Inter-cell Interference Cancellation Techniques In The Network MIMO Systems

In order to effectively use the limited frequency spectrum resource and improvethe spectrum efficiency, the networking of frequency reuse factor one is universalused in the LTE system, which leads to severe co-channel interference and degradesthe cell-edge throughput. To mitigate the inter-cell interference, this dissertationstudies the beamforming technique, the limited feedback technique and theinterference alignment technique in the network MIMO communication system. Themain contributions and results of this dissertation are listed as follows:1. If no channel state information is available at the transmitter, the sum-rateperformances of three classical beamforming transmission strategies are evaluatedbased on the limited feedback technique. The adaptive beamforming selectionstrategy of the eNodeB is investigated in the circumstances of flexibly changing userlocations. Analysis results demonstrate that when the user is noise-limited, theoptimal transmission strategy for the neighbor eNodeB is the eigen-beamformingtransmission, and when the user is interference-limited, the optimal transmissionstrategy for neighbor eNodeB is the signal to leakage and noise ratio beamformingtransmission.2. There exists a certain amount of error between the channel state informationachieved through limited feedback and perfect CSI, and mean loss in rate due tofeedback error is analyzed. A sub-optimal adaptive bits allocation algorithm isproposed by studying the dynamic allocation of the total number of feedback bitsbetween the desired eNodeB and the interference eNodeB, and a closed formexpression of the bit allocation is derived. The proposed algorithm can achieve bettersum rate performance than the equal bit allocation algorithm with the advantage oflow computation complexity.3. A precoding matrix group selection algorithm based on the minimum totalsystem interference leakage is proposed. This algorithm chooses each eNodeB in thecellular system in turn as the first eNodeB achieving the precoding matrix. Eachrotation can achieve a group of precoding matrixes. Then ergodic all of the precoding group in the system, choose a group of optimized precoding matrix according to thecriteria of minimum total system interference leakage, and this group of precodingmatrix derived by middle transformation matrix is the optimal matrix, then furtheroptimize the design of precoding matrix use this optimal matrix based on improvedscheme is further optimized. Simulation results show that the proposed algorithm hasgreat advantage in terms of the system sum-rate performance under perfect CSI andthe statistical CSI.