Research And Implementation Of Channel Estimation Algorithms For Maritime Wireless Communication Systems

With the economic development and the smooth implementation of the“Belt and Road,”maritime trade,scientific research,and military activities have become increasingly frequent and urgently require reliable and efficient maritime communication systems as guarantees.However,the reliability of maritime wireless transmission is easily affected by the marine environment.The data transmission rate of most current systems is low.How to ensure large-capacity and reliable communication between ship and land,ship and ship,ship and unmanned equipment in the time-varying channel environment at sea is one of the urgent problems to be solved in the development of maritime wireless communication.On the one hand,the unmanned equipment industry is developing rapidly.The maximum speed of the unmanned surface vehicle is 92.6km/h.The maximum flying speed of the unmanned aerial vehicle can reach 289km/h.The rapid relative movement between the receiving and transmitting antennas will produce a Doppler frequency shift,and the channel exhibits time-selective fading;On the other hand,the wireless signal produces multipath effect through sea surface reflection and diffraction,causing frequency selective fading,which seriously affects the transmission performance of the maritime wireless system.The dissertation aims at the maritime wireless communication system.Based on the time-varying channel environment at sea,this paper studies the channel estimation algorithm,considers the performance and complexity of the algorithm,selects the appropriate algorithm to design and implement the hardware structure,and applies it to practical engineering.The specific results of this paper are as follows:1.Based on the time-varying channel at sea,the thesis compares the Basis Expansion Model?BEM?channel estimation algorithm with the common pilot interpolation channel estimation algorithm.The simulation results show that for comb-pilot-based OFDM systems,The BEM channel estimation algorithm still has a performance advantage.However,the traditional BEM algorithm contains many high-dimensional matrix operations,so it has high computational complexity and is difficult to be applied to reality.Therefore,based on the BEM algorithm,a low-complexity FFT-BEM algorithm with equivalent performance is proposed.The algorithm uses fast Fourier transform instead of matrix operation to reduce the computational complexity of the traditional BEM from o?N³?to o?Nlog₂N?without reducing performance.2.Taking into account the design of the self-organizing network communication system supports multiple modes,so that users can be flexibly configured according to the actual application scenario,which requires the channel estimation module to support different bandwidth,different pilot pattern and other parameters change,making higher requirements for hardware implementation.In view of the above requirements,this paper designs a common hardware structure and interface of the pilot interpolation channel estimation module,reduces the degree of coupling between modules,and greatly enhances the portability of design modules.The paper uses Least Squares?LS?and linear interpolating channel estimation algorithms as an example,uses VHDL language to complete its FPGA implementation,and is verified in two independent OFDM systems.3.The hardware structure of the proposed FFT-BEM algorithm is designed.After two structural optimizations,the computational complexity of the FFT-BEM algorithm is further reduced.At the same time,the hardware structure is simplified,so that the FFT-BEM algorithm only needs one matrix and vector multiplication,and one fast Fourier transform to get the frequency domain channel estimate.The optimized structure is implemented in FPGA.After Vivado integrated wiring,the resource consumption of the lookup table,registers,etc.does not exceed 7%of the total resources.