|Publication number||US20060156198 A1|
|Application number||US 11/348,864|
|Publication date||Jul 13, 2006|
|Filing date||Feb 7, 2006|
|Priority date||Sep 22, 2000|
|Also published as||US7151754|
|Publication number||11348864, 348864, US 2006/0156198 A1, US 2006/156198 A1, US 20060156198 A1, US 20060156198A1, US 2006156198 A1, US 2006156198A1, US-A1-20060156198, US-A1-2006156198, US2006/0156198A1, US2006/156198A1, US20060156198 A1, US20060156198A1, US2006156198 A1, US2006156198A1|
|Inventors||Jill Boyce, Haitao Zheng|
|Original Assignee||Lucent Technologies Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (19), Referenced by (26), Classifications (17)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This application is a divisional application of U.S. patent application Ser. No. 09/668,242, filed on Sep. 22, 2000 and is related to U.S. patent application Ser. No. 09/668,243 entitled “Radio Link Protocol (RLP)/Point-to-Point Protocol (PPP) Design for Wireless Multimedia Packet Networks that Passes Corrupted Data and Error Location Information Among OSI Layers,” assigned to the assignee of the present invention and incorporated by reference herein.
The present invention relates generally to wireless packet networks, and more particularly, to methods and apparatus for reducing lost or corrupted packets in such wireless packet networks.
It is inevitable that future wireless services will support Internet Protocol (IP)-based multimedia applications. For example, current and emerging wireless networks allow (i) a user to download information from the Internet using a wireless communication device, (ii) Internet-to-mobile or mobile-to-mobile videoconferences, (iii) streaming of video or audio information (or both) from the Internet to a wireless communication device, and (iv) electronic-commerce applications. In general, the multimedia services can be classified into two categories. Real-time services generally have delay constraints but can tolerate channel errors, and include interactive services, such as voice, voice-over-IP, packet video/audio, videoconference applications. Non-real-time services, on the other hand, are generally sensitive to channel errors but have more relaxed latency requirements, and include Web browsing, electronic mail and file transfer protocol (FTP) applications. It is noted, however, that non-real-time services for wireless systems should still provide a reasonable level of latency in order to be compatible with performance on a wired network, such as the Internet.
The end-to-end path of many wireless multimedia sessions, such as an Internet-to-mobile communication, involves a number of heterogeneous network technologies, with the multimedia packets being sent from an originating server, through the Internet and then over one or more wireless packet networks to the mobile destination. Network congestion on the Internet leads to packet loss and degraded quality. Thus, most Internet-based real-time multimedia services employ the well-known User Datagram Protocol (UDP) as their transport protocol. Compared to the Transmission Control Protocol (TCP), the UDP protocol has low overhead and no retransmission delay, which makes it attractive to delay sensitive applications.
A UDP packet typically includes a header, containing source and destination address information, as well as a payload (the actual application data). The UDP protocol employs a cyclic redundancy check (CRC) to verify the integrity of packets, in a known manner. The UDP protocol can detect any error in the packet header or payload and discard the packet if an error is detected. Packet transmission based on the UDP protocol on the Internet is a “best efforts” protocol, where network congestion yields packet loss. When a packet fails to arrive before its processing time, the UDP protocol declares the packet as lost. Therefore, at the receiving host, packets are either perfect or completely lost.
Wireless packet networks encounter packet losses between a base station and a mobile receiver as a result of channel errors and network congestion. Furthermore, such packet losses can be random or bursty, depending on the environment, rate-of-motion and network loading. Therefore, it has been recognized that use of the UDP protocol in a wireless network will cause considerable packet losses, and as a result, poor audio/video quality and increased power consumption. The inefficiency of the UDP protocol in wireless networks arises from the discarding of a packet containing only a small portion of corrupted data. As such, the UDP protocol also discards error-free data within the packet. Indeed, current and emerging multimedia coding technologies are focusing on improved error resilience, such that the media decoder can tolerate a certain number of channel errors. Thus, a need exists for a revised UDP protocol that reduces or avoids unnecessary packet discarding.
