Coordinated Beamforming For Multi-cell Interference Mitigation

4G and 4G-beyond wireless communication system adopts long term evolution(LTE) technology and OFDM access. So users, especially ones on the edge of cells, suffer great inter-cell interference(ICI). The main problem in 4G system is to ensure spectrum effectiveness, simplicity, and fairness simutaniously. We need a kind of ICI-migration technology to improve the throughput of system and ensure users’ fairness. Coordinated beamforming, as a kind of Coordination of Multiple Points(CoMP) technology, only requires CSI and signal exchange among base stations. It has become an active topic because of its desirable performance and feasibility.Most of the existing beamforming algorithms are performed in a centralized way, which requires exhausting synchronization and signal exchange among base stations. Besides, very few of the existing researches take CSI error into consideration. So the algorithms probably generate great performance loss or even lead user into outage. So reducing the overhead and improving the robustness have now become two exceedingly active tops in this field. So this paper concerns the beamforming algorithm and multicell coordination mechanism. And the detailed works are listed as follow:(1)Concerning the overhead requirement, this paper proposes the Distributed Dynamic SINR Pricing Algorithm(DDSPA) in perfect CSI model. This algorithm tries to minimize transmit power subject to SINR constraints of users. Each base station announces a SINR pricing term to limit the interference it generates at non-local users. The pricing terms can be locally built only with a few parameters exchanged between base stations. Numerical results show that the SINR pricing algorithm closely approaches the optimal performance with a desirable convergence speed. Since the algorithm is performed in an iterative way, we also analyze convergence probability and signaling upper bound in one iteration.(2)This paper also proposes a game-theoretical algorithm that both improves the robustness and reduces signaling. There is no signal exchange between base stations. By reporting the user’s inter-cell interference up to base station, a local optimization is performed inside the cell. For robustness, we consider a synthesized CSI model including estimated error and delay error. The Average MSE is chosen as the optimization object. As simulation results shows obvious robustness against two kinds of CSI error compared to traditional MMSE schemes. Convergence speed and probability are also analyzed both theoretically and numerically.(3)This paper then proposes a close-formed robust algorithm that minimizes sum of AMSE subject to transmit power constraints. It can further improves throughput with reduced overhead. By adopting the famous KKT conditions when solving optimization problems, the optimum of system can be reached. As a close-formed scheme, this algorithm avoids iteration and further reduces the difficulty when applying in real scenarios. A coordination mechanism with limited signaling is also designed, which can be naturally performed in a distributed way in TDD system. Simulations shows this algorithm can further improves systems performance and suppresses impact brought by CSI errors.