Detection of Pilot Contamination Attack Using Random Training and Massive MIMO

摘要

Channel estimation attacks can degrade the performanceof the legitimate system and facilitate eavesdropping. It isknown that pilot contamination can alter the legitimate transmitprecoder design and strengthen the quality of the received signalat the eavesdropper, without being detected. In this paper, wedevise a technique which employs random pilots chosen from aknown set of phase-shift keying (PSK) symbols to detect pilotcontamination. The scheme only requires two training periodswithout any prior channel knowledge. Our analysis demonstratesthat using the proposed technique in a massive MIMO system, thedetection probability of pilot contamination attacks can be madearbitrarily close to 1. Simulation results reveal that the proposedtechnique can significantly increase the detection probability andis robust to noise power as well as the eavesdropper’s power.