IL200694A - Method and device for forwarding frames in a communications network - Google Patents

Description

Method and Device for Forwarding Frames in a Communications Network Veraz Networks Ltd. n"ji-i op-nun TXT I Inventors: David Matas oon -in iD'N'snn D025/IL FIELD OF THE INVENTION The present invention relates in general to telecommunication systems and methods for their management, and particularly to systems and methods for efficiently forwarding frames of packet switched coded communications .

BACKGROUND OF THE INVENTION In coded packetized communication such as a Voice over Internet Protocol (VoIP) system, one or some voice frame data, which are obtained by encoding a voice signal, are gathered to form a packet. After adding some further information to the packet, such as generation time, sequence number etc., the latter is transmitted along a transmission path, e.g. the Internet. The packets are successively transmitted along the transmission path and arrive at a processing point, e.g. a receiver. Typically, the processing point is provided with a buffer (queue) for re-arranging the packets received in accordance with their transmission time to allow overcoming different arrival delays resulting from various network problems such as congestion of various links, and when the packets are pulled from the buffer, they are decoded.

However, one of the major problems associated with such type of a network is, packets that either do not arrive to their final destination (e.g. being discarded along the way due to congestion or any other reason) or packets that finally arrive to their final destination but with such a delay that for practical purposes they may be considered as if they did arrive. When a packet does not arrive on time to the reception buffer, an attempt will typically be made, to derive the information included in that missing packet by interpolation processing using the prior frames received, a process which is called error concealment processing.

Several attempts were made to overcome these problems and enable reception of the signal in a good quality. Among the attempts made is US 20020169859A1 which describes a voice decoding apparatus with packet error resistance, which, in case a packet is not received at the appropriate time when it should be decoded, the decoded signal and a filter memory value would be calculated at that time by using a concealment processing. However, in case a packet is later received (a delayed packet), the packet would nevertheless be used in recalculating the filter memory value for frames that were formed later than the time stamp of the lost packet. This way, it would be possible to reduce/remove the long-term deterioration effect caused by the concealment processing in the filter memory value. However, no solution is provided by this publication to many cases where the packet recalculation of the filter memory value cannot be made .

WO 0230098 describes a method whereby voice is sampled and encoded to produce data that represents speech prior to its transmission. Adaptive multi-rate (AMR) speech codecs represent generation of coding algorithms that are designed to work with inaccurate transport channels, such as wireless transmission channels. The AMR speech codec has built-in mechanisms that make it tolerant to a certain level of bit errors introduced by the transport channel. Therefore, would be possible to restore the original speech with some degradation even though the coded speech is received with some bit errors.

In a publication entitled "Packet Loss and Control for Voice Transmission over the Internet" by Henning Sanneck, GMD Research Series No. 8/2000, of GMD -Forschungszentrum Informationtechnik GmbH, a further step was made. Relying on the fact that some of the coded voice frames, namely, frames where there has been a transition state form unvoiced signal to a voiced signal, comprise more important information than other frames, it has been suggested to provide these frames with extra protection. Two solutions were proposed in this publication. The first, to attach a replica of each of the so-called important frames that is carried by the N-th packets, to the N+2th packets, so that if such an Nth packet is lost, the important packet can still be regenerated from its replica carried by the N+2th packet. The other solution suggested, is, that a XOR operation is carried on the Nth and N+lth packets, irrespective of the information comprised therein, and the result obtain from this operation is attached to the N+2th packet. Still, both these solutions have their drawbacks as they introduce a delay in the regenerated packet, and more importantly, they lead to a waste of bandwidth consumed by carrying the replicas of all important frames according to the first solution or even the bandwidth used for carrying the results of all those XOR operations.

US 7177304 discloses methods for prioritizing between voice data packets for discard decision purposes. A perceptual importance of a voice data packet relative to the others is determined at encoding. The relative importance is represented as a comparative discardability code in the packet. If a discard decision is made, it takes into account the comparative discardability code of the packet, thus preferring to discard the unimportant packets more frequently.

Dongyan Xu et al in "QoS-directed error control of video multicast in wireless networks", Computer Communications and Networks, 1999, Proceedings, Eight International Conference On Boston Ma. USA 11-13 Oct. 1999, Piscataway, NJ, pp. 257-262, present a scheme for a QoS-directed error control for video multicast in wireless networks, providing both high transmission efficiency as well as high error recovery rate for QoS-essential video frames.

US 2008130678 of the present Applicant, discloses a method for enabling successful reconstruction of an encoded signal conveyed along a communication path in a packet switched network. The method described comprises determining which of a plurality of frames comprising the encoded signal is an essential frame, identifying which packets comprise at least one essential frame, and adding protecting information to a packet that proceeds a packet identified as a packet carrying at least one essential frame .

SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide a method to allow reducing the bandwidth required for transmission of encoded signals along packet switched networks while retaining a good signal quality when reconstructing the encoded signals.

It is yet another object of the present invention to provide an apparatus capable of improving the quality of signals conveyed in their encoded form along packet switched networks.

It is still another object of the present invention to provide a method to eliminate the need to carry out re-encoding of the signal (if decoded in determining which of the frames will be discarded) , thus providing a cheaper solution with results of a higher quality, due to the fact that the forwarded encoded frames do not have to be decoded and re-encoded along the transmission path.

It is yet another object of the present invention to provide a method and a device that enable reducing the delay associated with transmission of frames in a packetized network while minimizing the required bandwidth for that transmission.

Other objects of the invention will become apparent as the description of the invention proceeds.

In accordance with one embodiment of the present invention, there is provided a method for conveying an encoded signal along a communication path in a packet switched network, comprising the steps of: (i) providing a plurality of encoded communication frames belonging to the encoded signal; (ii) retrieving one or more parameters associated with at least one of the plurality of encoded communication frames; for at least one encoded communication frame that is not to be forwarded towards its destination, forwarding information associated therewith towards the destination of the respective communication frame and discarding the at least one frame, and wherein the information being forwarded relates to the retrieved one or more parameters.

As will be appreciated by those skilled in the art, the information associated with the encoded communication frame that is forwarded towards the destination of the respective frame which relates to the one or more parameters may be information that is derived based on the values of these parameters, or may be the parameters themselves or a combination of these two alternatives.

According to a preferred embodiment of the invention, the step of retrieving the one or more parameters comprises parsing the at least one of the plurality of encoded communication frames, without having the at least one encoded communication frame been decoded.

In accordance with another embodiment of the invention, the step of retrieving the one or more parameters further comprises a step of duplicating the at least one of the plurality of encoded communication frames prior to the retrieval of the parameters. Preferably, this embodiment also includes a step of at least partially synthesizing a signal based upon the retrieved parameters and determining whether the respective encoded communication frames that has been at least partially synthesized, is to be discarded. If in the affirmative - forwarding information associated therewith (e.g. information based on the retrieved (parsed) parameters) towards the destination of the original respective encoded communication frame, otherwise - forwarding the duplicated communication frame in its encoded form towards its destination.

By still another embodiment of the invention, the one or more parameters relate to the intensity of a respective portion of the signal comprised in the corresponding encoded communication frame. Preferably the one or more parameters relate to the gain of the portion of the encoded signal comprised in the corresponding frame .

Preferably, the one or more parameters are associated with information contained in a frame that will eventually be discarded, which loss would cause a relatively substantial deterioration in the quality of at least one frame that proceeds the frame that will be discarded, when that at least one proceeding frames is decoded at its destination.

By still another preferred embodiment of the invention, the bandwidth required for forwarding the information that relates to the one or more retrieved parameters is substantially less than the bandwidth that would have been required to forward the frame in its entirety .

The term "frame" as used herein and throughout the specification and claims should be understood to encompass also cases where the invention relates to a sub-frame (i.e. part of the frame, e.g. a sub-group of parameters that can be decoded to allow producing part of the encoded signal) and not necessarily to the whole frame, and the method and device provided by the present invention may be applied to frames, sub-frames and any combination thereof, mutates mutandis .

The term "packet switched communication network" as will be used hereinafter, should be understood to encompass the various types of networks known in the art, such as synchronous and asynchronous transfer networks, for example ATM networks, Ethernet networks, IP networks, frame relaying networks, Metropolitan Area Networks (MAN) , Wide Area Networks (WAN) , Local Area Network (LAN) , packet over SONET/SDH networks, wireless networks and the like. It should be understood by those skilled in the art that although the bearer along which the packets are conveyed belongs to a packet switched network, still frames may be originated elsewhere, e.g. in a Mobile Radio Access Network ("RAN") .

According to another aspect of the present invention there is provided a device for use in a communication system which comprises: an input interface adapted to receive a plurality of encoded communication frames belonging to an encoded communication signal; a processor adapted to: retrieve one or more parameters associated with information comprised in the plurality of encoded communication frames; and determine which of the plurality of encoded communication frames should be discarded; and a transmitter adapted to: forward encoded communication frames that were determined as frames that should not be discarded; and forward information based on the retrieved one or more parameters that are associated with encoded communication frames that should be discarded or partially discarded, towards these frames' destination .