When wireless channels have a fairly high bit error rate, packet-level forward error correction (FEC) coding techniques, such as those described in R. Blahut, Theory and Practice of Error Control Codes (Addison-Wesley, 1983), provide an effective way to mitigate channel unreliability and improve media quality. Typically, FEC techniques apply Maximal Distance Separable (MDS) codes, such as Reed-Solomon (RS) codes, across the packets to recover lost packets. Generally, an FEC encoder typically chooses k information packets and generates n−k parity packets of length n to construct an (n, k) RS code. For IP transmission, the packets are numbered and are assumed to arrive perfectly or never arrive at all. The missing packets can be detected by the receiver and declared as erasure packets. An (n, k) RS code can correct (n−k) erasures and thus recover up to (n−k) packet losses.
In a wireless network employing the UDP protocol, an FEC decoder would use only the packets that were received perfectly, and lost packets are considered erasures. However, the UDP protocol yields high packet loss rates even under low and medium physical layer data losses. Therefore, the (n−k) value employed by the FEC encoder should be sufficiently large to effectively reduce the packet loss, which implies increased overhead and less efficiency. On the other hand, in a wireless network employing the UDP Lite protocol, which performs a checksum based on the packet header, so that only corrupted packet headers result in packet loss, the FEC decoder may receive perfect or corrupted packets, and performs both error and erasure correction. Since MDS codes provide twice the erasure recovering capability compared to error correction capability, an (n, k) packet code can recover up to (n−k)/2 erroneous packets within every n packets.
We have recognized that the above-described shortcomings of the UDP protocol may be overcome by a revised UDP protocol, referred to herein as the complete User Datagram Protocol (CUDP), that reduces unnecessary packet discarding.
More specifically, we have developed the CUDP that includes new packet handling procedures and reduces unnecessary packet discarding. The disclosed CUDP protocol utilizes channel frame error information obtained from the physical and link layers to assist the packet level error recovery. In addition, the CUDP supports packet level FEC coding in order to reduce information loss.
The present invention forwards each packet, as well as the channel frame error information, to a given application. In particular, the disclosed CUDP protocol further assists the packet level FEC decoding process by forwarding the locations of corrupted frames to the FEC decoder. A transmitter in accordance with the present invention optionally applies FEC techniques employing Maximal Distance Separable (MDS) codes to a group of packets, to achieve robustness against packet loss within the Internet and against packet error over wireless channels. At the receiver, the CUDP protocol forwards the frame error information, as well as the packet data, to the application layer. An MDS decoder utilizes the frame error information to recognize the erasures within each packet.
The error information provided to the application layer can be represented in a number of forms. Upon receiving a packet, the CUDP protocol only performs a packet header CRC check (and not a payload CRC check). In one implementation, a set of logical transmission unit (LTU) error indicators associated with each packet is provided to the application layer (for FEC decoders requiring an erasure indicator). If the packet header is valid, the UDP layer forwards the indicator, the LTU size and the packet payload to the FEC decoder. In another implementation, a reformatted packet is provided to the application layer (for FEC decoders Recognizing Erasures). The frame (LTU) error information from the lower layers is incorporated in the packet payload. In this case, if a physical frame is corrupted, the payload within the frame is represented as a set of erasures, which can be recognized by the FEC decoder. For a valid packet header, the CUDP protocol passes the reformatted packet payload to the upper layers. An FEC encoder is also disclosed that encodes multimedia packets utilizing a packet-coding scheme.
At the receiver 140, the RLP can specify a limited number of retransmissions to compensate for LTU losses. The RLP forwards the received LTUs to the interface, e.g., Point-to-Point Protocol, to reconstruct the packet. For non-real-time services, PPP forwards the packets to the TCP layer, where the packet losses after RLP retransmissions can be recovered at the TCP level through packet level retransmission and congestion control. For real-time services employing UDP, upon receiving a packet from the PPP protocol layer, the conventional UDP layer performs a packet level cyclic redundancy check (CRC) to validate the information within the packet, including both the packet header and the payload. In this case, any LTU loss would result in the whole packet being discarded. Mathematically, the packet loss rate (PER) can be approximated as:
PER=1−(1−p)m ≈mp (for large m and small p),
where m is the number of LTUs per packet and p is the LTU error rate (LER) after retransmission. Therefore, using a conventional UDP in a wireless network will yield a considerable amount of packet loss, and as a result, poor video/audio quality and increased power consumption. The inefficiency of UDP in wireless networks arises from the discarding of a packet containing only a small part of corrupted data. As such, the UDP also throws out error-free data within the packet. Indeed, applications can utilize error-free data to recover corrupted data.
