An analytical framework for multicell cooperation via stochastic geometry and large deviations

Abstract

Multicell cooperation (MCC) is an approach for mitigating intercell interference in dense cellular networks. Existing studies on MCC performance typically rely on either oversimplified Wyner-type models or complex system-level simulations. The promising theoretical results (typically using Wyner models) seem to materialize neither in complex simulations nor in practice. To more accurately investigate the theoretical performance of MCC, this paper models an entire plane of interfering cells as a Poisson random tessellation. The base stations (BSs) are then clustered using a regular lattice, whereby BSs in the same cluster mitigate mutual interference by beamforming with perfect channel state information. Techniques from stochastic geometry and large-deviation theory are applied to analyze the outage probability as a function of the mobile locations, scattering environment, and the average number of cooperating BSs per cluster l. For mobiles near the centers of BS clusters, it is shown that outage probability diminishes as O(e-l[sup ν]¹) with 0 ≤ ν₁ ≤ 1 if scattering is sparse, and as O(l[sup -ν]²) with ν₂ proportional to the signal diversity order if scattering is rich. For randomly located mobiles, regardless of scattering, outage probability is shown to scale as O(l[sup -ν]³) with 0 ≤ ν₃ ≤ 0.5. These results confirm analytically that cluster-edge mobiles are the bottleneck for network coverage and provide a plausible analytic framework for more realistic analysis of other multicell techniques.