Iterative Sparse Channel Estimation And Equalization For Underwater Acoustic MIMO-OFDM

Orthogonal frequency division multiplexing(OFDM)has been intensively pursued for underwater acoustic(UWA)communications recently due to its ability to handle multipath channel.Multi-input multi-output(MIMO)techniques can improve the spectrum efficiency drastically.MIMO can be combined with OFDM for high-speed underwater acoustic communications.This paper mainly studies the channel estimation and detection technologies in underwater acoustic MIMO-OFDM communication system.First,the acoustic simulation tool Bellhop is used to model the underwater acoustic MIMO time-varying multipath channel in a statistical sense.Considering the broadband noise model,a capacity calculation method for band-limited underwater MIMO is given.The effects of center frequency,communication distance,transducer spacing and transducer depth on channel capacity are studied.Simulation results show that center frequency,communication distance and transducer spacing have a significant impact on the channel capacity,which provides a basis for the reasonable selection of parameters of the actual underwater acoustic communication system.Next,considering the limited bandwidth resources underwater,and the insufficient use of information among different processing modules in a conventional receiver,this paper studies the iterative reception of underwater acoustic MIMO-OFDM system based on Low-density Parity-check Codes(LDPC).Based on the underwater acoustic OFDM system,the iterative channel estimation algorithm is mainly studied.Five feedback strategies from the channel decoder to the channel estimator are studied,including full hard / soft decision feedback,as well as their threshold-controlled versions,and a decision feedback based on cost function is improved.Simulation results show that iterative channel estimation can achieve significant gain.The gain is more pronounced when the number of pilots is less.Soft decision feedback outperforms hard decision feedback generally.Based on the underwater acoustic MIMO-OFDM system,two kinds of iterative reception structures are studied,respectively,that is,channel estimation is excluded / included in the iteration loop.Simulation results show that the receiver with channel estimation included in the iteration loop can obtain larger iterative gain.Experimental results in sea trials and pools show that the iterative receiver can significantly improve the performance for OFDM and MIMO-OFDM systems.Finally,a progressive iterative receiver is studied to deal with inter-carrier interference(ICI)existing in underwater acoustic MIMO-OFDM communication system,and the orthogonal matching pursuit(OMP)algorithm for channel estimation considering ICI is given.During the iterations,it updates the system model to account for channels with large Doppler spreads,while fully utilizing soft information from the previous iteration for enhanced sparse channel estimation and ICI equalization.Simulation results show that for underwater acoustic OFDM and MIMO-OFDM systems with different degrees of ICI,the progressive iterative receiver can improve system performance significantly,while enjoying lower complexity than non-progressive iterative receiver.Experimental results show that the progressive iterative iterative receiver is effective,and when the ICI is minor,it has very close performance with the ICI-ignorant iterative receiver.