Coverage and Rate Analysis for Limited Information Cell Association in Stochastic-Layout Cellular Networks

Abstract

The complexity and uncertainty inherent in large cellular networks make the acquisition of location and channel information of all but perhaps a few neighboring network nodes (base stations (BSs)) difficult for a given user. Therefore, a cell association policy must operate with sparse information. Thus, the serving BS is proposed to be the one that provides the highest instantaneous signal-to-interference ratio (SIR) from among all BSs providing average received signal power exceeding a predetermined minimum. This policy is evaluated for the downlink of single-tier (homogeneous) and two-tier (heterogeneous) networks, and for the latter, the key advantage of the proposed policy is its capability to enable traffic offloading. Two methods to determine the minimum average signal power are given. Coverage probabilities and average rates of mobile stations in coverage are derived, accounting for path loss, multipath fading, and random locations of BSs in each tier. Analysis is verified by Monte Carlo simulations. We observe that the instantaneous SIR and average received signal power of a few BSs are sufficient to achieve the coverage corresponding to the highest SIR association, which in general requires instantaneous SIR information of a larger subset of a network. We also observe that in a two-tier network, the effect of strong interference from high-power BSs, such as macro BSs, can be limited by proper choice of minimum average received signal power for low-power BSs.