An Analytical Framework for Heterogeneous Partial Feedback Design in Heterogeneous Multicell OFDMA Networks

Abstract

The inherent heterogeneous structure resulting from user densities and large\nscale channel effects motivates heterogeneous partial feedback design in\nheterogeneous networks. In such emerging networks, a distributed scheduling\npolicy which enjoys multiuser diversity as well as maintains fairness among\nusers is favored for individual user rate enhancement and guarantees. For a\nsystem employing the cumulative distribution function based scheduling, which\nsatisfies the two above mentioned desired features, we develop an analytical\nframework to investigate heterogeneous partial feedback in a general\nOFDMA-based heterogeneous multicell employing the best-M partial feedback\nstrategy. Exact sum rate analysis is first carried out and closed form\nexpressions are obtained by a novel decomposition of the probability density\nfunction of the selected user's signal-to-interference-plus-noise ratio. To\ndraw further insight, we perform asymptotic analysis using extreme value theory\nto examine the effect of partial feedback on the randomness of multiuser\ndiversity, show the asymptotic optimality of best-1 feedback, and derive an\nasymptotic approximation for the sum rate in order to determine the minimum\nrequired partial feedback.\n