Pilot Allocation and Sum-Rate Analysis in Cell-Free Massive MIMO Systems

This paper deals with the challenging issue of the unaffordable channel training overhead in the dense cell-free massive multi-input multi-output (MIMO) system when a high number of users are being simultaneously served. By adopting the user-centric cluster method, a dynamic pilot reuse (DPR) scheme is proposed to allow a pair of users to share a single pilot sequence. Specifically, the proposed reuse scheme is achieved with the objective of maximizing the uplink achievable sum-rate subject to users' signal to interference plus noise ratio (SINR) requirements and pilot resources constraints. On this basis, the SINR expression is derived for any user sharing its pilot with another by utilizing both minimum mean squared error (MMSE) detection and channel estimation. A low complexity pilot reuse algorithm is then developed based on the separation distance between users. The iterative grid search (IGS) method is employed to find the threshold that can be utilized in the proposed algorithm to maximize the sum-rate. Finally, simulation results are presented to show the effectiveness of the DPR scheme with the optimized threshold in terms of the uplink achievable sum-rate.