The reliable UDP (RUDP) was proposed to provide reliable ordered delivery of packets up to a maximum number of retransmissions for virtual connections. For a more detailed discussion of RUDP, see, T. Bova and T. Krivoruchk, “Reliable UDP Protocol”, Internet Draft, Network Working Group, <draft-ietf-sigtran-reliable-udp-00.txt>, incorporated by reference herein. Generally, the RUDP protocol calculates the CRC checksum on the packet header alone or the packet header and the payload. This flexibility makes the RUDP protocol suitable for transport telecommunication signaling.
Similarly, the UDP Lite protocol was proposed to prevent unnecessary packet loss at the receiver if channel errors occur only in the packet payload. For a more detailed discussion of the UDP Lite protocol, see, L. Larzon et al., “Efficient Use of Wireless Bandwidth for Multimedia Applications,” 187-193, MoMuc 99, San Diego (November 1999), incorporated by reference herein. The UDP checksum is constructed based on the packet header, so that only corrupted packet headers result in packet loss. The UDP Lite protocol delivers the packet payload to the upper layers, whether the payload is perfect, lost or erroneous. Compared to UDP, UDP lite only performs CRC on the packet header. Therefore, if corrupted, a packet payload will not result in the packet being discarded. However, the error location within the packet payload is unknown to the application, which can improve packet level FEC performance.
When wireless channels have a fairly high bit error rate, packet-level forward error correction (FEC) coding techniques, such as those described in R. Blahut, Theory and Practice of Error Control Codes (Addison-Wesley, 1983), provide an effective way to mitigate channel unreliability and improve media quality. These FEC techniques are currently being considered by the Internet Engineering Task Force (IETF) for supporting real-time multimedia communications on the Internet and over wireless networks. See, for example, D. Budge et al., “Media-Independent Error Correction Using RTP,” Internet Engineering Task Force Internet Draft (May 1997); S. Wenger and G. Côté, “Using RFC2429 and H.263+ At Low To Medium Bit-Rates For Low-Latency Applications,” Packet Video '99, New York, N.Y., USA (April 1999); M. Gallant and F. Kossentini, “Robust and Efficient Layered H.263 Internet Video Based on Rate-Distortion Optimized Joint Source/Channel Coding”, Packet Video '00 (Italy, 2000); S. Wenger and G. Côté, “Test Model Extension Justification for Internet/H.323 Video Transmission”, Document Q15-G-17, ITU Q15, Video Coding Experts Group (February 1999); and J. Rosenberg and H. Schulzrinne, “An RTP Payload Format For Generic Forward Error Correction,” Internet Draft, February 1999, available from http://info.internet.isi.edu:80/in-drafts/files/draft-ietf-avt-fec-05.txt, each incorporated by reference herein.
