Estimating multiple transmitter locations from power measurements at multiple receivers

We consider the estimation of the locations of multiple transmitters based on received signal strength measurements at a network of randomly-placed receivers. We generalize the expectation-maximization (EM) method to create a quasi EM algorithm for localization under lognormal shadowing. Simulated performance is compared to a state-of-the-art global optimizer and to random guessing. Results reveal that the proposed quasi EM algorithm outperforms both alternatives in median and ninety-fifth percentile error, especially as the number of receivers increases.