Impact of real ranging on algorithms for power allocation in localization processes

Possession of information related to a target position, moving inside an indoor limited area, turns out to be a crucial application in the context of wireless sensor networks (WSNs). Localization algorithms, designed for a scenario consisting of a set of fixed nodes called beacons with known positions, aim to provide accurate estimates of target position. As a great concern for battery-powered nodes, energy saving is a fundamental issue for an effective and practical deployment. To this end, power allocation (PA) among beacons has been considered for optimizing localization performance; In general, optimal PA is achieved by considering ranging measures as necessary inputs of the localization algorithm. In the literature, on one hand the common made assumption about ranging measures is their ideal values equal to their corresponding Cramer-Rao bounds but, on the other hand, at high signal-to-noise ratios, real ranging estimators are characterized by different lower limits on their performance mainly as a result of maximum sampling rate and computational load available in the sensors. In this paper, the impact of a real ranging method on PA procedure for an UWB sensor network is investigated and a computationally efficient simplification is proposed