks WeChat every night, and the duration of each time. It is 1 to 2 hours. For such users, it is easy to predict the service type at the next moment (for example, 15 minutes) from the service type at the previous moment. Therefore, according to the suddenness and persistence of the user's service type, the prediction of the user's service type can be divided into service analysis in a moving state and service analysis in a static state.

Specifically, the service prediction model in the embodiment of the present invention can be divided into a service prediction model in a user motion state and a service prediction model in a stationary state of the user. When conducting business forecasting according to the business forecasting model, it is necessary to first judge the user's state according to the user's historical behavior habits and real-time trajectory data to determine whether it is a moving state or a static state. By default, the user is in motion state, unless the user reaches a certain destination, it is determined that the user enters the stationary state when entering the destination.

In the state of motion, the type of service used by the user is related to the transition of its trajectory. The service prediction model in the user's motion state is related to the user's current service use state and the service state switching probability after the user switches from the current location to the next prediction point. For example, when a user gets on the subway (cell switching), most of them will turn on the video to watch a movie, and turn off the video to get off the train after arriving at the station. The probability of such an event following a single user is high. If the state switching probability of using a certain service type a when the user's location changes is defined as but can be expressed as:

in, Represents the number of events that the user uses service state switching when the location changes, and N(n _i →n _i+1 ) represents the number of events that the user switches from the current location to the next prediction point. According to the service state switching probability, combined with the service type currently used by the user, the probability of the user using a certain service type after a preset time interval can be predicted.

According to the continuity of the service, the service prediction model in the static state of the user is a time series-related model, which is related to the user's current location, the preset time interval, and the probability of using the service at the current location after the preset time interval in the user's history. Predict the probability that a user will use a certain type of service after a preset time interval. If a user has consistent records of using the app in a fixed period of time every day, it is considered that the user has good continuity. For users with strong continuity, a higher weight is given to the historical business type, otherwise the previous time period is considered to be weighted. larger. In this way, in a static state, the probability of a user using a certain service type a after a preset time interval can be expressed as:

Among them, N _i,t,a represents the number of events that the user uses the service type a at the position n _i+1 and the preset time interval t, and A represents the total number of service types corresponding to the user.

Referring to Fig. 3, the process of using the service prediction model to predict the service type of the user after the preset time interval can be as follows:

Step 31: Acquire the current location of the user and the status of the current service used, that is, the detailed list of service data;

Step 32: According to the user's historical behavior habits and real-time trajectory data, determine whether the user is in a motion state or a static state;

Step 33: when the user is in a stationary state, based on the service prediction model in the stationary state of the user, predict the service type after the preset time interval of the user;

Step 34: When the user is in a motion state, predict the service type after the user preset time interval based on the service prediction model in the user's motion state and in combination with the user's current service type.

The embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

Referring to FIG. 4 , an embodiment of the present invention provides a cell processing method, which is applied to a network device, and the network device is used to process a cell. The cell processing method includes the following steps:

Step 401: Predict the traffic volume when the user is located in the target cell according to the current location of the user and the state of the current service usage of the user.

It should be noted that the user in this step specifically involves at least one user.

Among them, when predicting the traffic volume when the user is located in the target cell, in order to fully select the users that may be related to the target cell, the target area can be determined first, and the user (one or more) currently in the target area is determined as the analysis user. . The target area generally includes at least the target cell and the adjacent cells of the target cell, so that the users in the target area can be strongly correlated with the target cell and have the possibility of entering, staying and leaving the target cell. Correspondingly, step 401 may include:

determine at least one user currently within the target area;

According to the received trajectory data of each user in the at least one user, predict the probability that each user is located in the target cell after a preset time interval;

According to the location of the target cell, the current location of each user, and the current status of each user's use of services, the traffic volume when each user is located in the target cell after a preset time interval is predicted.

Step 402: Switch the target cell to the energy saving state according to the traffic volume when the user is located in the target cell.

In this embodiment of the present invention, when the target cell is switched to the energy-saving state according to the traffic when the user is located in the target cell, it may be considered based on a preset traffic volume threshold. The preset traffic volume threshold is preset according to the actual traffic situation of the target cell. For example, when the total traffic volume of the target cell is greater than the preset traffic volume threshold, it is necessary to ensure that the target cell is in an awake state, and when the total traffic volume of the target cell is less than When equal to or equal to the preset traffic volume threshold, the target cell can be switched to the energy-saving state.

