面向天基物联的SCMA低复杂度译码算法

Low Complexity SCMA Decoding Algorithm for Space-based Internet of Things

在天基物联场景中,当卫星星历和位置信息都较为准确时,即地面的各个用户可以同步接入到系统接收端。由此天基物联接入系统主要面临译码算法计算复杂度过高的问题。针对上述问题,基于部分边缘化以及球形译码策略,提出一种基于球形译码策略的部分边缘化消息传递算法。首先主要通过球形译码策略减少算法迭代中的冗余码字星座点,然后利用码字概率排序和设定动态阈值等方法,在每次算法迭代过程中选择出一个码字概率收敛性能最好的用户,并在之后的迭代中停止更新该用户信息,从而在每次迭代过程中降低了参与迭代更新的用户节点数和码字星座点数。仿真结果表明,该算法相比于现有的消息传递算法,计算复杂度得到显著降低的同时算法误码率性能以及收敛性能几乎不受影响。

In a space-based internet of thirgs scenario,when satellite ephemeris and position information are more accurate,each user on the ground can be synchronously connected to the system receiver.As a result,space-based objects are connected to the system,which is mainly faced with the problem that the decoding algorithm is too complex.In order to solve above problems,based on partial marginalization and spherical decoding strategy,a partial marginalization message delivery algorithm based on spherical decoding strategy is proposed.Firstly,the spherical decoding strategy is mainly used to reduce the redundant codeword constellations in the iteration of the algorithm.Then the codeword probability sorting and setting dynamic threshold are used to select a user with the best codeword probability convergence performance in each iteration of the algorithm,and to stop updating the user information in the following iteration.As a result,the number of user nodes and codeword constellation points participating in iterative updates are reduced in each iteration.Simulation results show that,compared with existing message passing algorithms,the computational complexity of the proposed algorithm is significantly reduced,while the BER performance and convergence performance of the algorithm are almost unaffected.