Research On Key Technologies Of Anti-jamming Transmission In Single-Carrier Frequency-Domain Equalization Communication System

Single-carrier frequency-domain equalization(SC-FDE)and orthogonal frequency-division multiplexing(OFDM)are two typical transmission technologies widely used in the field of wireless communication.Compared with the OFDM communication system,the SC-FDE communication system has the advantages of insensitive to carrier frequency offset,low peak-to-average power ratio and simple transmitter structure.Therefore,the SC-FDE communication system has been widely used in wireless communication,such as the uplink of 4th and 5th generation cellular communication system and the unmanned aerial vehicle communication link.However,in the increasingly complex electromagnetic environment,wireless communication signals are susceptible to jamming.In order to ensure the high-speed and reliable information transmission of the SC-FDE communication system in the complex electromagnetic spectrum environment,it is particularly important to study more efficient anti-jamming transmission technology.Because the traditional jamming suppression technology will cause the loss and distortion of the communication signal,the jamming suppression performance is limited and even error floor may occur,which fails to meet the performance requirements of reliable transmission of the system.Therefore,this dissertation studies the key technologies of anti-jamming transmission of SC-FDE communication system.First,an anti-jamming transmission scheme of SC-FDE communication system is proposed.Then,a new deep learning-based frequency-domain equalization(DL-FDE)technology is proposed for non-reconfigurable jamming,and the transmission performance is better than the traditional channel estimation and equalization algorithms.For reconfigurable jamming,such as single tone jamming,multi-tone jamming,real sinusoidal jamming,linear frequency modulated(LFM)jamming and sinusoidal frequency modulated(SFM)jamming,five new jamming cancellation algorithms are proposed to make the performance after jamming cancellation close to that without jamming.The main research work and innovations of this dissertation are as follows:1)An anti-jamming transmission scheme for SC-FDE communication system is proposed.Aiming at the non-reconfigurable jamming,the jamming suppression scheme is adopted automatically,and the jamming information is transmitted to DL-FDE module,which reduces the influence of communication signal distortion caused by jamming suppression and improves the transmission performance of the system.Aiming at the reconfigurable jamming,the jamming cancellation scheme is automatically adopted,and the corresponding jamming cancellation algorithm is adopted for different jamming.The proposed scheme can greatly improve the anti-jamming ability of SC-FDE communication system.2)A deep learning-based frequency-domain equalization algorithm is proposed for non-reconfigurable jamming.A sub-network is designed for channel estimation,noise power estimation and channel equalization respectively.The channel estimation and equalization are optimized together,and the jamming information is effectively used to reduce the influence of communication signal distortion caused by jamming suppression.A new channel training mechanism is proposed to improve the generalization ability of DL-FDE algorithm under different wireless channel characteristics.Simulation results show that the bit error ratio(BER)performance of the proposed algorithm is better than that of the traditional channel estimation and equalization algorithm.3)Aiming at reconfigurable single tone and multi-tone jamming,three new jamming cancellation algorithms for the SC-FDE system are proposed,and the theoretical performance analysis and performance simulation are carried out.The simulation analysis verifies the correctness of the theoretical analysis.Aiming at single tone jamming,a two-stage zero-padding algorithm is proposed to eliminate jamming.A new interpolator which supports arbitrary zero-padding length is used in the algorithm.By choosing different zero-padding length in two stages,the computational complexity is reduced and the estimation accuracy is improved.Aiming at multi-tone jamming,a two-stage windowed interpolation based jamming cancellation algorithm is proposed.By selecting different window functions at different stages and combining with a new windowed interpolator,the mutual jamming between different jamming components can be reduced.Aiming at real sinusoidal jamming,a jamming cancellation algorithm based on joint interpolation is proposed.By jointly interpolating the positive and negative frequency components,the algorithm avoids the influence of mutual jamming between the components.Compared with the existing algorithms,the proposed three jamming cancellation algorithms have higher parameter estimation accuracy,lower computational complexity and better jamming cancellation performance.The theoretical analysis and simulation results show that when the data processing length meets certain conditions,the performance of the proposed algorithms is obviously better than the existing algorithms,and the BER performance of the SC-FDE communication system after jamming cancellation almost reaches the BER performance without jamming.4)Aiming at the reconfigurable LFM jamming and SFM jamming,two new jamming cancellation algorithms are proposed,and the theoretical performance analysis and performance simulation are carried out.The simulation analysis verifying the correctness of the theoretical analysis.Aiming at LFM jamming,a jamming cancellation algorithm based on two-stage classification estimation is proposed,and different solutions are proposed for the two cases of continuous and discontinuous jamming frequencies.The estimation of LFM jamming parameter is simplified to single tone signal parameter estimation,which achieves higher parameter estimation accuracy,lower computational complexity and better LFM jamming cancellation performance than the existing algorithms.Aiming at SFM jamming,a jamming cancellation algorithm based on two stages and two branches is proposed.Based on the influence of modulation index on the spectral characteristics of SFM jamming,a double-branch strategy is adopted.By simplifying SFM jamming parameter estimation to real sinusoidal signal parameter estimation,the proposed algorithm achieves higher parameter estimation accuracy,larger effective parameter estimation range,lower computational complexity and better jamming cancellation performance than existing algorithms.Theoretical analysis and simulation results show that when the data processing length meets certain conditions,the two proposed jamming cancellation algorithms are obviously superior to the existing algorithms,and the performance after jamming cancellation is close to that of SC-FDE system without jamming.