Specifically, step 402 may include:

According to the user existence state information, the probability that each user is located in the target cell, and the traffic volume when each user is located in the target cell, determine the probability that the total traffic volume of the target cell is greater than the preset traffic volume threshold;

When the probability that the total traffic of the target cell is greater than the preset traffic threshold is smaller than the first preset probability threshold, the target cell is switched to the energy saving state.

Wherein, each user refers to a user located in the target area. The first preset probability threshold is preset according to the actual traffic situation of the target cell, for example, it can be 2%, 3%, etc., only when the probability that the total traffic volume of the target cell is greater than the preset traffic volume threshold is relatively small, can it be Reduce the impact on users after the target cell saves energy. And if the probability that the total traffic volume of the target cell is greater than the preset traffic volume threshold is greater than or equal to the first preset probability threshold, the target cell should be in a wake-up, ie, normal, working state.

In the cell processing method according to the embodiment of the present invention, by determining the traffic volume when the user is located in the target cell, and when the traffic volume when the user is located in the target cell satisfies the preset condition, the target cell is switched to the energy-saving state, so that energy saving of the cell can be performed based on user analysis. Processing, to realize that when performing energy saving processing on the target cell, consider the impact of users entering, staying and leaving the target cell on the traffic volume of the cell under the current actual situation, so as to more accurately understand the future network load of the target cell and improve the accuracy of cell energy saving. rate, reduce the chance of deviation, and improve network performance and user experience.

In this embodiment of the present invention, the probability that the user is located in the target cell may be predicted based on a preset trajectory prediction model. Therefore, in the embodiment of the present invention, according to the received trajectory data of each user, the process of predicting the probability that each user is located in the target cell after the preset time interval can be as follows:

According to the received trajectory data of each user and the preset trajectory prediction model, predict the destination of each user;

Calculate the possible multiple cell transfer probabilities for each user to reach the destination from the current location, and select the largest cell transfer probability from the multiple cell transfer probabilities;

According to the maximum cell transition probability, determine the probability that each user is located in the target cell.

Optionally, the traffic volume when the user is located in the target cell may be predicted based on a preset traffic prediction model. Therefore, in this embodiment of the present invention, according to the location of the target cell, the current location of each of the users, and the current service status of each of the users, it is predicted that each of the users will be located in the target after the preset time interval. The process of the traffic volume in the cell can be as follows:

According to the location of the target cell, the current location of each user, the current status of each user using services, and the preset service prediction model, it is predicted that after the preset time interval, each user will be located in The probability of using various service types in the target cell;

According to the probability of using various service types when each user is located in the target cell, and the traffic generated when each user uses various service types obtained from historical data, calculate when each user is located in the target cell. volume of business.

The preset service prediction model may be a service prediction model when the user is in motion or a service prediction model in a stationary state of the user, depending on the actual state of the user. The service prediction model in the user's motion state is related to the user's current service use state and the service state switching probability after the user switches from the current location to the next prediction point. The service prediction model in the stationary state of the user is related to the current location of the user, the preset time interval, and the probability of using the service at the current location after the preset time interval in the history of the user. In this way, the prediction accuracy can be improved by considering the user's use of services in a moving state and in a stationary state.

In the embodiment of the present invention, when multiple users are related to the target cell, it is determined that the total traffic volume of the target cell is greater than The process of presetting the probability of the traffic threshold may be:

According to the user existence state information, the probability that each user is located in the target cell and the traffic volume when each user is located in the target cell, calculate the various possible total traffic volumes of the target cell and the corresponding probability of each possible total traffic volume;

According to various possible total traffic volumes of the target cell and the corresponding probability of each possible total traffic volume, the probability that the total traffic volume of the target cell is greater than the preset traffic volume threshold is calculated.

In the following, the process of calculating the probability that the total traffic volume of the target cell is greater than the preset traffic volume threshold will be described by way of a specific example.

