Sparse Channel Estimation Based On Compressed Sensing For Underwater Acoustic Communication Experiment

Underwater acoustic(UWA)communication is an underwater communication technology which uses the sea water as the transmission medium to carry out information transmission.And it is also the only reliable wireless communication mode which can be realized at the present stage.The development of UWA communication has a great significance on the exploration of the ocean world and marine resources.But it is difficult to achieve the high quality and high efficiency of underwater communication in UWA channel which has multipath effect,Doppler effect,time-varying effect,serious signal attenuation,high noise and so on.Orthogonal Frequency Division Multiplexing(OFDM)is a special multi-carrier modulation technology,which has advantages of high spectrum efficiency and strong anti-channel muti-path interference.Considering the ignorance of the UWA channel sparse characteristics in traditional channel estimation methods,the compressed sensing technology was applied to the estimation of the UWA sparse channel.In this dissertation,we pursued research on the two aspects of theoretical simulation and experimental verification.In this dissertation,we first designed the transceiver module of OFDM UWA communication system.To design the pilot module and the PAPR reduction module,we put up several methods,and chose QPSK modulation with high spectrum efficiency and SLM without distortion.And we designed the frequency conversion module based on Hilbert transform which made the system independent of hardware frequency converter and reduced the performance degradation caused by frequency offset.In addition,a space diversity module based on the maximum ratio combining method was designed to reduce the bit error rate of the system.Finally,we verified the design of the data transceiver structure effective performance in the anti-channel multi-path interference,spectrum efficiency,anti-noise etc.On this basis,we focused on the estimation methods of UWA sparse channel based on compressed sensing(CS).The CS-based estimation channel methods can greatly improve the system performance and spectrum utilization by using a small quantity of pilots to accurately reconstruct the signal.First,we constructed a CS-based sparse multipath channel model,and presented the sparse representation of signal,the construction of the dictionary matrix and the realization of the reconstruction algorithm.Finally,we analyzed the performance of both OMP and BP algorithms under different parameters.In the end of this dissertation,a underwater communication experiment of thesignal transmission and data acquisition in Qiandao Lake was carried out.We applied these CS-based channel estimation algorithms to process experimental data.Compared with the traditional channel estimation algorithms,both OMP and BP algorithms under different parameters performed better in anti-channel muti-path interference and Doppler effects etc.Because of the experimental platform with the low-power amplifier and the swing transducer,the performance of the UWA communication system is not very ideal in the long-distance communication,which remains further study.