FI103464B - ATM cell utilization in a cellular network - Google Patents

Description

103464 Utilization of ATM cells in a cellular network

The invention relates to a method and apparatus for transmitting speech frames of a mobile communication system as packets in a packet network, in particular an ATM network.

The invention is described in connection with an ATM network, but it can be applied to other packet networks as well.

The invention is described in connection with speech processing and speech frames, but the same technique can also be applied to the transmission of music and video signals. Common to such signals is that the signal samples must be applied to the decoder isochronically, i.e. at substantially the same intervals as the samples formed in the encoder.

In a digital cellular system, the speech signal is encoded in some way before channel coding and transmission to the radio path. For example, in the case of the GSM system, digitized speech is processed frame by frame for periods of about 20 ms using different methods, resulting in a set of parameters describing the speech per frame. This information, i.e. a set of parameters, is coded and transmitted to the bearer. The speech coding algorithms used are RPE-LTP (Regular Pulse Excitation LPC with Long Term Prediction) and various code sensitive algorithms CELP (Code 20 Excited Linear Prediction), such as VSELP (Vector-Sum Excited Linear Prediction).

In addition to the actual coding, the following functions are built into digital speech processing: a) Transmitter-side Voice Activity Detection VAD, which can control the transmitter ... 25 only when there is speech to be transmitted (Discontinuous Transmission, DTX); b) evaluating the background noise on the transmitter side and generating the corresponding parameters, and generating noise (Comfort Noise) on the receiving side from the parameters in the decoder; and c) preventing acoustic rotation. Noise interrupts the connection to make it sound more comfortable than absolute silent V · 30.

·: · In a well-known GSM cellular system, the input of a speech encoder is! «» · R '* ·. either the 13-bit PCM signal from the network side or the A / D-converted 13-bit PCM from the audio portion of the mobile station. The speech frame resulting from the encoder output is 20 ms long and comprises 260 audio bits 35 formed by encoding 160 PCM coded speech samples. From the parameters of the speech frame, the Voice Activity Detector (VAD) determines whether the frame contains speech or not. If speech is recognized, there are 2 103464 radio links. frames derived from traffic frames are speech frames. After a speech burst and at intervals during speech breaks detected by the VAD, the traffic frames are SID frames (Silence Descriptor) containing noise parameters, whereby the receiver can also generate 5 of the original noises from these parameters during breaks.

The traffic frame thus includes a 260-bit speech block representing 20 msec of encoded speech / data or noise. In addition, the frame has 56 bits available for frame synchronization, speech and data indication, timing, and other information, so the total length of the traffic frame is 10,316 bits. The uplink and downlink traffic frames differ slightly for the 56 bits mentioned.

In wired packet networks, ATM technology is becoming more common. Asynchronous Transfer Mode (ATM) has been developed as a broadband ISDN transmission technology. In ATM data transfer, data is transferred in standard size 53 byte packets called ATM cells. In each ATM cell, five bytes are the cell header and the remaining 48 bytes are payload, or real information. ATM cells are specified in CCITT Recommendation 1.361 and CCITT Draft Recommendation 1.150. Simplified, indicating the user information to be transferred is cut into 20 fixed-length bit strings, and each bit string is placed in the information field of the ATM cell. The number of bit strings per time unit represents the transmission capacity required by the user. In addition, the following information is appended to the information field:: a more detailed heading to form a standard 53 byte ATM cell. An ATM cell is an independent data transport unit because it contains 25 indirectly information about the address of the recipient from which the recipient • ·. . ·. found online. There are several different service criteria defined in the ATM network, for which a large number of different parameters can be set. Such criteria «· · * are e.g. response time, bit error rate, and probability of packet loss.

ATM is a Connection Oriented packet network, which means that connections are established and terminated according to standardized practices. Contact between the two parties through the ATM network called · · to as ATM virtual channel (Virtual Channel). Advantages of ATM include 1 (1 · mm, including the flexibility to provide different services. For example, all bandwidths within an ATM network are equally natural within the physical layer capacity of 35 (between 1.5 and 622 Mbps in current technology).