According to another embodiment of this aspect of the invention, the device provided further comprising means adapted to retrieve the one or more parameters by parsing at least one of the plurality of encoded communication frames, preferably, without having the at least one encoded communication frame been decoded.

By yet another embodiment of the invention, the device provided further comprises means for duplicating the at least one of the plurality of encoded communication frames prior to the retrieval of the one or more parameters. Preferably, the device is further adapted to forward a duplication of at least one of the received encoded communication frames, upon determining that the respective at least one encoded communication frame should not be discarded. In other words, after having two copies of each (or some) of the encoded communication frames, one is parsed to enable determining if the respective frame should be discarded, partially discarded or not at all, and in the latter case, where the encoded communication frame should not be discarded at all, the non-processed copy (i.e. the one that was not used for the retrieval of the one or more parameters) will be forwarded towards the frame's destination.

In accordance with another embodiment of the invention, the one or more parameters relate to the intensity of a respective portion of the signal comprised in the at least one encoded communication frame.

According to an embodiment of the invention, the encoded communication signal is a member selected from the group consisting of voice type, audio type and video type. Preferably, if the communication signal is of a voice type, the information to be forwarded in place of a frame determined to be discarded, would typically relate to a transition from unvoiced signal (e.g. fricative phoneme) or silence, to a voiced signal.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 presents a block diagram showing a structure example of an apparatus according to an embodiment of the invention; FIG. 2 illustrates another option of constructing a an apparatus according to another embodiment of the invention; and Fig. 3 presents a method of carrying out an embodiment of the present invention.

DETAILED DESCRIPTION OF THE INVENTION The present invention will be understood and appreciated more fully from the following detailed non-limiting examples taken in conjunction with the accompanying drawings.

Let us consider now Fig. 1 which presents an example of a schematic block diagram of an apparatus constructed according to an embodiment of the present invention.

Assuming that the signal that reaches receiver 4 of communication device 2 is a voice signal that has been encoded while using AMR (Adaptive Multi-Rate) narrowband codec which is the 3GPP mandatory standard codec for narrowband speech and multimedia messaging services over 2.5G/3G wireless systems based on evolved GSM core networks (WCDMA, EDGE, GPRS) . The encoded frames would be therefore 244 bits long frames.

The encoded frames received by receiver 4 are parsed by parsing means 6 (i.e. extraction means) in order to retrieve one or more parameters that are associated with a portion of the encoded communication signal comprised in the respective frames, and then the frames are processed by processor 8 in order to determine which of the arriving frames may be discarded or partially discarded (preferably when congestion conditions at the egress of device 2 require such a discard) .

Processor 8 carries out a full or partial synthesis of the signal comprised in the frame (e.g. synthesis of the signal's excitation vector), or any other algorithm that may indicate from the parsed parameters the quality degradation that may occur upon discarding said frame, and decision making block 10 determines whether this frame should be discarded or not. The determination may be based on information contained in the parameters retrieved and/or based on the synthesized (or partially synthesized) signal. Now, let us assume that the frame being processed is a frame that comprises mainly silence, but at the end of the frame there is a beginning of a voice phoneme. Now, since the frame comprises mainly silence it will be determined by decision making block 10 to be a frame that should be discarded. However, if the frame is discarded in its entirety, the information regarding the beginning of the phoneme would be discarded as well, and when the proceeding frame will eventually be decoded at its destination (or anywhere else along the path extending towards the proceeding frame' s destination), this information will be missing, and consequently, the quality of the decoded proceeding frame will be substantially degraded when compared with the frame's quality that could have been achieved, had the decoder been provided with the information included in the discarded frame.

The one or more parameters that are retrieved are preferably such that would allow device 2 to forward information relating to frames being discarded, towards a decoding device located downstream to communication device 2. This information is preferably essential information which loss would cause a relatively substantial deterioration in the quality of at least one frame that proceeds the frame to be discarded, when that at least one proceeding frames is eventually decoded. The one or more parameters may belong to one or more groups of parameters. One such group of parameters is a group that comprises parameters associated with the gain of that signal's portion e.g. Codebook gain or pitch gain or combined gains, or combination thereof, etc. Another possible group of parameters contains parameters associated with the spectrum of a respective portion of the signal comprised in the encoded communication frame, e.g. line spectral frequency ("LSF") parameters. Obviously, the one or more retrieved parameters may be members of one or more of the parameters groups, and are preferably selected so as to provide the optimal representation of the signal's gain and/or of the signal's spectrum.

