Pilot Contamination Mitigation Techniques in Massive MIMO Systems: A Precoding Approach

Abstract

A massive MIMO system comprises of base stations with a very large number of antennas serving a considerably smaller number of users and providing substantial gains in spectral and energy efficiency in comparison to conventional MIMO systems. However, these benefits are limited by pilot contamination which is caused by the use of training sequences for channel estimation. This negative effect has given rise to various research works on schemes to mitigate pilot contamination and among them are precoding techniques.

This thesis reviews some of the precoding techniques that mitigate pilot contamination and studies the effect of pilot contamination on the performance of massive MIMO systems through simulations. It was found that pilot contamination leads to a severe degradation of the network performance. Furthermore, as the number of antennas at the base station increases, the effect of pilot contamination is more prominent on the probability of outage and the bit error rate but this is not the case for the average sum capacity. With the average sum capacity, the effect diminishes very gradually as the antenna array at the base station grows. However, overall, the presence of pilot contamination further lowers the network performance as the number of antennas at the base station increases.