面向FMCW激光雷达的修正Rife算法设计与硬件实现  

Design and hardware implementation of modified Rife algorithm for FMCW LiDAR

作　　者：叶茂 刘恒泉 赵毅强 孙泽文 胡彬 Ye Mao;Liu Hengquan;Zhao Yiqiang;Sun Zewen;Hu Bin(School of Microelectronics,Tianjin University,Tianjin 300072,China;Tianjin Key Laboratory of Imaging and Sensing Microelectronic Technology,Tianjin 300072,China)

机构地区：[1]天津大学微电子学院,天津300072 [2]天津市成像与感知微电子技术重点实验室,天津300072

出　　处：《红外与激光工程》2022年第12期130-136,共7页Infrared and Laser Engineering

摘　　要：FMCW激光雷达以其高精度、抗干扰能力强、同时测距测速等特点得到了广泛研究。针对FFT固有栅栏效应引入测距、测速误差的问题,通过分析频谱幅值和相角的规律,并结合正弦函数原理提出了一种易于硬件实现的修正Rife算法,有效地降低了传统Rife算法在估计频率接近FFT量化频率点时的误差。通过仿真和FPGA验证,修正Rife算法在信噪比为-10dB时相较于传统Rife算法平均误差降低了69.6%,均方根误差降低了50.7%,而计算量仅增加了两个乘法和加法,与N点FFT计算量相比可忽略不计。最后,通过搭建光学测试平台,模拟激光雷达中频回波信号验证了该算法的有效性。测试结果显示,该算法可在112m范围内实现同时测距测速,测距误差不大于5cm,测速误差不大于0.16km/h,满足实时性要求。FMCW LiDAR is being widely studied for its high accuracy,strong anti-interference capability and simultaneous ranging and speed measurement.The inherent fence effect of FFT will introduce errors in ranging and speed measurement.To solve this problem,firstly,this paper analyzes the law of spectrum amplitude and phase angle,and then combines with the principle of sine function,proposes a modified Rife algorithm that is easy to implement in hardware.When the estimated frequency is close to the FFT quantization frequency point,this method can effectively reduce the error of the Rife algorithm.The simulation and FPGA verification results show that when the SNR is-10dB,the mean error(ME)and root mean square error(RMSE)of the algorithm are reduced by 69.6%and 50.7%,respectively,compared with the traditional Rife algorithm.The calculation amount is only increased by two multiplications and additions,which is negligible compared to N-point FFT computation.Finally,in order to verify the effectiveness of the algorithm,this paper build an optical test platform to simulate the intermediate frequency echo signal of LiDAR.The results show that the algorithm can achieve simultaneous ranging and speed measurement within 112m.The ranging error is no more than 5cm,and the speed measurement error is no more than 0.16 km/h.The algorithm meets the demand of real-time ranging and speed measurement.

关 键 词：FMCW激光雷达 频率估计 修正Rife算法 FPGA 

分 类 号：TN958.98[电子电信—信号与信息处理]