«« I

3, 103464

Figure 1 shows the essential parts of the cellular mobile communication system for the invention. The mobile terminal MT (Mobile Terminal) communicates with the air interface Um over two radio stations 11 and 12. It is then assumed that the mobile station MT is not so close to the center of its cell as to hear only one radio station. The system compares the signals received by two (or more) radio stations and makes decisions about handover and macro diversity. The latter means that which of the signals transmitted through the two transmission branches is selected, whichever is better, according to one of the criteria - usually the bit error rate BER 10 (bit error ratio).

In a packet switched network, such as an ATM network, successive packets may travel from source to destination on different paths and have different propagation delay times. The ATM network is accessed through the User to Network Interface (UNI). The ATM network comprises ATM switches and transmission lines. 15 ATM switches direct packets to the destination indicated by their header. The packets that have traveled through the ATM network over different routes are combined at a connection point 13, which may be, for example, a Mobility Server installed on the network to support the mobile subscriber's mobility. The connection point 13 combines the packets passing from the radio stations 11 and 12 along different routes 20 into a common traffic stream and forwards this traffic stream to the destination, which is an B-subscriber terminal connected to the local exchange LE (Local Exchange) 14 of the PISN in Figure 1. the subscriber could equally well be another mobile communications area.

25 Consider the transmission in the uplink direction. Mat- • *. the mobile station MT transmits the air interface Um over the frame F (Frame) which speech. ···. in the case there is a speech frame, but in general the frame F may contain music, video signal, etc. Radio stations 11 and 12 receive each frame F and measure parameters describing the quality of the radio connection, such as signal strength and / or bit error rate BER.

• · * * ·] * As mentioned above, the ATM cell payload is 48 bytes.

♦: · Commonly used GSM full-speed speech frame is 316 bits, or about 40 bytes. This is relatively close to the length of the payload portion of the ATM cell. However, as mobile devices increase, device manufacturers must develop more efficient speech coding algorithms that can compress speech into a shorter speech frame without sacrificing quality. It is 4 103464 expected that the so-called. for next generation mobile stations, the speech frame length could be about one fifth of the payload of the ATM cell.

Referring to Fig. 2, according to the prior art, speech frames F are transmitted in payload portions of ATM cells such that each radio station 11,125, etc. transmits a speech frame in its own ATM cell 21-23 and 24-26. The oblique line represents the ATM cell header ) and the white part of the payload. The letters A, B, ..., J describe the different speech frames such that the first (oldest) speech frame of the period is A, the next is B, and so on.

One problem with the arrangement described above is the variation of propagation delays when isochronous communication is desired after combining two or more traffic streams. A straightforward solution to this problem would be to buffer traffic flows before connecting. This problem is illustrated in Figure 2 with the upper branch (from BTS1) being 6 packets (ATM cells or speech frame) behind the lower branch (from BTS2) due to a larger 15 propagation delay. The propagation delays are compensated by buffers 26 and 27 placed in front of the merge point 3. When only one packet (A) has been led along the upper leg 26, the lower leg 27 already has 7 packets (A ... G). Macro diversity means that the merge point 3 20 must select each speech frame from the buffer 26 and 27 where that speech frame is better according to some criteria (e.g., has a lower bit error rate). However, macro diversity cannot be used unless it is accepted that due to the higher propagation delay of the upper branch, transmission must be delayed (in the example of Figure 2) by six speech frames.

Another problem arises if the ATM cell is lost for some reason. . ·. arum. Similarly, a straightforward solution to this problem would be re-. ···. transmission where the recipient - upon discovering that the cell sequence is missing • · · one cell - asks the sender to resend the missing cell.

... If the variation of ATM cell propagation delays between different paths is • · <I * .. '30 greater than Cell Interval, the need for buffering increases. Possible uu-; delenium transmission significantly increases the delay and, at the same time, the water required for buffering.

It is therefore an object of the invention to provide a method and apparatus implementing the method so that the aforementioned problems associated with ATM cell transmission can be resolved. The objects of the invention are achieved by methods and systems which are characterized by what is stated in the independent claims. Preferred embodiments of the invention are claimed in the dependent claims.

