Research On Low-Complexity Integer-Forcing Detection Algorithm Based On Lattice Reduction

Compared with single-antenna systems,multi-antenna systems can effectively improve channel capacity and spectrum utilization,which are widely used in wireless communication.However,at the receiver,the complexity of signal detection increases with the increasing number of antennas.In order to fully take the advantages of multi-antenna technology,it is necessary to find low-complexity and high-performance detection algorithms.Recently,integer forcing(IF)detection algorithm has proposed,which is an important research direction for signal detection because it has the close-to optimal detection performance and it has the lower detection complexity than linear detection algorithms.Although the IF algorithm has lower detection complexity than the maximum likelihood algorithm,its detection complexity is high especially when the number of antennas is larger.Therefore,the aim of the thesis is to further reduce the complexity of the IF algorithm and achieve a better performance-complexity tradeoff.To reduce the complexity of the IF detection algorithm based on the exhaustive search,the IF detection will be based on the lattice reduction(LR)algorithm in this thesis.Then,the group IF(G-IF)detection algorithm is proposed.In this algorithm,the received signals are firstly grouped,and then each group utilizes the IF detection algorithm independently.Additionally,in each group,signals within other groups are treated as effective noise.The G-IF algorithm may lose some detection performance compared with the IF algorithm,but it reduces the complexity of the operation to obtain the optimal integer matrix from the lattice reduction algorithm by reducing the dimension of the channel matrix,so the complexity of the IF detection algorithm is reduced.The procedures of the G-IF detection algorithm will be specifically introduced in this thesis.The complexity of the G-IF algorithm is compared with the IF algorithm,and the detection performance is compared with the ZF algorithm,the G-IF algorithm has a better performance in terms of the BER performance.A full-diversity design criterion for space time block code based on the G-IF detection algorithm is proposed in this thesis,so full diversity can be achieved for the G-IF detection algorithm used at the receiver.The upper bound of the error probability for each layer of the G-IF algorithm is firstly obtained,which is combined with the linear STBC design to obtain the criterion based on the non-vanishing singular value.The results show when the linear STBC design has the non-vanishing singular value,the communication system which uses the STBC can obtain full diversity by G-IF detection.