Detection Algorithm And System Performance Of Media-based Modulation

Media-based modulation MBM is one of the newest members of the index modulation family proposed in recent years.In MBM scheme,numerous independent channel realizations are produced by controlling the on/ off states of special parasitic components placed near the transmit antenna.These indexes of different channel realizations are used to convey information,so the spectral efficiency of the system can be increased without increasing the transmitting energy or the number of transmit antennas.However,the complexity of the maximum likelihood ML detection algorithm of MBM scheme is high and increases exponentially with the number of parasitic components.Therefore,this paper focuses on the low complexity detection algorithm and system performance of MBM.The research content can be divided into the following three aspectsFirstly,the bit error rate BER performance of single-input multiple-output MBM SIMO-MBM system and multiple-input multiple-output MBM MIMO-MBM system are investaged.Then,taking SIMO-MBM system as an example,three detection algorithms,ML detection algorithm and sphere decoding detection algorithm and minimum mean squared error detection algorithm,are investigated.Simulation results show that the SIMO and MIMO systems with MBM get better BER performances when the spectral efficiency are equal.Analysis of SIMO-MBM system indicates,the BER performance of the sphere decoding detection algorithm and minimum mean squared error detection algorithm is almost the same as that of the ML detection algorithm while the computational complexity of both are improved.The computational complexity of the sphere decoding algorithm is effectively reduced with the increase of the number of receive antennas,and the complexity of minimum mean squared error algorithm can greatly reduce the detection complexity of modulation symbols.Secondly,the generalized spatial modulation MBM GSM-MBM system model is established,and the average bit error probability of GSM-MBM system is analyzed theoretically.To solve the problem that the complexity of the ML detector of GSM-MBM system increases exponentially with the modulation order,a low-complexity detection algorithm termed energy ordered-relaxation iteration method EO-RIM is proposed.First,the possible active transmit antenna combinations and corresponding mirror activation pattern combinations are sorted according to their signal energy.Then a relaxation iterative method is performed to obtain the corresponding modulated signals.According to a predefined threshold,the algorithm strikes a trade-off between complexity and performance.Simulation results show that the BER performance of EO-RIM algorithm approaches that of ML algorithm and is comparable to that of the ordered block minimum mean squared error OB-MMSE detection algorithm.The computational complexity of EO-RIM increases linearly rather than exponentially with the increase of the modulation order and it is reduced by an order of magnitude compared to the OB-MMSE.Thirdly,the uncoded M-ary quadrature amplitude modulation space time labeling diversity USTLD system is introduced,and the average bit error probability of the system is analyzed theoretically.Aiming at the problem that the mapper design method of the USTLD system can not be directly inferred to the phase-shift keying constellation and the expectation to improve the spectral efficiency of the system,a design method of mappers for phase-shift keying constellation and an USTLD system combined with MBM USTLD-MBM are proposed respectively.The theoretical bound of the average bit error probability of the proposed system is derived.The simulation results show that under the same spectral efficiency,the USTLD-MBM system achieves better performance than the original USTLD system.To conduct the high complexity of ML detection of the USTLD-MBM system,a low complexity detection algorithm based on sphere decoding is proposed.The proposed algorithm can reduce the detection complexity by about 50% while the BER performance is almost identical to that of the ML algorithm.