By one embodiment of the present invention, the frames that have been determined by decision making block 10 to be non-discardable frames are packaged by packetizer 12 as well as essential information associated with frames being discarded, to be included in the next packets prepared by packetizer 12. Preferably, processor 8 also minimizes the essential information that will be added to that packet in order to save on the bandwidth used. As mentioned before, this example deals with AMR encoded frames that are 244 bits long, and a typical bandwidth required for the essential information of a discarded frame would be about 18 bits. The essential information that would be forwarded may be for example the values of the one or more parameters themselves, any manipulation thereof (e.g. changing the parameters' values, deriving information based on the parameters' values, etc.) or any other information that is associated therewith. The packets are then forwarded from device 2 by transmitter 14, and conveyed along the path extending from device 2 towards the destination of the respective packets. The information that is forwarded could be associated with spectrum related parameters, preferably when there is a somewhat substantial change in the spectrum of the signal.

Fig. 2 illustrates another embodiment of the present invention, in which, similarly to the example illustrated in Fig. 1, the signal reaches receiver 24 of communication device 22. The received encoded frames are duplicated by duplication means 26 and one of the two copies (i.e. the two copies being the original encoded frame and a duplication thereof) of each of the frames is parsed by parsing means 28 to retrieve parameters that are carried in and/or relate to each of the respective parsed frames. The frame is then forwarded to processor 32 where it is synthesized (or partially synthesized) , preferably by using the retrieved parameters, thereby allowing the decoding of the respective frame. Still, it should be noted that the retrieved parameters may but not necessarily be used for the synthesis step, as they may also, or in the alternative, be forwarded directly towards decision making block 34. Decision making block 34 determines which of the frames may be discarded and what information that relates to the parameters retrieved from frames that are due for discard, will be conveyed towards the frame's original destination. As will be appreciated by those skilled in the art, the decision of the decision making block may be based upon decoded information and/or the values of the retrieved parameters. In addition, the decision may also take into consideration certain information that was retrieved from one or more additional frames (e.g. values of parameters retrieved from a preceding frame) , or directly from parsed parameters.

Once a decoded frame has been determined that it should not be discarded, its duplicated encoded copy that has meanwhile been stored in buffer 28 is forwarded to packetizer 36. The parameters retrieved by parsing means 30 are also stored in buffer 28 and are forwarded to packetizer 36 based on the decision of decision making block 34 on which frames should/could be discarded. Obviously the decision making block may decide that all of the stored parameters may be forwarded to packetizer 36, or only part thereof. Again, the one or more parameters that are retrieved and stored are (or at least some of them) preferably such that would allow device 22 to forward information relating to frames being discarded, towards a decoding device located downstream to communication device 22. Here again, this information is preferably essential information which loss would cause a relatively substantial deterioration in the quality of at least one frame that proceeds the frame to be discarded, when that at least one proceeding frames is eventually decoded. Thus, the non-discarded frames are packaged by packetizer 36 as well as essential information associated with frames being discarded, to be included in the next packets prepared by packetizer 30. The packets are then forwarded from device 22 by transmitter 38, and conveyed along the path extending from device 22 towards the destination of the respective packets .

When the packets arrive at the receiving side of a corresponding communication device located downstream of communication device 22 (not shown in this FIG.), the information received is used in a frame discard concealment process, in order to restore the discarded encoded frame to allow conveying the restored the discarded frames to their destination. Preferably, this frame discard concealment process relies on information retrieved from neighbor frames, on the one or more parameters associated with the discarded frame and the rules applied by the transmitting side when taking the decisions on which frames should be discarded. Once the encoded frames that were discarded are restored, they may be decoded at the destination to obtain the communication signal comprised therein.

Fig. 3 demonstrates a method of carrying out an embodiment of the present invention, by which an encoded signal is conveyed along a communication path in a packet switched network. Upon receiving a plurality of encoded communication frames that belong to the encoded signal (step 50), one or more parameters associated with each of the received plurality of encoded communication frames, are retrieved (step 52) . A decision is taken whether the frame may be discarded (step .54) and if in the affirmative, determine for each of the frames to be discarded, if and what information associated with the retrieved parameters, be sent (step 56) . This information may be considered as information that represents the signal included in the respective frames that are about to be discarded, whether included in the whole frame, or only in a part of such a frame. For example, if the frame comprises mostly silence and only the very last part thereof is a beginning of a phoneme that extends to the next frame (s), the information will preferably relate only to this very last part, e.g. last sub-frame. The non-discarded frames as well as the respective information that represents the discarded frames are packetized into packets (step 58) and the packets are forwarded towards their respective destination (s) (step 60) .

