Research On Underwater Acoustic Channel And Symbol Estimation Algorithm Based On System With Multiplicative Noise

System with multiplicative noise is the promotion of the traditional Kalmanfiltering linear system, which can solve the problem of signal adaptive estimationunder complex communications environment. Marine environment has many randominhomogeneities, such as time-space variant waves in sea surface, uneven randomwater mass in seawater and random topography, acoustic properties in seafloor. Theseinhomogeneities always affect the propagation of acoustic signals. Therefore, we cansay that the marine medium of underwater acoustic channel is a time-space-frequencyvariant channel with high noise, strong reverberation, narrow bandwidth and seriousmultipath effects interference. Its main characteristics are propagation loss, multipatheffects and dispersion effects.In this thesis, the system model with multiplicative noise is built based on thecharacteristics of underwater acoustic channel combining with the mechanism ofmultiplicative noise. For single input multiple output (SIMO) and multiple inputmultiple output (MIMO) underwater acoustic communications system, the main studyis introduced as follows:1、 For SIMO underwater acoustic communications system, the complicatedmulti-channel system model with multiplicative noise is established based on thedetailed analysis of the channel characteristics, which can depict more complexchannel characteristics. Using the system model with multiplicative noise to describethe underwater acoustic channel has a certain rationality as it has clear fadingcharacteristics and random time-variant factors. And on the basis of the model, thestate optimal filtering algorithm and fixed domain smoothing algorithm are givenbased on the orthogonal projection theorem and the linear minimum variance criterion.2、Based on the SIMO system, the system model with multiplicative noise isextended to MIMO underwater acoustic communications system, and the state-spacemodel of the system with multiplicative noise is established. And the optimal filteringalgorithm is given under the linear minimum variance criterion for MIMO channel andsymbol estimation.3、Computer simulations have been done to validate the optimal filtering and smoothing algorithms of SIMO and MIMO underwater acoustic communicationssystem. Moreover, we use the received data from the widely recognized multipathRayleigh fading channel to compare the optimal filtering algorithm of SIMO systemwith the equalization algorithm from the relative literature to verify the effectiveness ofthe proposed algorithm.