Error control codeword generating system and method based on a neural network
AssigneeTexas Instruments Incorporated
InventorGary A. Frazier
Life of this patent
Estimating the term…
Abstract
A communication system and method that translates a first plurality of information symbols into a plurality of code words, transmits the plurality of code words through a communication channel receives the plurality of code words transmitted through the communication channel, deciphers the plurality of code words transmitted through the communication channel into a second plurality of information symbols that correspond to the first set plurality of information symbols, wherein the plurality of code words are derived from a reverse dynamical flow within a first neural network.
Claims4
A method for generating a sequence of K N-bit binary code words for information transmission encoding and decoding, comprising the steps of:
- (a) providing a neural network with N neurons, each neuron has two states with output values 0 and 1, interconnection strengths T.sub.ij connecting the output of the jth neuron to an input of the ith neuron, and dynamical evolution given by
- V.sub.i .fwdarw.1 if .SIGMA..sub.j T.sub.ij V.sub.j <.theta..sub.i and
- V.sub.i .fwdarw.0 if .SIGMA..sub.j T.sub.ij V.sub.j >.theta..sub.i
- where V.sub.i is the output value of the ith neuron, V.sub.j is the output value of the jth neuron and .theta. is a threshold;
- (b) providing an initial N-bit binary code word U.sup.1 with components U.sup.1.sub.j ;
- (c) setting said interconnection strengths T.sub.ij =U.sup.1.sub.i U.sup.1.sub.j for all i and j except setting T.sub.ii =0;
- (d) inputting U.sup.1 into said neural network, and letting said neural network evolve to a stable state V;
- (e) taking said stable state V to define a new binary code word U.sup.2 ;
- (f) updating said interconnection strengths by adding U.sup.2.sub.i U.sup.2.sub.j to the current T.sub.ij for all i and j except i=j;
- (g) repeating steps (d)-(f) with U.sup.k replaced by U.sup.k+1 for k=1,2, . . . , K-1;
- and
- (h) applying said code words U.sup.1, U.sup.2, . . . U.sup.k to establish an information transmission encoder and decoder.
A method for generating a sequence of K N-bit binary code words for information transmission encoding and decoding, comprising the steps of:
- (a) providing a Hopfield neural network with N neurons but with dynamical evolution away from stored states defining the interconnection strengths;
- (b) providing an initial N-bit binary code word U.sup.1 ;
- (c) setting the interconnection strengths of said neural network T.sub.ij =U.sup.1.sub.i U.sup.1.sub.j for all i and j except setting T.sub.ii =0;
- (d) inputting U.sup.1 into said neural network, and letting said neural network evolve to a stable state V;
- (e) taking said stable state V to define a new binary code word U.sup.2 ;
- (f) updating said interconnection strengths by adding U.sup.2.sub.j U.sup.2.sub.j to the current T.sub.ij for all i and j except i=j;
- (g) repeating steps (d)-(f) with U.sup.k replaced by U.sup.k+1 for k=1, 2, . . . , K-1;
- and
- (h) applying said code words U.sup.1, U.sup.2, . . . U.sup.k to establish an information transmission encoder and decoder.
Dependencies are inferred from the claim text.