As will be appreciated by those skilled in the art, the method described herein enables reducing the delay associated with transmission of frames in a packetized network, while minimizing the required bandwidth. According to prior art solutions, when frames are to be discarded, there are three typical ways to take the decision which of the frames should be discarded. One way is to discard every frame that is suspected not to comprise meaningful information. However, by following this way we end up with discarding frames that should not have been discarded and hence receiving at the destination a signal with rather poor quality. Another way is to refrain from discarding any frame if it is not absolutely clear that the frame should be discarded. Here, although there will be no impact on the received signal's quality, no substantial reduction of the the bandwidth required can ever be achieved. The third way is to introduce a delay in order to enable investigating in a discard of a frame will have an adverse effect on its proceeding frame. The penalty here is that although we can save on the bandwidth required and not impacting the information required to reconstruct the received frames, still the delay that is introduced has its own negative impact upon the received frames. AS opposed to these methods, because according to present invention, all operations are carried out on the current frame, the result is enabling a substantial reduction in the bandwidth required without sacrificing the quality of the signal received at the destination, due to the information that is conveyed that relates to the discarded frames and to the fact that no additional delay was introduced in order to determine the importance of each frame to its proceeding frame.

It is to be understood that the present invention has been described using non-limiting detailed descriptions of embodiments that are provided by way of example and are not intended to limit the scope of the invention. For example, although certain means have been described in the above examples, such as duplication means, parsing means, synthesizer, etc., still it should be understood that the functionality obtained by all or some of them, may be obtained by using one or more processors, carrying out these functionalities. It should also be understood that features and/or steps described with respect to one embodiment may be used with other embodiments and that not all embodiments of the invention have all of the features and/or steps shown in a particular figure or described with respect to one of the embodiments. Variations of embodiments described will occur to persons of the art.

It is noted that some of the above described embodiments describe the best mode contemplated by the inventors and therefore include structure, acts or details of structures and acts that may not be essential to the invention and which are described as examples. Structure and acts described herein are replaceable by equivalents which perform the same function, even if the structure or acts are different, as known in the art. Therefore, the scope of the invention is limited only by the elements and limitations as used in the claims. When used in the following claims, the terms "comprise", "include", "have" and their conjugates mean "including but not limited to"

Claims (11)

Claims :

1. A method for conveying an encoded signal along a communication path in a packet switched network, comprising the steps of: (i) providing a plurality of encoded communication frames belonging to the encoded signal; (ii) retrieving one or more parameters associated with at least one of the plurality of encoded communication frames; (iii) for at least one encoded communication frame that is not to be forwarded towards its destination, forwarding information associated therewith towards its destination and discarding said at least one frame, and wherein the forwarded information relates to the retrieved one or more parameters.

2. A method according to claim 1, wherein the step of retrieving the one or more parameters is carried by parsing said at least one of the plurality of encoded communication frames.

3. A method according to claim 2, wherein the step of retrieving the one or more parameters further comprises at least partially synthesis:ing said at least one of the plurality of encoded communi nation frames.

4. A method according to claim 1, further comprising a step of duplicating said at least one of the plurality of encoded communication frames prior to the retrieval of the one or more parameters.

5. A method according to claim 4, further comprising a step of determining whether said at least one of the plurality of encoded communication frames is to be discarded, and if in the affirmative - forwarding the information associated with said at least one communication frame towards the destination of the respective communication frame, otherwise - forwarding the duplicated communication frame in its encoded form towards its destination.

6. A method according to claim 1, wherein said one or more parameters relates to the intensity of a respective portion of the signal comprised in said encoded communication frame.

7. A method according to claim 1, wherein the forwarded information that relates to the retrieved one or more parameters, is information which loss will cause a relatively substantial deterioration in the quality of at least one proceeding frame when that at least one proceeding frames is decoded at its destination.

8. A device for use in a communication system which comprises : an input interface adapted to receive a plurality of encoded communication frames belonging to an encoded communication signal; a processor adapted to: retrieve one or more parameters associated with information comprised in said plurality of encoded communication frames; and determine which of the plurality of encoded communication frames should be discarded; a transmitter adapted to: forward encoded communication frames that were determined as frames that should not be discarded; and forward information based on the retrieved one or more parameters that are associated with encoded communication frames that should be discarded or partially discarded, towards these frames' destination .

9. A device according to claim 8, further comprising means adapted to retrieve the one or more parameters by parsing at least some of said plurality of encoded communication frames.

10. A device according to claim 8, further comprising means for duplicating the at least one of the plurality of encoded communication frames prior to the retrieval of the one or more parameters, and wherein said device is further adapted to forward a duplication of at least one of the received encoded communication frames which has been determined as an encoded communication frame that should not be discarded.

11. A device according to claim 8, wherein said one or more parameters relate to the intensity of a respective portion of the signal comprised in said at least one encoded communication frame.