Dynamic joint processing: Achieving high spectral efficiency in uplink 5G cellular networks.

Coordinated Multi-Point (CoMP) processing is a promising method to mitigate the intra-cluster interference and improve the average Signal-to-Interference-plus-Noise Ratio (SINR). Such method, however, cannot mitigate the inter-cluster interference and only changes the boundaries of interference from cell to cluster. On the other hand, since clusters in the traditional CoMP are static and all the spectrum is used in each cluster, we cannot avoid cluster-edge MSs. In this article, the problems of the current CoMP method are addressed and the reasons why clustering in CoMP should be studied jointly with spectrum allocation are discussed. Then, in the context of Cloud Radio Access Network (C-RAN), an innovative uplink solution—called Dynamic Joint Processing (DJP)—is proposed to mitigate both intra- and inter-cluster interference and to address the fluctuation in capacity demand. Two coexisting clustering approaches are presented based on the level of mobility of MSs: 1) in the low-mobility approach, virtual clusters are formed for each subband and their size is dynamically changed based on the position of the MSs; 2) in the high-mobility approach, a different set of subbands are assigned to different neighboring hybrid cells and the boundaries of frequency bands are dynamically optimized so to address the unanticipated change in capacity demand. Monte Carlo simulations confirm the validity of our statements and show the potential of our solution towards next-generation green communications.