The Research On The Inter-cell Interference Coordination Strategy With Local Channel State Information

In LTE-Advanced system, how to well utilize the system resource and enhance theperformance of the cell-edge users gradually becomes a research focus. Accordingly, theLTE-A system adopts full frequency reuse strategy, which will cause the strong inter-cellinterfenrence and degrade the system performance rapidly. To overcome this challenge,the ICIC（Inter-Cell Interference Coordination） technique is studied in depth tocoordinately allocate the system resources for interference avoidance. However, theexisted centralized ICIC algorithms all consume huge system overhead and delay, whichbecomes a big burden to the system. In this thesis, we will pay more attention to thedistributed ICIC strategy with local CSI（Channel State Information）, which caneffectively improve the performance of the cell edge users and suppress the inter-cellinterference, without sharing the global CSIs among all the base stations.To solve the above problems, this thesis focuses on the research on the inter-cellinterference coordination stategy based on local channel state information, thecontributions are as follows:Firstly, we propose a distributed precoder design algorithm based on virtualsignal-to-interference-plus-noise ratio（VSINR） for multi-cell multi-user multiple-inputmultiple-output（MU-MIMO） system. We seek for the equivalence between thedistributed VSINR maximization problem and the original centralized throughputmaximization problem by designing the proper correction factor of VSINR. Then we canbuild a VSINR model which is proved to be pareto optimal, and we can distributedlyobtain the corresponding pareto optimal precoder only with local CSI. Simulation resultsshow that the proposed precoder can bring a significant increase to the average cellthroughput while effectively saving the system overhead. Secondly, a distributed joint optimization ICIC algorithm based onsignal-to-leakage-plus-noise ratio（SLNR）is proposed. Through the joint optimization ofthe precoding vector and the power allocation factor, we use lagrangian to decouple thejoint optimization problem into two related sub-optimization problems. Then we derivethe closed-form solutions from the problems by matrix analysis and decompositiontechniques. The jointly optimal precoder and power factor can effectively suppress theinter-cell interference while improving the resource utilization rate. The simulationresults show that, the performance of the proposed algorithm is superior to the existingSLNR-based precoding design algorithms.Finally, we propose a distributed ICIC algorithm with adaptive CoMP（CoordinatedMulti-Point） strategy to enhance the throughput performance of the cell-edge users. Foreach active user, we choose the optimal CoMP mode according to the channel situation,then, based on the local CSI, we design the downlink beamforming strategy bymaximizing the SLNR metric. Simulation results show that the proposed algorithm caneffectively suppress the inter-cell interference and well utilize the system resources, alsoreduce the system complexity.