CB-REFIM: A Practical Coordinated Beamforming in Multicell Networks (1407.1395v3)

Abstract: Performance of multicell systems is inevitably limited by interference and available resources. Although intercell interference can be mitigated by Base Station (BS) Coordination, the demand on inter-BS information exchange and computational complexity grows rapidly with the number of cells, subcarriers, and users. On the other hand, some of the existing coordination beamforming methods need computation of pseudo-inverse or generalized eigenvector of a matrix, which are practically difficult to implement in a real system. To handle these issues, we propose a novel linear beamforming across a set of coordinated cells only with limiting backhaul signalling. Resource allocation (i.e. precoding and power control) is formulated as an optimization problem with objective function of signal-to-interference-plus-noise ratios (SINRs) in order to maximize the instantaneous weighted sum-rate subject to power constraints. Although the primal problem is nonconvex and difficult to be optimally solved, an iterative algorithm is presented based on the Karush-Kuhn-Tucker (KKT) condition. To have a practical solution with low computational complexity and signalling overhead, we present CB-REFIM (coordination beamforming-reference based interference management) and show the recently proposed REFIM algorithm can be interpreted as a special case of CB-REFIM. We evaluate CB-REFIM through extensive simulation and observe that the proposed strategies achieve close-to-optimal performance.