FI102344B - Integrated voice coding - Google Patents

Description

102344

Integrated speech coding

The invention relates to an arrangement according to the preamble of claim 1 for integrating the functions of mobile communication systems.

Figure 1 shows the essential parts of a cellularly implemented mobile communication system 5 according to the invention. Mobile stations MS (Mobile Station) communicate with base transceiver stations (BTSs). The base stations are controlled by base station controllers BSC (Base Station Controller), which are connected to the mobile switching centers MSC (Mobile Switching Center) via transcoders TC (Transcoder). The subsystem controlled by the base station controller BSC - which already includes the base stations BTSn controlled by it - is collectively referred to as the base station subsystem BSS (Base Station Subsystem). The interface between the exchange MSC and the base station subsystem BSS is called the A-interface. The part of the mobile communication system on the MSC side of the A-interface exchange is called the Network Subsystem (NSS).

15 The mobile switching center MSC takes care of the connection of incoming and outgoing calls. It performs the same type of tasks as a PSTN exchange in a public switched telephone network. In addition to these, it also performs functions specific only to mobile call traffic, such as subscriber location management, in cooperation with the subscriber registers of the network. The GSM system has at least (not shown) a home location register HLR and a visitor location register VLR as subscriber registers.

•; The mobile switching centers MSC are connected to other networks, such as the public switched telephone network (PSTN) and / or the integrated service network ISDN, etc.

::: In the prior art mobile communication system described above, a speech signal is processed and converted from one representation to another at several different points.

Echo Cancellation EC (Echo Cancellation) prevents the transmission of echo returning from the fixed telephone network (PSTN) to a mobile subscriber. These loudspeakers are usually located in the mobile switching center on the interconnectors.

• ♦ · 30 Transcoders TC convert speech from digital format • · v .: to another. The traffic between the exchange and the transcoder TC can be, for example, 64 kbit / s A-law encoded PCM data. Alternatively, coding according to the μ law can be used. The traffic between the transcoder TC and the base station controller BSC can be, for example, 13 kbit / s coded speech.

'' The rate matching for the data shall be performed at a rate of 64 kbit / s and a rate of 3,6; Between 6 or 12 kbit / s.

2 102344

Soft Handover (SH) together with Diversity Combining (DC) can be implemented, for example, as follows. In the downlink direction, the signal is transmitted to the mobile station via a plurality of base stations BTS, and in the uplink direction 5, signals from the mobile station MS are received through a plurality of base station BTSs. At the receiver, the speech frames are combined, for example, by selecting the best of the speech frames received along different routes according to some criteria. Suitable criteria are, for example, SINR (Signal to Interference plus Noise Ratio) and a bit error rate BER (Bit Error Rate) can be used. This function is initiated by the mobile services switching center MSC (or the base station controller BSC) after detecting that the mobile station MS has moved to a common field area of two or more base stations BTS.

The dashed arrows in Figure 1 illustrate that at the points indicated by the arrows - not necessarily at each point - processing steps related to soft handoff SH and macrodiversity DC are performed.

The above-mentioned measures for processing and converting the signal from one representation to another at several points in the system include run-20 saving similar processing. The problem with such a mobile communication system is that performing similar processing many times at several different points increases the complexity of the system.

: Complexity increases e.g. because the elements of the system in which said processing steps are performed each require their own signal processors, memories, etc. The processing steps performed separately can only utilize the results of the previous processing step, but they cannot take advantage of useful intermediate results. Adding Complexity .v. increase costs and degrade system reliability.

The object of the invention is thus to develop an arrangement suitable for a mobile communication system * \ 30, comprising the above-mentioned processing steps, so that the above-mentioned problems related to the complexity, cost and reliability of the system can be solved. The objects of the invention are achieved by an arrangement which is characterized by what is stated in the independent claims. Preferred embodiments of the invention are the subject matter of the dependent claims.

The invention is based on the fact that as many processing steps as possible of the mobile communication system are performed in an integrated manner, preferably in a single integrated network element. Such an integrated network element may be a separate unit located in the network hierarchy at the location of the transcoder TC according to Figure 1. The integrated network element can also be physically located in the mobile switching center MSC, however, so that the responsibilities of the device manufacturers according to the A-interface 5 are clearly defined. The advantage of the arrangement according to the invention over the solution according to the prior art is a simpler structure, which reduces costs and improves the reliability of the system. The invention is characterized by vertical integration, i.e. the integration of different functions. In order to increase the signal processing capacity, it is advantageous to install several signal processors in parallel. It is also possible to install several integrated network elements according to the invention in parallel, either so that the integrated network elements are located at the same point or alternatively so that they are connected, for example, to the base station controllers BSC. In connection with the invention, it is thus possible to speak of vertical integration 15, i.e. integration of functions and at the same time horizontal decentralization.

