多用户双向AF MIMO中继系统的联合预编码算法

近年,双向放大转发(amplify-and-forward, AF)多输入多输出(multiple input multiple output, MIMO)中继系统中的预编码设计吸引了越来越多研究学者的兴趣。然而,在实际MIMO中继通信系统中,针对双向传输下的系统,对多用户模型进行的研究较少。由此，针对多用户双向AF MIMO中继系统模型，提出了一种联合预编码算法。采用在所有节点功率限制下,以最小和均方误差(minimun sum mean square error, MSMSE)设计准则去求解联合信源、中继和信宿端多用户的非凸优化问题,将最初的非凸优化问题转化成多个子优化问题,并利用半正定规划设计、平方约束二次规划设计以及二阶锥规划设计求解子优化问题。之后通过交替迭代方法进一步求解每个子凸优化问题局部优化值,达到优化所有节点矩阵变量的目的。所提算法较现有的算法,不仅在优化问题方面考虑更全面、更实用,而且实验仿真结果验证,提出的算法在系统和均方误差(sum mean-square-error, SMSE)性能、和速率以及误码率(bit-error rate, BER)上有均有改善。

Joint precoding algorithm for multi-user two-way AF MIMO relay system

Precoding design for two-way amplify-and-forward (AF) multiple input multiple output (MIMO) relay systems has attracted more and more research interest recently. However, there is little research for multi-user two-way AF MIMO relay systems, which has important significance for the practical communication systems. A novel precoding design for multi-user two-way AF MIMO relay systems is investigated in which the minimum sum mean square error (MSMSE) is taken as the design criterion to solve the non-convex optimization problem, e.g. joint source, relay, and multi-user with power constraints. The initial non-convex optimization problem is transformed into multiple sub-optimization problems and the semi-definite programming (SDP) design, quadratically constrained quadratic programming (QCQP) design and second-order cone programming(SOCP) design are used to solve the sub-optimization problems. Afterwards, the local optimization value of each sub-convex optimization problem is further solved by alternate iterative methods to achieve the optimization of all node matrix variables. The proposed algorithm is more comprehensive and practical than the existing ones in terms of the optimization problems, and the simulation results show that the proposed iterative algorithm can significantly improve the sum mean square error (SMSE) performance, sum-rate and bit-error rate(BER) of the systems.