Multi-cell Resource Allocation Algorithm In Time-varying Underwater Acoustic Channels

With the development of hydroacoustic network research and application,multi-cell networks with buoys as control nodes are widely used in ocean exploration to cover a wide range of data tasks.In a multi-cell network,soft frequency reuse(SFR)technology can certainly increase the system's spectrum efficiency,and it can also greatly alleviate inter-cell interference.However,this technology does not consider the influence of transmission delay and time-varying characteristics in the underwater acoustic channel.In the cell,the adaptive resource allocation algorithm of the Orthogonal Frequency Division Multiple Access(OFDMA)system is used to allocate various resources to the data nodes,and its design depends on the accuracy of the channel state information(CSI).However,the channel time change and transmission delay of the actual underwater acoustic system often cause the delay and inaccuracy of the CSI,which in turn affects the performance of the adaptive system.Based on the existing technology of the underwater acoustic system,this thesis makes an in-depth study on the resource allocation problem in the time-varying multi-cell.Aiming at the problem of interference between edge nodes in the underwater acoustic multi-cell network,this thesis is based on the traditional SFR technology spectrum planning strategy,taking into account the slow propagation speed of the acoustic signal of the underwater acoustic channel,resulting in the presence of multiple signals received by the same node.Due to the characteristics of delay difference,SFR(T-SFR)technology based on delay difference is proposed to better alleviate inter-cell interference.In this scheme,the formulas for the signal to interference and noise ratio(SINR)and sub-band width of the edge node are re-derived.Aiming at the problem of inaccurate CSI caused by time-varying underwater acoustics,this thesis incorporates the finite state Markov chain(FSMC)model into the linear predictor,and establishes a linear finite state Markov chain(LFSMC)predictor to obtain more accurate CSI for adaptive resource allocation.This scheme combines the advantages of the FSMC model with low complexity and reasonable use of statistical CSI and the linear predictor which adapts to fast-changing channels and has excellent prediction accuracy.In the time-varying underwater acoustic multi-cell system,this thesis combines and verifies the proposed T-SFR scheme with the LFSMC prediction model.Through the simulation experiments,it can be concluded that compared with the traditional frequency reuse scheme,the T-SFR scheme proposed in this thesis can ensure the SINR and spectrum efficiency requirements of the cell edge nodes,and effectively alleviate the interference of the cell edge nodes.The LFSMC prediction scheme is superior to the existing least-squares prediction and MC-based feedback prediction scheme in terms of system throughput.When the T-SFR scheme is used to divide the frequency spectrum and the LFSMC predictor is used to obtain more accurate CSI,the time-varying underwater acoustic multi-cell system can obtain better throughput.Finally,the advantages of the proposed algorithm can be summarized.