Error detection/correction and fault detection/recovery – Pulse or data error handling – Digital data error correction
Reexamination Certificate
2000-05-19
2003-06-10
Ton, David (Department: 2133)
Error detection/correction and fault detection/recovery
Pulse or data error handling
Digital data error correction
C714S787000
Reexamination Certificate
active
06578172
ABSTRACT:
FIELD OF THE INVENTION
The invention relates to a method and an arrangement for decoding and channel correcting a convolutionally encoded signal in a receiver, by means of a feedback neural network, said signal being received over a transmission path.
BACKGROUND OF THE INVENTION
In telecommunication connections, the transmission path used for transmitting signals is known to cause interference to telecommunication. This occurs regardless of the physical form of the transmission path, i.e. whether the transmission path is a radio link, an optical fibre or a copper cable. Particularly in telecommunications over a radio path there are frequently situations where the quality of the transmission path varies from one occasion to another and also during a connection.
Radio path fading is a typical phenomenon that causes changes in a transmission channel. Other simultaneous connections may also cause interferences and they can vary as a function of time and place.
In a typical radio telecommunication environment the signals between a transmitter and a receiver propagate on a plurality of routes. This multipath propagation mainly results from signal reflections from surrounding surfaces. The signals that propagated via different routes reach the receiver at different times due to different transit delays. Various methods have been developed in order to compensate the interference caused by this multipath propagation.
In order to reduce the effects of interference caused by the transmission path, a digital signal is encoded so as to make the transmission connection more reliable. Thus the errors caused by the interference in the transmitted signal can be detected and, depending on the encoding method used, also corrected without retransmission. Conventional encoding methods used in digital telecommunication include, for instance, block coding and convolutional coding. In block coding the bits to be encoded are grouped into blocks at the end of which are added parity bits, by means of which the correctness of the bits of the preceding block can be checked.
In the receiver, the errors caused by multipath propagation are typically first corrected, for instance, by a linear transversal corrector and thereafter the convolutional code is decoded.
The efficiency of convolutional coding depends on the code rate and constraint length used. A large constraint length enables good error correction capability, but on the other hand, decoding by known methods is then very complicated.
In general, convolutional coding is decoded by using a Viterbi algorithm that has nearly optimal performance. However, the Viterbi algorithm has a drawback that its complexity increases exponentially as the constraint length increases. This restricts the constraint lengths available.
Another known decoding method is sequential decoding that is described in greater detail in
Digital Communications
, J. G. Proakis, 3rd edition, pp. 500-503. One drawback of this algorithm is that the decoding delay does not remain constant but varies.
Yet another known decoding method is a so-called stack algorithm that is described in greater detail in the above-mentioned publication
Digital Communications
, J. G. Proakis, 3rd edition, pp. 503-506. The performance of this algorithm is not so good as that of the Viterbi algorithm.
Of the known methods, the Viterbi algorithm has the best performance for decoding the convolutional code, but its implementation has turned out to be extremely difficult in practice as the constraint length increases. The difficult implementation of the complicated Viterbi algorithm by circuitry has restricted the constraint lengths available for use.
A separate channel corrector and decoder are a suboptimal solution. Use of channel data, in particular within the Viterbi algorithm when computing the metrics, leads to increased complexity and the implementation of practical applications is impossible.
BRIEF DESCRIPTION OF THE INVENTION
The object of the invention is to provide a method and an arrangement implementing the method such that the above-mentioned drawbacks can be solved. This is achieved by a method of the invention for decoding a convolutionally encoded signal received over a transmission path, which signal comprises code words and in which method a transmission channel is estimated, decoding is carried out by means of an artificial neural network, the neural network comprising a set of neurons which comprise a set of inputs and an output, the received code word set is conducted to the inputs of the neuron, at least some of the inputs of the neuron are multiplied, after multiplication some of the inputs of the neuron are combined, some of the output signals of the neural network neurons are fed back to the inputs of each neuron, initial values of the neurons are set and the network is allowed to stabilize.
Further, in the method of the invention the multiplication of the neuron inputs depends on the convolutional code used in signal encoding and on the estimated channel and on the fact that an estimate of the decoded and channel-corrected symbol is conducted from the output signal of a predetermined neuron to the output of the network after the network has reached a stabilized state, the set of code words in the shift register is updated and the above-described four last steps are repeated until the desired number of code words are decoded.
The invention also relates to an arrangement for decoding and channel correcting a convolutionally encoded signal received over a transmission path, the signal comprising code words and the arrangement comprising a neural network which comprises a set of neurons which comprise a set of inputs and an output, the received code words being applied to the inputs of said neurons and to at least some of the output signals of the neural network neurons, the neurons comprising means for multiplying at least some of the inputs of the neuron prior to combining means, the arrangement comprising means for estimating the transmission channel. Further in the arrangement of the invention, estimated channel data is applied to the inputs of the neurons, and a predetermined neuron is arranged to give an estimate of the channel-corrected and decoded symbol in its output signal.
The preferred embodiments of the invention are disclosed in the dependent claims.
In the solution of the invention, convolutional code decoding and channel correction are performed by means of a feedback neural network. Several advantages are achieved with the solution of the invention. Performance that is close to that of the Viterbi algorithm is achieved with the solution of the invention by means of considerably simpler circuitry. In the solution of the invention, equally complicated circuitry permits a larger constraint length and thus improved error correction over the Viterbi algorithm.
The neural network of the invention can be readily constructed by semiconductor technology, since the neurons of the neural network are very similar to one another in structure, only the input couplings vary. Consequently, to design and implement even a large network does not involve a great amount of work. The solution can also be advantageously implemented as a VLSI circuit, which makes it fast.
REFERENCES:
patent: 4885757 (1989-12-01), Provence
patent: 5134396 (1992-07-01), Sirat et al.
patent: 5166938 (1992-11-01), Chung
patent: 5168551 (1992-12-01), Jeong
patent: 5268684 (1993-12-01), Allen et al.
patent: 5293453 (1994-03-01), Frazier
patent: 5548684 (1996-08-01), Wang et al.
patent: 5632006 (1997-05-01), Peterson et al.
patent: 5764858 (1998-06-01), Sheu et al.
“Digital Communication”, J.G. Proakis, 3rd Edition, pp 500-506.
Hämäläinen Ari H.
Henriksson Jukka A.
Nokia Mobile Phones Ltd.
Perman & Green LLP
Ton David
LandOfFree
Method and arrangement for implementing convolutional decoding 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 arrangement for implementing convolutional decoding, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and arrangement for implementing convolutional decoding will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3115237