Multiplex communications – Communication over free space – Combining or distributing information via time channels
Reexamination Certificate
2000-05-08
2001-04-24
Nguyen, Chau (Department: 2663)
Multiplex communications
Communication over free space
Combining or distributing information via time channels
C714S748000, C714S780000
Reexamination Certificate
active
06222835
ABSTRACT:
BACKGROUND OF THE INVENTION
FIELD OF THE INVENTION
The invention relates to a method for packet-oriented data transmission in a digital transmission system, in which an ARQ protocol and channel coding are used for error minimization with coders and decoders which emit soft information for evaluation of a retransmitted data frame, and to an configuration for performing the method.
A method and a configuration of the type mentioned initially are described in IEEE COMMUNICATIONS LETTERS, Volume 1, No. 2, pages 49-51, K. R. Narayanan et al., “A Novel ARQ Technique using the Turbo Coding Principle”. According to the method described there, logarithmic likelihood ratios (LLR) for the data to be decoded are determined during an earlier attempt to decode data in a data frame. If a decoding attempt fails, then the corresponding data frame is retransmitted. When the data in the retransmitted data frame are being decoded, the logarithmic likelihood ratios determined during the earlier decoding attempt are used as a-priori information in a forward-progressing manner.
In a conventional transmission path for transmission and for reception of digital data, the data from a data source are combined into blocks and are fed to a convolution coder. The coded data blocks are encrypted in an interleaver and are then transmitted via an ISI channel (ISI=Intersymbol Interference). At the receiver end, the data blocks are equalized in an equalizer. Once the equalizer has removed the interference, the equalized data blocks are processed in a deinterleaver, and this emits the data blocks, after deinterleaving, to a decoder, which then corrects the majority of the errors and passes its output signals to the data sink.
The aim in such a system is to use the channel coding to correct the influence of those errors which occur in the ISI channel during transmission. The coding increases the bit rate, and all the arriving information is subjected to channel coding. Depending on which channel is used, various us coding methods are used, for example different coding versions are used for data transmission than for voice coding. Forward error correction (FEC) may be used for error correction in the physical layer, and error correction with the ARQ protocol may be used for layer
2
, and these types of error correction lead to a transparent or nontransparent transmission class, respectively.
When error correction is carried out using the ARQ protocol (ARQ=Automatic Request for Retransmission), error correction is carried out deliberately for the data transmission. The ARQ method based on flow control leads to nontransparent transmission, and is particularly effective in conjunction with FEC. in the GSM system, ARQ is used in the RLP (RLP=Radiolink Protocol).
In ARQ protocols, data frames which, for example, are rejected during transmission are requested once again and are retransmitted. When the data are received once again, information which was generated when these data were previously received can be reused. The principle proposed for this purpose is known from IEEE Transactions on Communications, Volume COM-33, No. 5, May 1985, D. Chase, “Code Combining—A Maximum Likelihood Decoding Approach for Combining an Arbitrary Number of Noisy Packets”. In this case, the information is transmitted in packets which are coded with a code having a relatively high code rate, and which are repeated in order to achieve reliable communication, if the redundancy in the code is not sufficient to overcome the channel interference problems. The receiver combines packets that are subject to noise signals, in order to obtain a packet at a code rate which is sufficiently low that even channels having extremely high error rates can be combined reliably. The aim is to combine a minimum number of packets to reduce the code rate and the delay to the minimum that is required to decode a predetermined packet (data frame). This is the classic code combining method, which is to be improved. Diversity methods, for example metric combining, are used.
SUMMARY OF THE INVENTION
It is accordingly an object of the invention to provide a method and a device for data transmission in a digital transmission system using a packet switching service which overcomes the herein afore-mentioned disadvantages of the heretofore-known methods and devices of this general type in such a way that the effort for processing retransmissions of data frames within an ARQ protocol is reduced, and the error correction is improved.
With the foregoing and other objects in view there is provided, in accordance with the invention, a method for packet-oriented data transmission in a digital transmission system that initially includes forming a first data frame by combining data. The first data frame is then coded, transmitted, and received. If an attempt to decode the received first data frame fails, an ARQ protocol is used to retransmit the first data frame. Specifically, a second data frame is transmitted that is essentially identical to the first data frame. The transmitted second data frame is received. Soft information and decoded data are produced from the received first data frame and from the received second data frame using decoders. The received second data frame is evaluated using the soft information obtained from the received first data frame as a-priori information. The received first data frame is evaluated using the soft information obtained from the received second data frame as a-priori information.
If it has been channel-coded, the data frame that is retransmitted in accordance with the ARQ protocol (i.e. the second data frame) has already been decoded once (i.e. the first data frame). During the decoding process, a-posteriori and extrinsic likelihoods are obtained for the coded bits. The extrinsic or a-posteriori likelihoods may be used as a-priori information or likelihoods for reception. A-priori information can thus be obtained via the channel decoder and the ARQ protocol, and the receiver processes the additional information together with the retransmitted data frame in a forward-progressing manner. In other words, the ARQ protocol allows information which the decoder has produced to be used as a-priori information for the equalization of the retransmitted data frames, even if metric combining according to the prior art is impossible because of the characteristic properties of the structures.
Furthermore, the soft information obtained during decoding of the data frame n+1 is processed in a backward-progressing manner with one or more of the data frames 1 to n. The previously transmitted data frames are thus combined with the extrinsic or a-posteriori information which has been obtained with the data frame n+1, enabling the data frame to be sufficiently free of errors such that it no longer need be retransmitted. In order to carry out the method according to the invention, the extrinsic information which has been obtained during the transmission of a data frame is thus used both for conditioning data frames which are transmitted later, and for conditioning data frames which have already been previously transmitted.
In accordance with an added mode of the invention, the method is carried out alternately in the forward-progressing and in the backward-progressing manner.
In accordance with an additional mode of the invention, the method is repeated until no retransmission within the ARQ protocol is requested.
In accordance with another mode of the invention, in order to limit the number of repetitions that a data frame is transmitted, the process is terminated after a predetermined number of repeated transmissions of the data frame.
In accordance with a further mode of the invention, the soft information that is processed in the equalizer is processed by using the soft information directly in the algorithm. This method has been proven in practice for the combination of transmitted data frames, and is thus also advantageous for the present invention.
DE 42 24 214 C2 discloses a method for source-controlled channel decoding by
Emmer Dieter
Franz Volker
Greenberg Laurence A.
Lerner Herbert L.
Nguyen Chau
Siemens Aktiengesellschaft
Stemer Werner H.
LandOfFree
Method and configuration for packet-oriented data... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and configuration for packet-oriented data..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and configuration for packet-oriented data... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2497773