Research On OTFS Equalization Technology For Ship-to-air Mobile Communication

As a highly dynamic communication scenario at sea,ship-to-air communication can not only be an extension of maritime satellite communication and shore-based communication but also constitute an important part of the future air-space-sea-land integrated communication network,which has great development potential.However,ship-to-air communication faces serious fading and multipath effects caused by the complex environment,and the large Doppler shift generated by the high-speed movement of its transceiver makes the widely used Orthogonal Frequency Division Multiplexing technique in 5G seriously impaired.The Orthogonal Time Frequency Space(OTFS)technique describes the time-frequency dualselective channel as an approximate time-invariant channel in the Delay-Doppler(DD)domain through a time-frequency two-dimensional extension,which is suitable for highly dynamic scenarios.Therefore,in order to realize reliable communication in ship-to-air channel environment,this thesis investigates the application of OTFS to the ship-to-air communication scenario,with a focus on the equalization technique of the OTFS system on the base of establishment of ship-to-air channel model.The main contents of the thesis are as follows:Firstly,the thesis analyzes the complex environmental factors in ship-to-air communication,based on which the maritime Physical Optics(PO)channel model is studied and the concept of effective diffuse reflection region is introduced.By limiting the diffuse reflection to a certain range,the computational effort of PO model is reduced.And the way to calculate the delay of reflection and diffuse reflection paths is given while the Doppler effect caused by the relative movement of the transceiver is considered as well.As a result,a multipath channel model suitable for the ship-to-air mobile communication scenario is proposed.Secondly,to address the high computational complexity of the traditional linear equalization algorithm,a matrix chunking triangular decomposition equalization algorithm is proposed for the Cyclic Prefix OTFS(CP-OTFS)system based on a thorough study of the OTFS time domain and DD domain channel matrix structure.Based on the traditional Minimum Mean Squared Error(MMSE)algorithm,the proposed algorithm derives the structural properties of the chunked MMSE equalization matrix by using the block diagonal structure of the CP-OTFS time-domain channel matrix and the quasi-band structure of its block matrix,and on this basis,the triangular decomposition and the antecedent recursive back substitution algorithm are used to avoid the matrix inversion,which can greatly reduce the complexity of the MMSE algorithm without sacrificing its performance.Finally,a low-complexity nonlinear equalization algorithm based on alternating minimization is proposed in this thesis to address the problem of insufficient performance of OTFS linear equalization in coping with harsh ship-to-air channel conditions.The proposed algorithm,for the first time,approximates OTFS maximum likelihood equalization to the optimization problem of the sum of convex function in the real number field and iteratively solves it using the idea of alternating minimization.Based on this,the complexity of the proposed algorithm is further reduced by adding preprocessing and optimizing the iterative process by exploiting the sparsity,circularity,and two-norm invariance of the DD domain equivalent channel matrix of CP-OTFS.The superiority of the proposed algorithm in terms of BER performance and the effectiveness of the complexity reduction method are verified through simulations in this thesis.