Utilization of multi-dimensional source correlation in multi-dimensional single parity check codes

Abstract

This paper proposes a joint source-channel coding (JSCC) technique that well utilizes multi-dimensional (MD) source correlation using MD single parity check codes (MD-SPCCs). The source is assumed to be described by the coupling of multiple first-order binary Markov processes. The knowledge about the source correlation is utilized in the channel decoding process where each component decoder utilizes a single dimension correlation of the MD source. To enhance performance and reduce the error floor, a rate-1 recursive systematic convolutional code is serially concatenated to the MD-SPCC via a random interleaver. Two decoding techniques are proposed for each component decoder, and the selection of the decoding technique depends on the strength of the source correlation, which may further enhance the performance of the proposed JSCC technique. Simulation results reveal that a significant performance gain can be achieved by exploiting the MD source correlation with the proposed JSCC technique compared with the case in which the source correlation is not utilized; more significant gains can be achieved with stronger source correlation, and with a larger dimensionality source correlation as well.