For example, if the preset traffic threshold is v _Th , the number of users currently in the target area is m, the number of services that each user can use is s, that is, s types of services, and v _ij indicates that user i is in the target cell. Using the traffic of service type app(j), the traffic matrix V of m users can be expressed as:

u _ij represents the probability that user i uses the service type app(j) in the target cell, then the probability matrix U of m users can be expressed as:

w _i represents the traffic when user i is located in the target cell, then the traffic matrix W of m users is:

The state matrix E of m users in the target cell is:

p _i represents the probability that user i is located in the target cell, then the probability matrix P of m users is:

Therefore, the total traffic matrix S of the target cell is:

That is to say, in this example, the total possible traffic volume of the target cell is 2 ^m in total. Among them, the probability matrix corresponding to the 2 ^m total traffic volumes is p(e _i )=E*p.

Note that when calculating p(e _i ), it is not a simple matrix calculation. For example, if the number of users m is equal to 4, Then the probability p(e ₁ ) corresponding to S ₁ is: p(e ₁ )=p ₁ ·p ₂ ·p ₃ (1−p ₄ ).

That is, the corresponding relationship between the total traffic matrix of the target cell and its probability is:

Based on this, the probability p (S>V _TH ) that the total traffic volume of the target cell is greater than the preset traffic volume threshold is:

In this embodiment of the present invention, when performing energy saving processing on the target cell, in addition to considering the probability that the total traffic volume of the target cell is greater than the preset traffic volume threshold, the resource utilization rate of the target cell can also be further considered. The target cell is switched to the energy-saving state only when the probability that the total traffic volume is greater than the preset traffic volume threshold and the resource utilization rate of the target cell meet the preset conditions.

When determining the resource utilization rate of the target cell, the determination may be performed based on a preset resource utilization rate mapping model. The preset resource utilization mapping model is obtained by using historical data training. Referring to FIG. 5 , the input parameters include at least the total traffic volume of the cell, the number of users in the cell, the RSRP average value of the historical reference signal received power in the cell, and the time information (current time, week, holiday information), etc., and the output parameter is the resource utilization rate of the cell. Among them, when the model is trained, the training data can be updated according to a fixed period, and the corresponding model will also be updated. Because the user needs a certain amount of traffic, the signal strength of the user's location will affect the signal-to-noise ratio of the received signal, which in turn affects the resource occupation of the wireless network. For example, in the LTE system, the user needs a certain amount of traffic. When the user's signal-to-noise ratio is poor, more physical resource blocks (PRBs) will be allocated to ensure that the user's traffic demand is met. Therefore, when the total traffic size of the cell is mapped to 2/3/4G When the system resource utilization is concerned, the current signal strength of the user needs to be considered. Ideally, the historical RSRP of each user in each cell should be modeled and trained. User historical (eg one month) RSRP average.

Specifically, on the basis of calculating a variety of possible total traffic volumes of the target cell and the corresponding probability of each total traffic volume, the process of switching the target cell to the energy-saving state may be as follows:

Determine the total traffic with the largest probability among the various possible total traffics of the target cell as the total traffic of the target cell;

According to the historical data of the target cell, predict the number of users in the target cell and the RSRP average value of the target cell after the preset time interval;

Determine the resource utilization rate of the target cell according to the total traffic volume of the target cell, the number of users in the target cell, the RSRP average value of the target cell, the time information after the preset time interval, and the preset resource utilization mapping model;

When the probability that the total traffic of the target cell is greater than the preset traffic threshold is less than the first preset probability threshold, and the resource utilization rate of the target cell is less than the second preset probability threshold, the target cell is switched to the energy saving state.

Wherein, the second preset resource utilization threshold is preset according to the actual service situation of the target cell.

Hereinafter, the cell processing process of the embodiment of the present invention will be described from the perspectives of offline learning and online prediction with reference to FIG. 6 .

Offline learning: First, obtain historical location data, historical service data, and historical RSRP and KPI of the cell from the detailed data list. The data in the detailed data list can be updated in real time; then, perform feature training on the historical location data to obtain trajectory prediction. model, and perform feature training on historical service data to obtain a service prediction model, and conduct cell traffic-to-resource utilization mapping model training on the cell's historical RSRP and KPI to obtain a resource utilization mapping model.

