On the Fairness of Rate Allocation in Gaussian Multiple Access Channel and Broadcast Channel

Abstract: The capacity region of a channel consists of all achievable rate vectors. Picking a particular point in the capacity region is synonymous with rate allocation. The issue of fairness in rate allocation is addressed in this paper. We review several notions of fairness, including max-min fairness, proportional fairness and Nash bargaining solution. Their efficiencies for general multiuser channels are discussed. We apply these ideas to the Gaussian multiple access channel (MAC) and the Gaussian broadcast channel (BC). We show that in the Gaussian MAC, max-min fairness and proportional fairness coincide. For both Gaussian MAC and BC, we devise efficient algorithms that locate the fair point in the capacity region. Some elementary properties of fair rate allocations are proved.