Error detection/correction and fault detection/recovery – Pulse or data error handling – Digital data error correction
Reexamination Certificate
2000-01-11
2003-12-02
Baker, Stephen M. (Department: 2133)
Error detection/correction and fault detection/recovery
Pulse or data error handling
Digital data error correction
C375S259000, C375S265000, C370S389000, C370S395430, C370S535000
Reexamination Certificate
active
06658620
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to error detection and correction coding for communication systems. In particular, the present invention relates to packet and burst data transmission using product codes and iterative decoding.
There is a continuing need for more efficient use of wireless digital transmission channels. With increased efficiency come many benefits, including increased overall data throughput, greater flexibility in service level options, and increased revenues. Within the constraints of available bandwidth and transmit power, the goal is generally to maximize the amount of data transported at a specified quality of service. The quality of service maybe expressed, for example, as a desired bit error rate (BER) or cell loss rate (in an Asynchronous Transfer Mode (ATM) system).
While continual improvement in quality of service is the topic of much research in the field, Shannon's channel capacity theorem imposes a theoretical limit on the BER performance of any code. The use of modern error detection and correction coding techniques has already dramatically increased digital transmission efficiency, however, established techniques do not yet approach the theoretical limits. Furthermore, the signal structure of modern communications systems complicates the search for the “most efficient” coding technique.
In part, complications arise due to the burst or packet nature of modern communication systems. As an example, in Time Division Multiplexed (TDM) and Time Division Multiple Access (TDMA) communications systems, each terminal transmits signals in the same frequency spectrum but in different time slots or frames so that the transmissions arrive in a preselected order at a receiver with no overlap. Each terminal, of course, generates data independently and at different times than the other terminals. Each frame gives rise to a potential discontinuity in the data stream decoded at the receiver.
The most powerful conventional coding technique, concatenating coding, suffers from a reduction in efficiency in the form of overhead needed to accommodate packet oriented transmissions. Efficiency loss occurs in part because the current forms of concatenated codes were intended for continuous transmission. In particular, a concatenated code typically uses a convolutional code as the inner code of the concatenated code. The discontinuous nature of the data stream at the decoder causes premature termination of the decoding process. The convolutional decoder at the receiver, for proper operation, therefore requires the transmitted frames to contain overhead (i.e., non information, non parity) “flush bits” that reset the decoder at the end of the frame in preparation for decoding data in the next periodic frame.
TDM systems will remain extremely popular for the foreseeable future. As it stands, many cellular telephone systems operate in a TDM/TDMA fashion. In addition, most satellite communications systems incorporate a TDMA/Frequency Division Multiplexed (FDM) uplink, and a TDM downlink. With the well entrenched nature of TDM communications comes the need for error detection and correction coding that increases efficiency beyond that obtained with established concatenated coding schemes.
A need has long existed in the industry for an improved error detection and correction coding structure for packet and burst transmission systems.
BRIEF SUMMARY OF THE INVENTION
One aspect of the present invention is a communication system that uses a product coded data packet in the data section of a discrete communication channel frame.
Another aspect of the invention is a communication system particularly adapted to efficient transmission of product coded Asynchronous Transfer Modem (ATM) cells.
A further characteristic of the present invention is an efficient transmission technique for packetized data using a product code adapted to a data section or data unit length.
A preferred embodiment of the present invention provides a communication subsystem for transmitting error correction coded data in packets. The communication subsystem comprises an input buffer storing unencoded data, a product coder coupled to the input buffer that accepts unencoded data from the input buffer and outputs product coded data packets having a packet size, and a time division transmitter transmitting the product coded data packets in a data section of a frame. The data section has a length substantially equal to an integer multiple of the packet size.
In one embodiment, the communication subsystem uses a (60, 53)×(63, 56) product coder and the integer multiple is 4. More generally, a (s, t)×(n, m) product code may be used with the product of t and m adapted to a data unit size. The parameter s is greater than t, and n is greater than m.
A preferred embodiment of the present invention is also found in a method for communicating error correction coded data in packets. The method includes storing unencoded data in an input buffer, product coding the unencoded data in the input buffer, and outputting product coded data packets having a packet size. The method also includes time division transmitting the coded data packets in a data section of a frame in which the data section length is substantially equal to an integer multiple of the packet size. As noted above, the integer multiple may be 4, as an example, and the data to be encoded may comprise ATM cells.
A novel time division multiplexed frame format for communicating error correction coded data is also disclosed. The frame format comprises a header section and a data section. The data section carries product coded data packets having a packet size and has a length substantially equal to an integer multiple of the packet size. As one example, the data section length may be 7560 QPSK symbols (15120 bits), and the data section may carry four (60, 53)×(63, 56) product coded data packets.
REFERENCES:
patent: 4675863 (1987-06-01), Paneth et al.
patent: 5663962 (1997-09-01), Caire et al.
patent: 5767912 (1998-06-01), Bunting et al.
patent: 5898708 (1999-04-01), Tateishi et al.
patent: 5930251 (1999-07-01), Murakami et al.
patent: 5943344 (1999-08-01), Keller et al.
patent: 6052370 (2000-04-01), Ogura
patent: 6118825 (2000-09-01), Ikeda et al.
patent: 6198775 (2001-03-01), Khayrallah et al.
patent: 6356537 (2002-03-01), Jaakkola et al.
patent: 6430722 (2002-08-01), Eroz et al.
patent: 6519732 (2003-02-01), Li
patent: 8-714028 (1996-08-01), None
patent: 11-145969 (1999-05-01), None
Text pages: Error Control Coding—Fundamentals and Applications, pp. 274-278, Shu Lin and Daniel J. Costello, Jr., Prentice Hall, Inc., 1982.
Berger Harvey L.
Saunders Oliver W.
Baker Stephen M.
McAndrews Held & Malloy Ltd.
Northrop Grumman Corporation
Whittington Anthony T.
LandOfFree
Burst and packet wireless transmission using product codes... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Burst and packet wireless transmission using product codes..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Burst and packet wireless transmission using product codes... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3176022