一种基于消息传递的时域重叠复用译码算法

An OvTDM Decoding Algorithm based on Message Passing

时域重叠复用(Overlapped Time Division Multiplexing,OvTDM)技术作为一种非正交的传输技术,人为地引入了符号间干扰来实现高传输速率,但这也使得OvTDM系统在进行最大似然序列检测时面临着极高的运算复杂度。为解决这一问题,研究了OvTDM系统的卷积编码方式,利用发送序列和接收信号之间的关系构建了对应的因子图模型,结合消息传递的原理,提出了一种基于消息传递的OvTDM系统译码算法。该算法通过节点之间信息的不断迭代更新来实现信号的检测。最后,对该算法的性能进行了仿真和分析,并与Fano算法进行比较,仿真结果表明该算法的译码性能优于Fano算法并且具有很低的计算复杂度。

As a non-orthogonal transmission technology,OvTDM(Overlapped Time Division Multiplexing)introduces inter-symbol interference artificially to achieve high transmission rates.However,it leads to a high computational complexity of Maximum Likelihood Sequence Detection in OvTDM system.To deal with this problem,after studying the convolutional coding method of the OvTDM system,this paper constructs a factor graph model between transmission sequences and receiving signals.Then,it proposes an OvTDM decoding algorithm based on message passing.The algorithm detects the signals by continuous iterative updating of information between nodes.Finally,this paper simulates and analyzes the performance of the algorithm,and compares them with Fano algorithm.Simulation results indicate that the decoding performance of this algorithm is better than the Fano algorithm and has a very low computational complexity.