As previously indicated, the UDP and UDP Lite protocols do not perform well with packet based FEC. One significant reason is that the UDP and UDP Lite protocols ignore useful channel information from the RLP layer 330. The present invention recognizes that such information can be exploited to maximize FEC coding efficiency. However, the current protocol design does not support information communications from the RLP layer 330 to the PPP/IP/UDP layers and above. The present invention proposes a new system protocol design that allows the exchange of certain information in both directions among the layers 310. For a more detailed discussion of communications between the RLP and PPP layers in the conventional open system interconnection (OSI) model, see co-pending U.S. patent application Ser. No. 09/668,243 entitled “Radio Link Protocol (RLP)/Point-to-Point Protocol (PPP) Design for Wireless Multimedia Packet Networks that Passes Corrupted Data and Error Location Information Among OSI Layers,” incorporated by reference above. A revised UDP protocol, referred to herein as the complete User Datagram Protocol (CUDP), is disclosed that reduces or avoids the discarding of unnecessary packets by passing the LTU error information as well as the corrupted packets to the Packet FEC decoder. Depending on the implementation of the FEC decoder, the error information can be represented in two forms:
LTU Error Indicator: (For FEC Decoders Requiring Erasure Indicator)
Upon receiving a packet, the UDP protocol only performs a packet header CRC check (and not a payload CRC check), in a similar manner to the UDP Lite protocol. The LTU error information obtained from the lower layers is represented in terms of a set of error indicators that are associated with each packet. The error indicators point to the starting and ending location of the erroneous data. If the packet header is valid, UDP forwards the indicator, the size of LTU as well as the packet payload to the FEC decoder.
Reformatted Packet: (For FEC Decoders Recognizing Erasures)
Upon receiving a packet, the UDP protocol performs a packet header CRC check. The LTU (referred to herein as “frame”) error information from the lower layers is incorporated in the packet payload. In this case, if a physical frame is corrupted, the payload within the frame is represented as a set of erasures, which can be recognized by the FEC decoder. The erasure format depends on the system implementation. Under a valid packet header, UDP passes the reformatted packet payload to the upper layers. The proposed UDP protocol captures all of the available information, i.e., the error-free frames and the location of erroneous frames.
According to another aspect of the present invention, an FEC coding technique is disclosed that uses the available wireless frame error information. Conventional MDS codes were designed for Internet multimedia applications, where packet loss is congestion related. For wireless multimedia applications, most packets are partially damaged by channel errors. Thus, if a conventional UDP protocol was utilized, such partially damaged wireless packets would be discarded. Similarly, if a conventional UDP Lite protocol was utilized, such partially damaged wireless packets are forwarded to the application, but the location of the error is discarded, thereby leading to information loss.
Vertical Packet Coding (VPC) Scheme
The FEC encoder 400 selects k packets of length X units. The proposed system encodes multiple packets of the same size together. For real-time applications, the packets should have the same or similar delay constraint, e.g., packets correspond to a single video frame. The packets can be of different length. If so, they are bit stuffed to match the longest packet length. Indeed, the source coding and packetization scheme can be designed to generate packets with equal or similar size, as described in co-pending U.S. patent application Ser. No. 09/668,243 entitled “Radio Link Protocol/Point-to-Point Protocol Design for Wireless Multimedia Packet Networks that Passes Corrupted Data and Error Location Information Among OSI Layers,” (Attorney Docket Number Lu 7-1), incorporated by reference above. The channel encoder at the application layer takes one data unit from each packet and generates (n−k) parity units to construct (n−k) additional packets in a vertical packet coding (VPC) structure, shown in
The VPC scheme provides transparent Internet-to-Wireless communications. In this case, the transmitter at the Internet multimedia database 105 is unaware of the wireless network 200 further downstream, and performs the same coding scheme to support packet flows over the Internet or a wireless network. On the other hand, for a Wireless-to-Internet packet flow, the gateway 115 between these two networks should embed the frame error information in the packet, and forward it to the receiver. Another option would be for the gateway to perform packet FEC decoding based on the frame error information. The gateway 115 discards any packet with an unrecoverable frame error without sending it to the Internet 110. As such, the UDP protocol within the Internet remains unchanged. However, this action increases both computational complexity and transmission delay at the gateway 115.
Another advantage of the CUDP with VPC of the present invention is that even if the decoder fails, part of the erroneous packets can still be recovered. Using the scenario in
Long Vertical Packet Coding (LVPC) Scheme
(n, k) MDS codes achieve better error/erasure correction efficiency as n increases, for the same redundancy ratio (n−k)/n. The FEC encoder 400 can increase n by coding L multiple columns of data units together. This generates X/L coded streams, each corresponding to a (nL, kL) MDS codes. This coding scheme is referred to as long vertical packet coding (LVPC), shown in
Using the scenario shown in
Horizontal/Vertical Packet Coding (HVPC) Scheme
In a further variation, FEC coding is applied to code the wireless frames in a single packet. As such, the encoder 400 applies vertical packet coding across the packets, and then applies an outer horizontal coding scheme within each packet. Similarly, the decoder first performs horizontal decoding to recover frame errors within the same packets, and then performs vertical decoding. The horizontal packet coding requires knowledge of the frame size.
