Open AccessPerformance and optimisation of agent compression and forwarding system for correlated sources
Author(s) -
Xu Dazhuan,
Chen Dan,
Zhang Ruidan
Publication year - 2019
Publication title -
iet communications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.355
H-Index - 62
eISSN - 1751-8636
pISSN - 1751-8628
DOI - 10.1049/iet-com.2018.5704
Subject(s) - relay , computer science , additive white gaussian noise , node (physics) , encoder , distributed antenna system , relay channel , topology (electrical circuits) , channel (broadcasting) , gaussian , computer network , antenna (radio) , power (physics) , telecommunications , electrical engineering , engineering , physics , structural engineering , quantum mechanics , operating system
A system for the correlated Gaussian sources to perform compression forwarding via the multi‐relay network in the additive white Gaussian noise channel is proposed in this study. The characteristic of the newly proposed system is that it is composed of two parts: an analogue sensor network from the sources to multiple relay nodes and a digital communication network from multiple relay nodes to the destination node. The sensors can only perform simple analogue signal transmission, but the encoder set up at each relay can carry out complex distributed source coding DSC and then send the coded digital signals to the destination node. In this study, the authors establish the rate–distortion function of DSC based on the Chief Executive Officer problem. Combined with the channel capacity theorem, an optimal power allocation scheme is first proposed. The quantisation signal‐to‐noise ratio (SNR) at the destination node is used to evaluate the performance of the system under different correlations between sources. For the case of independent sources, a closed‐form expression for the quantisation SNR is derived and they compare the optimised agent compress‐and‐forward system with the traditional amplify‐and‐forward system to verify that the proposed system can achieve better SNR performance.