Online prediction: First, obtain real-time location data and real-time business data from the detailed data list; secondly, perform user trajectory prediction based on real-time location data and trajectory prediction model, and perform user trajectory prediction based on real-time business data and business prediction model. Traffic volume prediction; again, summarize each user's prediction, and predict the traffic volume of the target cell, that is, predict the probability that the total traffic volume of the target cell is greater than the preset traffic volume threshold; then, according to the total traffic volume of the target cell, and the resource utilization rate A mapping model to predict the resource utilization rate of the target cell; finally, when the probability that the total traffic of the target cell is greater than the preset traffic threshold is less than the first preset probability threshold, and the resource utilization rate of the target cell is less than the second preset probability threshold (the preset condition is met), the target cell is switched to the energy-saving state.

The above embodiments describe the cell processing method of the present invention, and the cell processing apparatus of the present invention will be described below with reference to the embodiments and the accompanying drawings.

Referring to FIG. 7 , an embodiment of the present invention further provides a cell processing apparatus, including:

The prediction module 71 is used for predicting the traffic volume when the user is located in the target cell according to the current position of the user and the state of the user's current use of services;

The processing module 72 is configured to switch the target cell to an energy-saving state according to the traffic when the user is located in the target cell.

The cell processing apparatus according to the embodiment of the present invention can switch the target cell to an energy-saving state by determining the traffic volume when the user is located in the target cell, and when the traffic volume when the user is located in the target cell satisfies a preset condition, and can perform cell energy saving based on user analysis Processing, to realize that when performing energy saving processing on the target cell, consider the impact of users entering, staying and leaving the target cell on the traffic volume of the cell under the current actual situation, so as to more accurately understand the future network load of the target cell and improve the accuracy of cell energy saving. rate, reduce the chance of deviation, and improve network performance and user experience.

Optionally, in this embodiment of the present invention, as shown in FIG. 8 , the prediction module 71 includes:

a first determining unit 711, configured to determine at least one user currently in the target area;

a first predicting unit 712, configured to predict the probability that each of the users is located in the target cell after a preset time interval according to the received trajectory data of each of the at least one user;

The second prediction unit 713 is configured to predict, according to the location of the target cell, the current location of each of the users, and the current state of service usage of each of the users, that each of the users will be located after the preset time interval. Traffic volume in the target cell.

Further, as shown in FIG. 4 , the processing module 72 includes:

The second determining unit 721 is configured to determine that the total traffic volume of the target cell is greater than a preset amount according to the user presence status information, the probability that each of the users is located in the target cell, and the traffic volume when each of the users is located in the target cell the probability of the traffic threshold;

The processing unit 722 is configured to switch the target cell to an energy saving state when the probability that the total traffic volume of the target cell is greater than a preset traffic volume threshold is less than a first preset probability threshold.

Optionally, the second determining unit 721 is specifically configured to:

According to the user existence state information, the probability that each user is located in the target cell, and the traffic volume when each user is located in the target cell, calculate various possible total traffic volumes of the target cell and each possible total traffic volume the probability corresponding to the quantity;

Calculate the probability that the total traffic volume of the target cell is greater than a preset traffic volume threshold according to a plurality of possible total traffic volumes of the target cell and a probability corresponding to each possible total traffic volume.

Optionally, the processing unit 722 is specifically configured to:

Determining the total traffic with the largest probability among the various possible total traffics of the target cell as the total traffic of the target cell;

According to the historical data of the target cell, predict the number of users in the target cell after the preset time interval and the RSRP average value of the reference signal received power of the target cell;

According to the total traffic volume of the target cell, the number of users in the target cell, the RSRP average value of the target cell, the time information after the preset time interval, and the preset resource utilization mapping model, determine the resource utilization rate of the target cell;

When the probability that the total traffic of the target cell is greater than the preset traffic threshold is less than the first preset probability threshold, and the resource utilization rate of the target cell is less than the second preset probability threshold, the target cell is switched to Energy saving state.

Optionally, the first prediction unit 712 is specifically configured to:

Predict the destination of each of the users according to the received trajectory data of each of the users and a preset trajectory prediction model;

Calculate the possible multiple cell transfer probabilities for each of the users to reach the destination from the current location, and select the largest cell transfer probability from the multiple cell transfer probabilities;

According to the maximum cell transition probability, the probability that each of the users is located in the target cell is determined.

