A Variable Rate Slepian-Wolf Code Construction Using Concatenated Convolutional Codes

A variable rate Slepian-Wolf (SW) code is constructed which is vital for wireless sensor network applications. A practical scheme to construct SW codes is to use the syndrome of a channel code as a compressed representation of a codeword in presence of a side information (Pradhan and Ramchandran, 2003). The proposed solution is based on an efficient and practical algorithm to compute the syndrome of rate-compatible convolutional codes (RCPC). By using this algorithm, there is no need to compute the syndrome of punctured version of the mother code for each puncturing matrix which is complex. Instead, the the syndrome of the punctured code is designed to be the punctured version of the syndrome of the mother code using the same pattern of puncturing. The algorithm is general for all convolutional codes in Z_q. The strategy is also generalized for parallel and serial concatenated convolutional codes.