Pilot spoofing detection for large-scale MIMO communication systems

Abstract

For large-scale MIMO systems, upon the investigation of pilot spoofing on system security capacity, a novel detection scheme is proposed and the contaminated channel state information (CSI) is corrected. Specifically, a two-parts pilot sequence is exploited to implement channel estimation. The first one is conventional, and the other is partially randomized on the top of the first one. Leveraging the difference between the statistical characteristic of the estimated CSIs in the two parts, a pilot spoofing detector is established based on the principle of the maximum likelihood ratio. Meanwhile, the contaminated CSI is purified with the obtained information in pilot spoofing detection, and thus system security capacity can be improved greatly. Finally, simulation results are presented to evaluate the superior performance of the proposed scheme.