The synergy advantages are greatest if the integrated network element according to the invention is adapted to perform processing steps related to as many functions as possible. In some cases, however, it may be appropriate to leave some functions, such as echo cancellation, to a separate unit 20 and combine other functions with an integrated network element that may; be a separate element or connected to, for example, a base station controller BSC.

In principle, it is also conceivable for all functions to be • integrated on the NSS side of the A-interface, ie on the MSC side of the exchange. Such a change requires at least such major changes to the GSM system that it is no longer possible to talk about the GSM system.

• · · • · · • ·

The invention will now be described in more detail in connection with preferred embodiments, with reference to the accompanying drawings, in which: • · ·

Figure 1 shows a mobile communication system essential to the invention. 30 parts; Fig. 2 shows a preferred embodiment of the invention and an exemplary structure of an integrated speech coding unit according to the invention; Γ *. Figure 3 shows a flow chart of the operation of an exemplary * '* 35 integrated speech coding unit according to the invention.

4 102344

Referring to Figure 2, transcoding TC, echo cancellation EC, soft handover SH and / or macrodiversity DC are integrated into a network element called processing block ISP (Integrated Speech Processing). The I / O (Input / Output) block comprises all interfaces in the uplink and downlink 5 directions, including multiplexers and demultiplexers, which are not shown separately. Block I / O can be a conventional technique. The functions according to the invention can be implemented, for example, by digital signal processors DSP, which are connected to the memory MEM via the BUS. For reasons of efficiency, the BUS is preferably a parallel bus. For horizontal spreading, i.e. to achieve sufficient processing capacity, there are usually several signal processors DSP. Since all signal processors DSP have access to a common memory MEM, it is not necessary to perform processing steps related to one voice or data connection with the same signal processor DSP, but it is possible that a new task is assigned to the next free processor.

15 The same memory MEM may also be associated with speech frame processing logic FPU (Frame Processing Unit), which depending on the architecture of the equipment supplier may be wired logic or a digital processor with software. The support circuits SUP (Support) comprise all the support circuits necessary for the operation of the processors, such as clocks, interrupts, voltage supply, etc. The speech processing unit 20 SPU consists of a signal processor DSP and its software stored in the memory MEM.

In the arrangement according to Figure 2, the processing block ISP implements the '' tasks included in the functions TC, EC, SH and DC according to the prior art.

25 The basic function of transcoding TC is to convert speech samples, for example, between formats encoded in accordance with Act A and otherwise encoded formats.

Echo cancellation EC can take place, for example, as follows. Speech sample «· is coded in the form of A-law; the speech thus encoded is sent to the mobile switching center MSC; in addition, the encoded speech is stored in a memory MEM, from which it can be retrieved by the Speech Processing Unit SPU if necessary; when the same speech sample arrives echo from the direction of the mobile switching center MSC back to the transcoder TC, the speech processing unit SPU retrieves from the memory: MEM a sample corresponding to the speech frame in question and removes the echo signal from the downlink speech frame based on the sample. The speech processing unit SPU consists, for example, of a digital signal processor DSP with its peripherals. The memory MEM comprises at least read / write memory (RAM, 5 102344

Random Access Memory), but may also include Read Only Memory (ROM) and / or flash memory, a hard disk, etc. to store the software.

For soft handoff SH and diversity combining in the DC var-5 processing block, the ISP has at its disposal several samples arriving along alternative paths. The urban area with the highest interference also tends to have a higher number of simultaneously heard base stations than outside the cities. According to the state of the art, the best of several parallel frames is selected according to a certain criterion. As an alternative to selecting frames according to the invention, parts of several frames can also be combined, whereby by combining a frame can be obtained which is better than any single frame. By combining several samples, it is thus possible to form a signal whose quality is better than that of a signal which could have been formed from only one input signal.

Fig. 3 shows exemplary processing steps when diversity combining DC and echo cancellation EC are performed in the circuit of Fig. 2. In Figure 3, index i refers to consecutive speech frame samples and N parallel diversity branches. P describes a counter to ensure that all branches of diversity are processed. The index j and the stored samples EC are related to: 20 echo cancellation.

:; In steps 302 (1 ... N), samples of the speech signals are formed. In steps 304 (1 ... N), the samples are stored in memory, at addresses 1 ... N.

t I i: In steps 306 to 308, it is determined whether all parallel processes have been processed • and if so, in step 310, the samples are processed by signal processors DSP. 25 In step 312, the next sample is processed.

