Research On Multiuser OFDM Resource Allocation In Multipath Underwater Acoustic Channel

Nowadays,with the deepening of exploration of marine resources,underwater acoustic communication has become the main medium of underwater communication in the middle and long distance communication.The background noise of the underwater acoustic channel is characterized by large noise,narrow bandwidth and obvious multipath effect,which seriously affects the quality of underwater acoustic communication.The underwater acoustic OFDM technology provide a high flexibility for resource allocation and received extensive attention in the field of underwater acoustic communication in recent years.The key of user resource allocation and scheduling is to obtain accurate channel state information(CSI).In order to reduce the impact of non-ideal feedback CSI,it needs to be predicted.In this paper,the problem of multi-user dynamic resource allocation under non-ideal feedback is studied.(1)By studying the transmission characteristics of complex and variable underwater acoustic channels,the loss model is established and compared with the radio channel to show the particularity of the underwater acoustic channel.Further,the multi-path simulation model of the underwater acoustic channel is established.The key parameters are gived by acoustic channel tool BELLHOP in the simulation.(2)The underwater acoustic OFDM resource allocation system is constructed based on the underwater acoustic channel transmission characteristics.The key technologies in the OFDM system are analyzed in detail.The advantages of applying OFDM multi-carrier modulation technology to underwater acoustic resource allocation system are discussed.(3)Under the different optimization models,several adaptive resource allocation algorithms are studied,including Lagrangian subcarrier scheduling algorithm,Hughes-Hartogs resource allocation algorithm,Chow algorithm,Fischer algorithm.Compare the performance of these algorithms with MATLAB simulation.The bit error performance of three algorithms is given in the constructed underwater acoustic system.(4)Based on the statistical correlation between the feedback time channel response and the transmission time to establish the Markov state transition probability matrix,a feedback prediction scheme based on Markov state space is proposed to determine the average modulation rate on the user subchannel.The bit and power of each user subcarrier are allocated according to the Chow algorithm,the system performance was compared with the least squares algorithm.It can be concluded from the simulation that the Markov prediction scheme based on non-ideal feedback information has better prediction performance than the least squares prediction algorithm,which can effectively improve the throughput of the system,the scheme can bring diversity gain under different user numbers and improve system performance.