Joint Design of Iterative Training-Based Channel Estimation and Cluster Formation in Cloud-Radio Access Networks

A dilemma in cloud radio access networks (C-RANs) is how to keep a balance between the performance and the efficiency of centralized processing. To solve this problem, the joint design of training-based channel estimation and cluster formation are studied in this paper. To provide efficient cooperation strategies in C-RANs, individual C-RAN clusters are formed by the remote radio heads (RRHs), and a data-assisted channel estimation scheme is studied, which can reduce the redundant cost of training sequences. To ensure the performance of channel estimation and data transmissions, the cluster formation and the channel estimation are optimized jointly. In particular, an iterative training-based channel estimation scheme is designed by using convex optimization and the Broyden-Fletcher-Goldfarb-Shanno algorithm jointly. Moreover, a utility function of cluster formation can be established based on the estimates and the mean squared error of our proposed channel estimation algorithm, and the cluster formation of RRHs can be formulated as a coalitional formation game. Furthermore, a sub-optimal algorithm is also proposed to reduce the computational complexity. Finally, the simulation results are shown to evaluate the performance of our proposed algorithms.