An efficient detection algorithm of pilot spoofing attack in massive MIMO systems

Abstract

In massive MIMO systems, the training phase provides opportunity for an eavesdropper to attack legitimate communications, which is known as pilot spoofing attack (PSA). This paper proposes an algorithm to detect the PSA based on the transmission of dedicated signals. In contrast to existing methods, the proposed scheme doesn’t rely on a prior knowledge of large-scale factors. Specifically, the user sends pilots to the base station, where the channel estimation is performed. After that, dedicated signals are constructed using the channel estimation result and then sent back to users. These signals are divided into two parts, where the first part helps acquire the large-scale fading factors and the second one is exploited for the PSA detection. In addition, performance analysis concerning the probability of detection is presented, particularly the impact of key parameters on it. Finally, numerical simulations are carried out to validate the proposed PSA detection method.