Fair resource allocation in one-dimensional cellular systems

We consider the problem of fair resource allocation in interference channels, for the case of one-dimensional cells. We propose simple interference channel models based upon the deterministic channel model, assuming zero queuing delay. This enables study of how the interference induces coupling among the rates available in the channels and evaluation of resource allocation. A sequence of models of increasing complexity provides a simple means of dissecting the different effects, and rapid simulation. We show that there is a strong benefit in explicitly pairing channels to minimize interference to users with disadvantaged locations, as compared to an averaging approach such as CDMA. We also show the benefits of power and bandwidth allocation when fair allocations are desired. A distributed algorithm is proposed that embodies the principle of interference channel pairing. We discuss the consequences for cognitive and multi-band radios.