Random parity coding system
US5825808A · kind A · utility
Assignee
Inventors
Key dates
| Filing date | Apr 4, 1996 |
| Grant date | Oct 20, 1998 |
| Priority date | — |
| Expiry date | Apr 4, 2016 |
Classification
- Technology area (CPC H)Electricity
- CPC primaryH04L1/0045
- WIPO fieldDigital communication
- WIPO sectorElectrical engineering
Abstract
An encoder functions to perform a modulo-two dot product of an n-bit information word with a succession of n-bit projector words which are n-bit words chosen at random or generated pseudorandomly. These projectors are also known by the decoder. The parity of the successive dot products is sent as the output bits of the coder. The decoder receives the incoming bits which have been generated at the encoder and subsequently passed through a noisy channel. In the decoder, the projectors serve as successive n-bit addresses to a table. If the incoming i-th bit is a zero, the content of the indicated table location is incremented. If the incoming bit is a one, the content is decremented. After the table has been filled, the table contents are considered as a vector, V. The Hadamard transform is then taken. The largest (most positive) coefficient of the transform is identified. The table address corresponding to the location of the largest coefficient is the estimate of the decoded information word.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.