Optimizing Cluster Size Through Handoff Analysis in User-Centric Cooperative Wireless Networks

User-centric base station (BS) cooperation has been regarded as an effective solution for improving network coverage and throughput in next-generation wireless systems. However, it also introduces more complicated handoff patterns, which may potentially degrade user performance. In this paper, we aim to theoretically quantify the tradeoff between handoff cost and data rate. Two user-centric clustering modes are investigated: number-based cooperation (NBC), which is easier to implement, and distance-based cooperation (DBC), which gives higher data rate performance. In the NBC mode, a user is served by its $K$ closest BSs, while in the DBC mode, it is served by all BSs within a given distance. However, due to the randomness of network topology, it is a challenging task to track handoffs and to characterize data rates. To address this issue, we propose a stochastic geometric analysis framework on user mobility, to derive a theoretical expression for the handoff rate experienced by an active user with arbitrary movement trajectory. Then, we characterize the average downlink user data rate under a common non-coherent joint-transmission scheme, which is used to illustrate the tradeoff between handoff rate and data rate in optimizing the cooperative cluster size. We conclude that in the NBC (resp. DBC) mode, the optimal cluster size is asymptotically inversely (resp. inversely) proportional to the square of the user speed and asymptotically inversely (resp. inversely) proportional to the BS intensity. Finally, computer simulation is conducted to validate the correctness and usefulness of our analysis.