Distributed Joint Source-Channel Coding With Copula-Function-Based Correlation Modeling for Wireless Sensors Measuring Temperature

Wireless sensor networks (WSNs) deployed for temperature monitoring in indoor environments call for systems that perform efficient compression and reliable transmission of the measurements. This is known to be a challenging problem in such deployments, as highly efficient compression mechanisms impose a high computational cost at the encoder. In this paper, we propose a new distributed joint source-channel coding (DJSCC) solution for this problem. Our design allows for efficient compression and error-resilient transmission, with low computational complexity at the sensor. A new Slepian-Wolf code construction, based on non-systematic Raptor codes, is devised that achieves good performance at short code lengths, which are appropriate for temperature monitoring applications. A key contribution of this paper is a novel Copula-function-based modeling approach that accurately expresses the correlation amongst the temperature readings from colocated sensors. Experimental results using a WSN deployment reveal that, for lossless compression, the proposed Copula-function-based model leads to a notable encoding rate reduction (of up to 17.56%) compared with the state-of-the-art model in the literature. Using the proposed model, our DJSCC system achieves significant rate savings (up to 41.81%) against a baseline system that performs arithmetic entropy encoding of the measurements. Moreover, under channel losses, the transmission rate reduction against the state-of-the-art model reaches 19.64%, which leads to energy savings between 18.68% to 24.36% with respect to the baseline system