Sparsity-aware multi-source RSS localization

We tackle the problem of localizing multiple sources in multipath environments using received signal strength (RSS) measurements. The existing sparsity-aware fingerprinting approaches only use the RSS measurements (autocorrelations) at different access points (APs) separately and ignore the potential information present in the cross-correlations of the received signals. We propose to reformulate this problem to exploit this information by introducing a novel fingerprinting paradigm which leads to a significant gain in terms of number of identifiable sources. Besides, we further enhance this newly proposed approach by incorporating the information present in the other time lags of the autocorrelation and cross-correlation functions. An interesting by-product of the proposed approaches is that under some conditions we can convert the given underdetermined problem to an overdetermined one and efficiently solve it using classical least squares (LS). Moreover, we also approach the problem from a frequency-domain perspective and propose a method which is blind to the statistics of the source signals. Finally, we incorporate the so-called concept of finite-alphabet sparsity in our framework for the case where the sources have a similar power. Our extensive simulation results illustrate a good performance as well as a significant detection gain for the introduced multi-source RSS fingerprinting methods.