It is to be understood that the embodiments and variations shown and described herein are merely illustrative of the principles of this invention and that various modifications may be implemented by those skilled in the art without departing from the scope and spirit of the invention.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US5717689 *||Oct 10, 1995||Feb 10, 1998||Lucent Technologies Inc.||Data link layer protocol for transport of ATM cells over a wireless link|
|US6148422 *||Oct 7, 1997||Nov 14, 2000||Nortel Networks Limited||Telecommunication network utilizing an error control protocol|
|US6208620 *||Aug 2, 1999||Mar 27, 2001||Nortel Networks Corporation||TCP-aware agent sublayer (TAS) for robust TCP over wireless|
|US6289054 *||Apr 27, 2000||Sep 11, 2001||North Carolina University||Method and systems for dynamic hybrid packet loss recovery for video transmission over lossy packet-based network|
|US6377799 *||Jun 17, 1999||Apr 23, 2002||Ericason Inc.||Interworking function in an internet protocol (IP)-based radio telecommunications network|
|US6421387 *||Aug 10, 2000||Jul 16, 2002||North Carolina State University||Methods and systems for forward error correction based loss recovery for interactive video transmission|
|US6430233 *||Aug 30, 1999||Aug 6, 2002||Hughes Electronics Corporation||Single-LNB satellite data receiver|
|US6530055 *||Nov 12, 1999||Mar 4, 2003||Oki Electric Industry Co, Ltd.||Method and apparatus for receiving and decoding coded information, including transfer of error information from transmission layer to coding layer|
|US6542490 *||Jan 29, 1999||Apr 1, 2003||Nortel Networks Limited||Data link control proctocol for 3G wireless system|
|US6571212 *||Aug 15, 2000||May 27, 2003||Ericsson Inc.||Mobile internet protocol voice system|
|US6591382 *||Aug 17, 1999||Jul 8, 2003||Skyworks Solutions, Inc.||Performance improvement of internet protocols over wireless connections|
|US6651214 *||Jan 6, 2000||Nov 18, 2003||Maxtor Corporation||Bi-directional decodable Reed-Solomon codes|
|US6697352 *||Jul 15, 1999||Feb 24, 2004||Telefonaktiebolaget Lm Ericsson||Communication device and method|
|US6718500 *||Oct 15, 1999||Apr 6, 2004||Samsung Electronics Co., Ltd.||RLP communication device and method for mobile communication system|
|US6850519 *||Sep 3, 1999||Feb 1, 2005||Kabushiki Kaisha Toshiba||Communication node and packet transfer method|
|US6999432 *||Jun 28, 2001||Feb 14, 2006||Microsoft Corporation||Channel and quality of service adaptation for multimedia over wireless networks|
|US7031257 *||Sep 22, 2000||Apr 18, 2006||Lucent Technologies Inc.||Radio link protocol (RLP)/point-to-point protocol (PPP) design that passes corrupted data and error location information among layers in a wireless data transmission protocol|
|US7151754 *||Sep 22, 2000||Dec 19, 2006||Lucent Technologies Inc.||Complete user datagram protocol (CUDP) for wireless multimedia packet networks using improved packet level forward error correction (FEC) coding|
|US20020054578 *||Jun 28, 2001||May 9, 2002||Qian Zhang||Channel and quality of service adaptation for multimedia over wireless networks|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7233875 *||Nov 21, 2005||Jun 19, 2007||Sigmatel, Inc.||Test set for testing a device and methods for use therewith|
|US7328393 *||Apr 13, 2004||Feb 5, 2008||Cisco Technology, Inc.||Forward error correction in packet networks|
|US7822869 *||Oct 15, 2009||Oct 26, 2010||Patentvc Ltd.||Adaptation of data centers' bandwidth contribution to distributed streaming operations|
