The Research On Physical Layer Technology Of UAV High Speed Data Linkchannel Estimation,Equalization,Quantization

The amount of communication data exchanged between the unmanned aerial vehicle and the ground station increases rapidly with the unmanned aerial vehicle applied in military,remote sensing mapping and other areas more and more.As a method of highspeed data transmission,the orthogonal frequency division multiplexing has the advantage of anti-multipath interference ability and high spectral efficiency.Therefore,it is suitable for application in the UAV data link.The multipath effect and Doppler spread can cause the problem of inter-symbol interference in the process of communicating with the ground station,due to the complexity of the environment in which the UAVs are located.Thus it is necessary to estimate and balance the channel to compensate the signal caused by the channel.In this paper,the channel estimation algorithm and interpolation algorithm are studied based on the characteristics of the channels of UAVs and the algorithms are implemented on the hardware platform.In addition,the quantitative problem of log likelihood ratio is also studied.The propagation characteristics of the channels of UAVs,the multipath effect and Doppler spread of the small scale fading are analyzed.Then the characteristics of the channels in different states are studied and the model of the UAV channels in discrete time is established based on the fading type,delay power spectrum and Doppler power spectrum.Finally the rationality of the simulation model of the channel is verified by MATLAB simulation.The channel estimation algorithms based on least squares(LS),minimum mean square error(MMSE)and singular value decomposition(SVD)and the channel interpolation algorithms based on the constant interpolation,linear interpolation,quadratic interpolation,DFT interpolation are analyzed.The channel estimation algorithm of LS and the channel interpolation algorithm of linear interpolation are selected for implementation by FPGA,considering the performance of the algorithm and the complexity of hardware implementation.Compared with the simulation results in MATLAB,the final hardware test results show that the selected algorithms have achieved good results.Finally,the LLR of the transmitted bits is obtained by soft output equalization based on the channel estimation.The dynamic range and distribution of the LLR in additive white Gaussian noise channel and multipath fading channel are analyzed,and then the performance of different quantization algorithms are compared.The simulation results show that: the bit error rate of the uniform quantization algorithm is 1dB better than the non-uniform quantization algorithm in the AWGN channel,and the bit error rate of the non-uniform quantization algorithm is 2d B better than the uniform quantization algorithm in the multipath fading channel.Besides,an adaptive quantization method is proposed in this paper since the actual channel condition is unknown.The uniform quantization is used at low signal to noise ratio.However,when the SNR exceeds the threshold,the nonuniform quantization is adopted.