Quantizer Design for Energy-Based Source Localization in Sensor Networks

We consider energy-based source localization applications, where distributed sensors quantize acoustic signal energy readings and transmit quantized data to a fusion node, which then produces an estimate of the source location. We propose an iterative quantizer design algorithm that allows us to take into account localization accuracy for quantizer design in the framework of the generalized Lloyd algorithm. Since source coding methodologies based on the Lloyd algorithm suffer from the presence of numerous poor local optima depending on initialization of quantizers, we introduce an efficient initialization, the equally distance-divided quantizer (EDQ), designed so that quantizer partitions correspond to a uniform partitioning in terms of distance. Our simulations demonstrate that improved performance over traditional quantizer designs can be achieved by using our proposed application specific strategy.