Optionally, the second prediction unit 713 is specifically configured to:

According to the location of the target cell, the current location of each user, the current status of each user using services, and the preset service prediction model, it is predicted that after the preset time interval, each user will be located in The probability of using various service types in the target cell;

According to the probability of using various service types when each user is located in the target cell, and the traffic generated when each user uses various service types obtained from historical data, calculate when each user is located in the target cell. volume of business.

Optionally, the preset service prediction model is the service prediction model under the user's motion state or the service prediction model under the user's stationary state; the service prediction model under the user's motion state and the user's current service use state, and the user's current service usage state. After the current location is switched to the next prediction point, the service state switching probability is related; the service prediction model in the user's stationary state is related to the user's current location, a preset time interval, and the user's probability of using the service at the current location after a preset time interval in history related.

In addition, an embodiment of the present invention also provides a network device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the computer program is executed by the processor During execution, each process of the above cell processing method embodiments can be implemented, and the same technical effect can be achieved. In order to avoid repetition, details are not repeated here.

Specifically, as shown in FIG. 9 , an embodiment of the present invention further provides a network device, where the network device includes a bus 91 , a transceiver 92 , an antenna 93 , a bus interface 94 , a processor 95 and a memory 96 .

In this embodiment of the present invention, the network device further includes: a computer program stored on the memory 96 and executable on the processor 95, wherein the computer program can implement the following steps when executed by the processor 95:

According to the current location of the user and the state of the user's current use of services, predict the traffic volume when the user is located in the target cell;

The target cell is switched to an energy-saving state according to the traffic when the user is located in the target cell.

In FIG. 9, the bus architecture (represented by bus 91), which may include any number of interconnected buses and bridges, will include one or more processors represented by processor 95 and memory represented by memory 96 The various circuits are linked together. The bus 91 may also link together various other circuits such as peripherals, voltage regulators and power management circuits, etc., which are well known in the art and therefore will not be described further herein. Bus interface 94 provides an interface between bus 91 and transceiver 92 . Transceiver 92 may be a single element or multiple elements, such as multiple receivers and transmitters, providing a means for communicating with various other devices over a transmission medium. The data processed by the processor 95 is transmitted on the wireless medium through the antenna 93 , and further, the antenna 93 also receives the data and transmits the data to the processor 95 .

The processor 95 is responsible for managing the bus 91 and general processing, and may also provide various functions including timing, peripheral interface, voltage regulation, power management, and other control functions. Instead, memory 96 may be used to store data used by processor 95 in performing operations.

Optionally, the processor 95 may be a CPU, ASIC, FPGA or CPLD.

Embodiments of the present invention further provide a computer-readable storage medium on which a computer program is stored, wherein when the computer program is executed by a processor, each process of the foregoing cell processing method embodiments can be achieved, and the same technical effects can be achieved , in order to avoid repetition, it will not be repeated here.

Computer-readable media includes both permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology for storage of information. Information may be computer readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), Flash Memory or other memory technology, Compact Disc Read Only Memory (CD-ROM), Digital Versatile Disc (DVD) or other optical storage, Magnetic tape cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media, such as modulated data signals and carrier waves.

It should be noted that, herein, the terms "comprising", "comprising" or any other variation thereof are intended to encompass non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements, It also includes other elements not expressly listed or inherent to such a process, method, article or apparatus. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in a process, method, article or apparatus that includes the element.

The above-mentioned serial numbers of the embodiments of the present invention are only for description, and do not represent the advantages or disadvantages of the embodiments.

From the description of the above embodiments, those skilled in the art can clearly understand that the method of the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is better implementation. Based on this understanding, the technical solutions of the present invention can be embodied in the form of software products in essence or the parts that make contributions to the prior art, and the computer software products are stored in a storage medium (such as ROM/RAM, magnetic disk, CD-ROM), including several instructions to make a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in the various embodiments of the present invention.

The above are only the preferred embodiments of the present invention. It should be pointed out that for those skilled in the art, without departing from the principles of the present invention, several improvements and modifications can be made. It should be regarded as the protection scope of the present invention.

Claims (11)

1. A cell processing method, comprising:

predicting the service volume of the user when the user is positioned in a target cell according to the current position of the user and the current service using state of the user;

and switching the target cell to an energy-saving state according to the service volume of the user when the user is positioned in the target cell.

