基于稀疏矩阵的GFDM系统低复杂度算法

GFDM System Low-Complexity Algorithm Based on Sparse Matrix

针对广义频分复用(GFDM)系统复杂度偏高,不能灵活应对未来通信系统多种应用场景的问题,提出了基于稀疏矩阵的GFDM低复杂度调制解调算法。首先,分析GFDM的系统模型,得到2种不同的调制结构;然后,分别对2种调制结构进行不同的预处理,使调制矩阵表现出不同的稀疏性;再对稀疏矩阵重新排列,得到不同的块对角矩阵;最后,利用块对角矩阵的特殊性质,可有效降低GFDM系统的计算复杂度。仿真实验结果表明,在大量窄带子载波或少量宽带子载波场景下,2种低复杂度算法不仅能取得较好的复杂度性能,还能突显GFDM灵活时频结构的优势。

In response to the high complexity of generalized frequency division multiplexing(GFDM)systems and their inability to flexibly adapt to various future communication system applications,a low-complexity modulation and demodulation algorithm based on sparse matrices for GFDM isproposed.Firstly,the system model of GFDM is analyzed,resulting in two different modulation structures.Subsequently,different preprocessing techniques are applied to the two modulation structures to demonstrate varying degrees of sparsity in the modulation matrices.The sparse matrices are then rearranged to obtain distinct block diagonal matrices.Ultimately,by exploiting the special properties of block diagonal matrices,it is possible to effectively reduce the computational complexity of CFDM systems.Simulation results evidently demonstrate that under scenarios with a large number of narrowband subcarriers or a small number of wideband subcarriers,both low-complexity algorithms not only achieve good complexity performance but also highlight the advantages of the flexible time-frequency structure of GFDM.