· The possible implementation of the processing step 310 is shown as steps 320 to 336. In step 324, a sample 1 ... N is read from the memory 1 ... N, which is stored in step 304. In step 326, the result information is generated, for example • · so that the ISP frame processing units FPU receiver. 30 samples of the speech frames and store them in the memory MEM. The speech processing unit SPU generates a signal according to the A-law (or: ···: μ-law) from the samples stored in the MEM. In step 328, the echo information EC is stored in memory and the signal is sent to the trunk line. In step 334, echo cancellation is performed by any suitable algorithm. In step 336, the sample is sent to Ti-35 wide.

When combining the functions of the transcoder TC and the echo canceller EC, it should be noted that these functions are normally performed on different sides of the A-interface 6 102344, i.e. possibly under the responsibility of different equipment manufacturers. In addition, it is possible that, under the control of the MSC, a call is connected between two telephones that perform echo cancellation independently. In this case, the echo cancellation would occur twice, which could degrade the signal quality. These 5 factors can be taken into account by a suitable negotiation practice, in which it is possible to negotiate on a connection-by-connection basis whether or not, for example, echo cancellation is used in the connection.

The speech coding integration according to the invention can also be implemented incompletely. In this case, for example, the echo cancellation EC 10 can be left to a separate unit and the integrated network element ISP can be adapted to perform the processing steps related to the other functions TC, SH and DC. The integrated network element ISP can be implemented as a separate unit, which is located in the location of the transcoder TC according to the prior art in terms of signal flow. The physically integrated network element ISP can be installed as a separate unit-15 or it can be placed in the same space as the exchange MSC. Alternatively, the integrated network element ISP can be installed in each base station controller BSC. Such a solution may be desirable in a situation where, in order to increase computing power, it is desired to decentralize digital processing from the horizontal, for example because it is not possible or desirable to implement a network element 20 that is efficient enough to handle the entire MSC traffic.

It will be apparent to one skilled in the art that as technology advances, the basic idea of the invention can be implemented in many different ways. The primary areas of application of the invention are the so-called third generation mobile communication systems, especially when using CDMA radio technology, where it is natural to take advantage of macrodiversity. However, the invention can be used in connection with existing systems, such as the GSM / DCS system, where applicable.

• ·

The invention and its embodiments are thus not limited to the examples described above but may vary within the scope of the claims.

• · · • · · • · ··· • · • · • ·

Claims (11)

An arrangement for integrating the functions of a mobile communication system, the functions comprising at least code conversion (TC), echo cancellation (EC) and soft switching (SH), preferably also diversity combination (DC), characterized in that utilizing the similarity between the processing functions of the different functions (TC, EC, SH and DC), the at least three different processing-related processing steps are performed with the same digital signal processor (DSP). Arrangement according to claim 1, characterized in that at least one of the various functions (TC, EC, SH and DC) uses the intermediate results of the other functions. Arrangement according to claim 1 or 2, characterized in that all processing steps (TC, EC, SH and DC) belonging to all functions are performed with the same digital signal processor (DSP). Arrangement according to claim 1 or 2, characterized in that in order to increase the processing capacity, several digital signal processors (DSPs) are arranged in parallel, and each signal processor (DSPs) performs all functions performed in the arrangement (TC, EC, SH and DC). . 5. Arrangement according to any one of claims 1 to 4, characterized in that the functions include soft switching and diversity combining (DC), and that at least in individual speech frames the output signal is formed by joining parts of a plurality of parts. Speech frames arrive along different roads. 6. Integrated network element (ISP) for a mobile communication system :: whose functions include at least code conversion (TC), echo cancellation (EC) • · v: 25 and soft switching (SH), preferably also diversity combination (DC), since the network element (ISP) comprises means (I / O) for connection to the connection line: V: ring, means (FPU) for processing speech frames, memory (MEM) for storage of: 1 · 1; signal sample, and at least one digital signal processor (DSP), support circuits (SUP) and one bus (BUS) joining the parts, characterized in that at least one digital signal processor (DSP) is arranged to execute at least - • · ·. · .1 tone three functions (TC, EC, SH and DC) in the mobile communication system using closed processing steps. Integrated network element (ISP) according to claim 6, characterized in that it comprises a plurality of digital signal processors (DSPs) which operate via the bus (BUS) in contact with the common memory (MEM). An integrated network element (ISP) according to claim 6, characterized in that it is arranged in the mobile telephone exchange (MSC) of the mobile communication system. An integrated network element (ISP) according to claim 6, characterized in that it is arranged in the base station controller (BSC) of the mobile communication system. Mobile communication system, characterized in that it comprises at least one integrated network element (ISP) according to claim 6. 11. Mobile communication system, characterized in that it comprises a plurality of integrated network elements (ISPs) according to claim 6. · «· · · · · · · · · · · · · · · ♦