Device for recognizing information conveyed by a received signal

• US 5,909,675 A
• Filed: 10/13/1995
• Issued: 06/01/1999
• Est. Priority Date: 10/17/1994
• Status: Expired due to Fees

• Current Assignee: Alcatel Mobile Communication France (Nokia Corporation)
• Original Assignee: Alcatel Mobile Communication France (Nokia Corporation)
• Inventors: Chiodini, Alain
• Primary Examiner(s): Downs, Robert W.
• Application Number:
  US08/542,841
• Time in Patent Office: 1,327 Days
• Field of Search: 395/23, 395/20, 395/27, 395/22, 706/20
• US Class Current: 706/20
• CPC Class Codes: H04L 2025/03464 Neural networks
• CPC Class Codes: H04L 25/03165 using neural networks

Abstract

For recognizing information conveyed by a received signal, represented by convention by possible elementary forms of the signal to be transmitted, a device includes a correlator for establishing a correlation between the received signal and various possible forms of signal, in accordance with the convention. A neural network using correlation coefficients obtained from the correlator is trained by application to its input of correlation coefficients corresponding to a received signal conveying given information whilst imposing the given information at its output. The network supplies recognized information.

Claims

1. A device for recognizing information conveyed by a received signal, said information being represented by convention by possible elementary forms of the signal to be transmitted, said device including:

correlator means for establishing a correlation between said received signal and at least some of said possible elementary forms of said signal and generating correlation coefficients, and

a neural network receiving correlation coefficients from said correlator means and generating at its output recognized information conveyed by said received signal, said network being trained by application to its input of correlation coefficients, which coefficients have been generated by said correlator means upon receipt by said correlator means of a received signal conveying given information whilst imposing said given information at the output of said neural network.

2. The device according to claim 1 further including training means for continuous and adaptive training of the neural network by means of a training sequence comprising said given information and inserted into said information.

3. The device according to claim 1, for demodulating signals modulated by means of modulation with a memory aspect, wherein each of said various possible signal forms over a time period equal to n information symbols is obtained by:

generating a modulated sequence signal by applying said modulation to a modulation sequence, said modulation sequence including one of 2ⁿ possible sequences of said n information symbols, preceded by a plurality of symbols having the same value as the first symbol of said one sequence and followed by a plurality of symbols having the same value as the last symbol of said one sequence, and

thereafter isolating the portion of the modulated sequence signal that corresponds to said n information symbols.

4. The device according to claim 1, wherein said recognized information comprises a demodulation product of said received signal.