The invention is based on the fact that the payload of ATM cells, which would normally utilize only one speech frame length, is filled with other information packets related to the same connection. If the payload portion of an ATM cell holds N whole speech frames, each speech frame is transmitted N times in successive ATM cells. The speech frame is transmitted for the first time when it is the newest speech frame in the ATM cell and for the last time when it is the oldest speech frame in the ATM cell. According to a preferred embodiment of the invention, the portion of the payload of an ATM cell that is still unused when the cell is filled with as many complete speech frames as possible is filled with parameters describing the connection quality, such as signal strength and / or bit error rate.

First of all, the method and system of the invention have the advantage that retransmission requests and retransmissions due to the disappearance of ATM cells are almost completely eliminated, thereby reducing the maximum value of ATM cell propagation delays. Another benefit is the harmonization of ATM cell propagation delays. Because the invention corrects some of the transmission errors that may occur in the ATM network and standardizes propagation delays, it allows the use of cheaper, lower-grade connections while still maintaining satisfactory connection quality.

The invention will now be described in more detail in the preferred embodiments. : with reference to the following drawings, of which:

Figure 1 shows the following essential parts of the cellular telephone network for the invention: 1 25; • · ·

Figure 2 illustrates transmission of speech frames according to the prior art; Fig. 3 shows a transmission of speech frames according to the invention; and

Figure 4 illustrates how an arrangement according to the invention overcomes a situation where multiple ATM cells are lost by an ATM network.

Referring to FIG. 3, according to the invention, speech frames are transmitted in payload sections of ATM cells such that each radio station 11, 12, etc. transmits a speech frame in as many consecutive ATM cells as the speech frame can be co-located. The ATM cell. In this example, one ATM. · '··. the payload portion of cell 31-36 holds 5 speech frames, and each radio station 35-11-12 transmits each speech frame in five consecutive ATM-6 103464 cells. The example of Figure 3 shows a situation in which ATM cells (34-36) from the lower leg (from radio station 12) are new to buffer 27, i.e. buffer 27 does not have corresponding speech frames H - N. Therefore, all speech frames H - N of next received ATM cell 34 Instead, L is placed in Buffer 5 27. Instead, in the upper branch (from radio station 11), the last speech frame of buffer 26 is E. The speech frames B-E of the next ATM cell 31 are already in buffer 26, so they can be discarded and stored in buffer 26 31 Newest Speech Frame F.

By comparing the technique of Figures 2 and 3, it can be seen that the arrangement of the invention 10 reduces the maximum size of buffers 26 and 27 required at the junction point 13. Although the memory saving is not very high, but if one or both subscribers have to move the connection point 13 on the network, it may also be necessary to move the state of the previous connection point, i.e. all speech frames in buffers 26 and 27. Further, the reduction in memory requirement is achieved by utilizing that part of the ATM cell which is not utilized at all in the prior art.

Referring to Figure 4, it is considered how the arrangement of the invention overcomes the situation whereby the ATM network for some reason loses three ATM cells 41, 42 and 45, which in Figure 4 are denoted by a large X. In this 20 examples, an ATM cell containing speech frames A through E has already been transmitted along the upper leg (from radio station 11). Next, radio station 11 transmits an ATM cell 41 having speech frames B to F, followed by an ATM cell 42 having speech frames C to G, etc. Correspondingly, along the lower leg (from radio station 12), ATM cells 44 to 46 are coming, which si-25 contain speech frames A to E, B to F, and C to G, respectively. 26 and 27 are, despite the loss of three ATM cells, · · · an uninterrupted speech frame sequence and a merge point 13 may select from buffers 26 and 27 according to a particular criterion, such as a bit error rate BER. better speech frame. This comparison is represented by dashed line 48.

When the merge point 13 receives an ATM cell containing a speech frame from one radio station 11 or 12, it cannot endlessly wait for the corresponding ATM cell containing the speech frame from the other radio station. It is possible that only one radio station has heard the transmission of a mobile station. It is preferable to set a maximum wait time T, which is also a maximum-35 target delay at which the transmission of ATM cells may be delayed at the junction point. If an ATM cell is received from one radio station and the corresponding ATM cell is not received from other radio stations during this instant T, the only received ATM cell is used.

