Jensen–Shannon Divergence Based Location Verification in Emerging Wireless Networks Under Non-Gaussian Noises

In this article, the performance of location verification system (LVS) under non-Gaussian noises is analyzed. The existing works generally study location verification in Gaussian noise whereas in practice, non-Gaussian noise is common in wireless systems. Here, having obtained system and observational models, we prove the best operation point of LVS in non-Gaussian noisy environments, by exploiting an information theoretic criterion called Jensen-Shannon (JS) divergence, and considering Cauchy and Gaussian mixture noises. Numerical evaluations of theoretical findings show effectiveness of the proposed criterion in increasing the capability of LVS to correctly detect malicious users in non-Gaussian noise. Also, it is shown that this criterion is more efficient compared to other criteria such as Kullback–Leibler mutual information and error probability.