Reciprocity-based cognitive transmissions using a MU massive MIMO approach

Cognitive radio (CR) aims at enabling the coexistence between legacy systems called primary users (PU) and secondary users (SU), provided the SU do not (significantly) cause interference on PU links. Massive MIMO systems on the other hand yield significant performance and reliability enhancement. In this paper, we investigate the capacity of both PU and SU systems using multi-user (MU) massive MIMO and time division duplex (TDD). TDD indeed permits to estimate the channel state information (CSI) at the PU/SU transmitters to perform linear precoding, taking advantage of the channel reciprocity. The capacity is derived considering the CSI imperfections, the non-reciprocal radio frequency front-ends as well as antenna coupling effects. These impairments can be compensated using reciprocity calibration techniques we recently developed. Extending these techniques to MU massive MIMO, we evaluate interference cancellation precoders. Simulations reveal performance improvements using the massive MIMO reciprocity-based precoder in a CR approach.