|US7966540||May 2, 2007||Jun 21, 2011||Sprint Communications Company L.P.||Transmitting error correction information|
|US8230316||Jan 22, 2009||Jul 24, 2012||Nevion Usa, Inc.||Forward error correction for burst and random packet loss for real-time multi-media communication|
|US8406134 *||Mar 26, 2013||At&T Intellectual Property I, L.P.||Scaling content communicated over a network|
|US8473833||May 22, 2012||Jun 25, 2013||Nevion Usa, Inc.||Forward error correction method|
|US8495421||Jan 24, 2008||Jul 23, 2013||Thomson Licensing||Method for packet-switching transmission of media data and device for processing media data with insertion of error information data|
|US8693406||Aug 9, 2007||Apr 8, 2014||Intel Corporation||Multi-user resource allocation and medium access control (MAC) overhead reduction for mobile worldwide interoperability for microwave access (WiMAX) systems|
|US8819259 *||Oct 14, 2009||Aug 26, 2014||Aster Risk Management Llc||Fast retrieval and progressive retransmission of content|
|US8819260 *||Oct 15, 2009||Aug 26, 2014||Aster Risk Management Llc||Random server selection for retrieving fragments under changing network conditions|
|US8819261 *||Oct 15, 2009||Aug 26, 2014||Aster Risk Management Llc||Load-balancing an asymmetrical distributed erasure-coded system|
|US8825894 *||Oct 14, 2009||Sep 2, 2014||Aster Risk Management Llc||Receiving streaming content from servers located around the globe|
|US8832292 *||Oct 15, 2009||Sep 9, 2014||Aster Risk Management Llc||Source-selection based internet backbone traffic shaping|
|US8832295 *||Nov 10, 2010||Sep 9, 2014||Aster Risk Management Llc||Peer-assisted fractional-storage streaming servers|
|US8874774||Oct 15, 2009||Oct 28, 2014||Aster Risk Management Llc||Fault tolerance in a distributed streaming system|
|US8874775||Oct 15, 2009||Oct 28, 2014||Aster Risk Management Llc||Balancing a distributed system by replacing overloaded servers|
|US8938549 *||Oct 15, 2009||Jan 20, 2015||Aster Risk Management Llc||Reduction of peak-to-average traffic ratio in distributed streaming systems|
|US8949449 *||Oct 15, 2009||Feb 3, 2015||Aster Risk Management Llc||Methods and systems for controlling fragment load on shared links|
|US20050229074 *||Apr 13, 2004||Oct 13, 2005||Cisco Technology, Inc.||Forward error correction in packet networks|
|US20110055420 *||Nov 10, 2010||Mar 3, 2011||Patentvc Ltd.||Peer-assisted fractional-storage streaming servers|
|US20110317543 *||Dec 29, 2011||At&T Intellectual Property I, L.P.||Scaling content communicated over a network|
|US20130155838 *||Feb 15, 2013||Jun 20, 2013||At&T Intellectual Property I, L.P.||Scaling content communicated over a network|
|DE102007004951A1 *||Jan 26, 2007||Jul 31, 2008||Deutsche Thomson Ohg||Verfahren zum paketvermittelten Übertragen von Mediendaten sowie Vorrichtung zum Bearbeiten von Mediendaten|
|EP2719105A2 *||Jun 8, 2012||Apr 16, 2014||Samsung Electronics Co., Ltd.||Apparatus and method for transmitting and receiving packet in broadcasting and communication system|
|WO2010018250A1 *||Jul 20, 2009||Feb 18, 2010||Universidad Politécnica de Madrid||Method for transmitting multimedia contents|
|International Classification||H03M13/00, H04W80/06|
|Cooperative Classification||H03M13/1515, H03M13/373, H04L1/0057, H04L1/0045, H04L1/0041, H04W80/06, H04L1/1838, H04L1/0072|
|European Classification||H03M13/37E, H03M13/15P1, H04L1/00B5, H04L1/00B7B, H04L1/00B3, H04L1/00B8|