2. The method of claim 1, wherein predicting the traffic volume when the user is located in the target cell according to the current location of the user and the current service usage status of the user comprises:

determining at least one user currently in the target area;

predicting the probability of each user in a target cell after a preset time interval according to the received trajectory data of each user in the at least one user;

and predicting the service volume of each user when the user is positioned in the target cell after the preset time interval according to the position of the target cell, the current position of each user and the current service using state of each user.

3. The method of claim 2, wherein the switching the target cell to the energy-saving state according to the traffic volume of the user when the user is located in the target cell comprises:

determining the probability that the total traffic of the target cell is greater than a preset traffic threshold according to the user existence state information, the probability that each user is located in the target cell and the traffic of each user when the user is located in the target cell;

and when the probability that the total traffic of the target cell is greater than a preset traffic threshold is smaller than a first preset probability threshold, switching the target cell to an energy-saving state.

4. The method of claim 3, wherein the determining the probability that the total traffic of the target cell is greater than the preset traffic threshold according to the user presence status information, the probability that each of the users is located in the target cell, and the traffic of each of the users when located in the target cell comprises:

calculating a plurality of possible total traffic of the target cell and the probability corresponding to each possible total traffic according to the user existence state information, the probability of each user in the target cell and the traffic of each user in the target cell;

and calculating the probability that the total traffic of the target cell is greater than a preset traffic threshold according to the multiple possible total traffic of the target cell and the probability corresponding to each possible total traffic.

5. The method of claim 4, wherein switching the target cell to the energy-saving state when the probability that the total traffic of the target cell is greater than the preset traffic threshold is smaller than a first preset probability threshold comprises:

determining the total traffic of the target cell with the highest probability in multiple possible total traffic of the target cell as the total traffic of the target cell;

predicting the number of users in the target cell and the Reference Signal Received Power (RSRP) average value of the target cell after the preset time interval according to the historical data of the target cell;

determining the resource utilization rate of the target cell according to the total traffic of the target cell, the number of users in the target cell, the RSRP average value of the target cell, the time information after the preset time interval and a preset resource utilization rate mapping model;

and when the probability that the total traffic of the target cell is greater than a preset traffic threshold is smaller than a first preset probability threshold and the resource utilization rate of the target cell is smaller than a second preset probability threshold, switching the target cell to an energy-saving state.

6. The method of claim 2, wherein predicting the probability of each of the users being located in the target cell after a preset time interval according to the received trajectory data of each of the at least one user comprises:

predicting the destination of each user according to the received trajectory data of each user and a preset trajectory prediction model;

calculating various cell transition probabilities that each user is possible to reach a destination from a current position, and selecting a maximum cell transition probability from the various cell transition probabilities;

and determining the probability that each user is positioned in the target cell according to the maximum cell transfer probability.

7. The method of claim 2, wherein the predicting the traffic volume of each user located in the target cell after the preset time interval according to the location of the target cell, the current location of each user and the current service usage status of each user comprises:

predicting the probability of using various service types when each user is positioned in the target cell after the preset time interval according to the position of the target cell, the current position of each user, the current service using state of each user and a preset service prediction model;

and calculating the traffic of each user when the user is positioned in the target cell according to the probability of using various service types when each user is positioned in the target cell and the traffic generated when each user uses various service types obtained according to historical data.

8. The method according to claim 7, wherein the preset service prediction model is a service prediction model in a user motion state or a service prediction model in a user static state;

the service prediction model in the user motion state is related to the current service using state of the user and the service state switching probability after the user is switched to the next prediction point from the current position;

the service prediction model in the user static state is related to the current position of the user, a preset time interval and the probability of using the service at the current position after the user historically presets the time interval.

9. A cell processing apparatus, comprising:

the prediction module is used for predicting the service volume of the user when the user is positioned in the target cell according to the current position of the user and the current service using state of the user;

and the processing module is used for switching the target cell to an energy-saving state according to the service volume of the user when the user is positioned in the target cell.

10. Network device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the computer program, when executed by the processor, implements the steps of the cell processing method according to any of claims 1 to 8.

11. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the cell processing method according to any one of claims 1 to 8.