A novel ℓ1-regularized LS formulation for target localization and malicious anchor identification

Secure target localization in the presence of malicious anchors is a critical issue in wireless sensor networks (WSNs), where compromised anchors attempt to mislead the target to a false position by broadcasting incorrect self location information. In this paper, we explicitly incorporate anchors´ misplacements into the distance measurement model and explore the pairwise sparse nature of the misplacements. We formulate the secure target localization problem as an ℓ₁-regularized least squares (LS) problem, whose objective is to simultaneously locate the target as well as identify the compromised anchors. We establish the sparsity threshold which defines the upper bound for the number of identifiable malicious anchors, and propose a simple projected gradient search algorithm to solve this novel ℓ₁-regularized LS problem in WSNs. Simulation results and possible future extensions are also provided.