Distributed dynamic SINR pricing for multi-cell beamforming with limited backhaul signaling

This paper considers a distributed approach in designing the downlink beamformers in a multiple-input-single-output (MISO) wireless system. The base stations (BSs) aim to minimize transmit power subject to a set of signal-to-interference-plus-noise ratio (SINR) targets of users. To avoid using a central processor (CP), a distributed iteration framework is adopted in which each BS announces a SINR pricing term to limit the interference it generates at non-local users. The pricing term is announced on these users´ SINR constraints by using interior-point method. With a second-order Taylor approximation the pricing term can be locally built only with a few parameters exchanged between BSs. The main contribution of this paper is to propose an original SINR pricing mechanism that achieves a performance quite close to the optimum with limited backhaul signaling. Simulation is presented, which compares the algorithm with the non-cooperative game and the existing interference pricing algorithm. Numerical results show that the proposed SINR pricing algorithm closely approaches the optimal performance with a desirable convergence speed.