In practice, radio stations 11 and 12 and connection point 13 include a programmed digital processor and memory, not shown separately in the figures. The practice of the invention is generally a corresponding change in the software of the processor in question. On the transmitting side, i.e., at radio stations 11 and 12, the change includes transferring the contents of the memory area corresponding to the payload of the ATM cell by one traffic frame so that the oldest traffic frame "" falls off "and the space 10 . On the receiver side, i.e., at the connection point 13, the change includes separating the traffic frames from the received ATM cells and storing each separated traffic frame in a buffer memory. If the corresponding traffic frame transmitted by the same radio station is already stored in the buffer memory, it is not stored a second time. When the connection point 13 transmits a traffic frame received from two different radio stations 11 or 12 in the direction of the B-subscriber 5, the connection point selects from the traffic frames transmitted along different radio paths the one with the best quality of radio link quality parameter. The bit error rate BER is generally used for comparison, but the technique of the invention 20 can be applied when any parameter describing the quality of the transmission path - air interface or wired network - is used.

It will be obvious to a person skilled in the art that as technology advances, the inventive concept can be implemented in many different ways. The words and examples used are intended to illustrate the invention and not to limit it. Thus, embodiments of the invention are not limited to the examples described above, but may vary within the scope of the claims.

• «« «« i «« • «#« · «« «« t «« «.

Claims (9)

A method of transmitting a digital mobile communication system's transmission signal as packets in a packet network, which packets are provided with a header portion and a payload portion, in which method: 5. The transmit signal is encoded into parameter quantities arranged in traffic frames; - the traffic frames and / or parts thereof are arranged in the payload portion of the packages; - the packets are sent to their destination where they are received and the traffic frames are separated from the payload and the signal corresponding to the transmission signal is returned; characterized in that: - in the payload portion of at least a portion of the packets, in addition to the most recently formed traffic frame, earlier traffic frames or parts of these ice which relate to the same connection are arranged until the payload portion of the package is substantially filled. 2. A method according to claim 1, characterized in that the part of the payload part which is not filled with traffic frames provides parameters indicating the quality of the connection in question, advantageously at least the bit error density. Method according to claim 2, characterized in that in the destination, where traffic frames which relate to the same connection and which: are sent via different transmission routes are received, by the traffic frames sent via those in. different transmission paths the traffic frame is chosen where any of the parameters indicating the quality of the connection, advantageously the bit error density, is best. T: Method according to claim 2 or 3, characterized in that the transmission paths comprise a radio path and that at least one of the parameters indicating the quality of the connection specifically indicates the quality of the connection on the radio path. 5. Node in a digital mobile communication network, in particular a base station (11, 12), comprising means for transmitting the traffic frames (F) of the mobile communication system as packets (21-26; 31-36) in the packet network, which packets "103464 have a header part and a payload portion, characterized in that the node (11, 12) further comprises means by which, in addition to the most recently formed traffic frame (ΑΝ), in the at least part of the payload portion of the package (31-36), earlier traffic frames or parts thereof are arranged. relating to the same connection until the payload portion of the package (31-36) is substantially filled. Node (11,12) according to claim 5, characterized in that it further comprises means by which parameters in the portion of the payload portion of the package (31-36) which are not filled with traffic frames (AN) are arranged which indicate the connection of the relevant connection. quality, advantageous at least the bit error density. A second node of a digital mobile communication system, in particular a mobility server (13) comprising: - means of receiving traffic frames (AN) transmitted as packets (21-26; 31-36) in a packet network from at least one first node (11) , 12); and ice - a buffer memory (26, 27) for storing the traffic frames (A-N) received; characterized in that the second node (13) further comprises: - means by which the traffic frames (A-N) included in the same packet (31-36) are separated; and - means by which each etched traffic frame (AN) is stored in the buffer memory (26, 27), unless a corresponding traffic frame (AN) transmitted by a first node (11, 12) is already stored in the buffer memory (26, 27). ). A second node (13) according to claim 7, characterized in that it further comprises means by which one of the traffic frames: T: (AN) is transmitted to the different first node points and which relates to the same connection the traffic frame is selected where any of the parameters indicating the quality of the connection, advantageously bit error density, is best. 9. A node according to claim 6 or 8 (11 -13), characterized in that c - -. The transmission paths comprise a radio path (Um1, Um2) and that ··· ···· at least one of the parameters indicating the quality of the connection, especially I f '· · ·' indicates the quality of the connection on the radio path. «