Reliable communication and sensing via parallel redundancy in noisy digital receivers

Recent work in the statistical physics literature has demonstrated an intriguing noise-enhanced signal transmission phenomenon known as Suprathreshold Stochastic Resonance (SSR). This occurs in a simple network of threshold devices designed to capture the most important properties of biological signal transduction in sensory neurons, as well as artificial sensors like sonar arrays. It has been demonstrated, by calculations of mutual information in the model, that the best signal transmission in the model occurs for an input SNR around 0 dB. The question of whether this model has any relevance for modern communications systems has not previously been addressed. In this paper the operational significance of previously reported mutual information calculations in this model is discussed in the context of modern communication systems such as distributed sensor networks, multi-antenna digital receivers, and UWB analog-to- digital converter circuits.