| Overlapped X-Domain Multiplexing(OVXDM)technology is a new non-orthogonal and highly spectrally efficient coding scheme.Compared with conventional communication systems,OVXDM systems reduce the spacing between transmitting symbols by violating the Nyquist criterion,so that adjacent symbols are regularly shifted and overlapped in the X-domain(time or frequency domain).This improves the transmission efficiency of information per unit time or bandwidth,thus achieving the effect of improving the spectral efficiency of the system.Unlike conventional communication systems,the overlapping principle considers the overlapping between symbols not as an interference,but as a beneficial coding constraint relationship.Moreover,the more severe the overlapping between adjacent symbols,the more the corresponding spectral efficiency is improved.However,the problem also arises that the overlapping between symbols will lead to a sharp increase in the computational complexity of the decoding algorithm at the receiver side.When the maximum likelihood algorithm is used for decoding,the computational complexity of the decoding algorithm grows exponentially with increasing K,which limits its practical application.Therefore,in order to promote the practical application of OVXDM technology,it is important to study the decoding algorithm with low computational complexity suitable for OVXDM system.In this paper,the following aspects are mainly studied for OVTDM systems.Firstly,this paper analyzes three communication techniques that enhance spectral efficiency by violating the Nyquist criterion:FTN,SEFDM,and OVXDM.And briefly analyzes the differences and connections between these three similar techniques.The system model of the OVTDM technology studied in this paper is highlighted,and the decoding principles and decoding processes of various decoding algorithms currently proposed for OVTDM systems are briefly outlined.These works lay the foundation for the research of low computational complexity decoding algorithms and provide new research ideas.Secondly,on the basis of the above theoretical analysis,a new low computational complexity decoding algorithm is proposed,Bidirectional Viterbi Algorithm(BVA).The simulation results show that the bidirectional Viterbi decoding algorithm reduces nearly half the decoding time and memory space compared to the optimal decoding algorithm,and there is essentially no loss in decoding performance.BVA reduces the hardware requirements for the receiver-side decoder,which will avoid the occurrence of decoded information erasure and buffer overflow.These advantages are important for the application of OVTDM in high spectral efficiency scenarios.In the end,based on the above study,I propose a novel decoding algorithm with low computational complexity,the k-best Viterbi decoding algorithm.This algorithm reduces the decoding complexity of Viterbi decoding algorithm by reducing the surviving paths.The simulation results show that the proposed k-best Viterbi decoding algorithm has better decoding performance than the Fano decoding algorithm.In addition,the proposed k-best Viterbi decoding algorithm has lower computational complexity than the Viterbi decoding algorithm,and its decoding performance is close to the optimal decoding performance.In particular,under the condition that the number of overlapping multiplexes is relatively large,the k-best Viterbi decoding algorithm can achieve the near-optimal decoding performance even with a relatively small number of surviving paths,which shows the advantage of the algorithm in decoding at high spectral efficiency. |
PDF Full Text Request