Secure AoA-Based Localization in the Presence of Malicious Nodes

This work addresses angle of arrival (AoA)-based localization problem in the presence of malicious attackers that manipulate their measurements in the attempt to obstruct the localization process. Unlike the state-of-the-art (SoA) that considers the problem in a 2-dimensional space, the current work provides a solution in a 3-dimensional case. Based on geometry embedded within the considered problem, the work builds triangular relationships between a pair of reference points and a target to derive a weighted least squares (WLS) estimator whose solution is given in analytic form. Moreover, in contrast with the SoA which is exclusively focused on the localization problem, the solution of the WLS together with simple geometry are exploited to define a scheme for attacker detection. Lastly, by treating measurement noise as von Mises random variables, the current work derives a final expression of the Cramér-Rao lower bound (CRLB) for the secure localization problem. The proposed method is evaluated in terms of different parameters relevant to the problem, and numerical results corroborate its effectiveness in terms of computational complexity, localization accuracy and attacker detection.