基于BP神经网络的VCO非线性校正方法  

A VCO nonlinearity correction method based on BP network

作　　者：肖玮 涂亚庆 刘思蔚 XIAO Wei;TU Yaqing;Liu Siwei(Army Logistics University of PLA,Chongqing 401311,China)

机构地区：[1]陆军勤务学院,重庆401331

出　　处：《自动化与仪器仪表》2021年第4期8-11,共4页Automation & Instrumentation

基　　金：国家自然科学基金(No.61871402);国家重大科技专项(No.2018YF2003900);重庆市自然科学基金(No.cstc2019jcyj-msxmX0245);重庆市研究生创新基金(No.CYS19376)。

摘　　要：压控振荡器(VCO,Voltage Controlled Oscillator)非线性校正技术是实现高精度及高分辨率线性调频连续波(Linear Frequency Modulation Continuous Wave,LFMCW)雷达的关键技术之一。针对现有VCO非线性校正方法在校正精度、工程实现、成本等方面等存在的不足,提出了一种基于BP(Back Propagation)神经网络的VCO非线性度校正方法。该方法通过反向训练BP网络得到控制线性调频信号所需要的调频电压,并引入定时校正机制用以跟踪校正因元件老化和环境变化带来的微小非线性。仿真结果表明,与现有方法相比,本方法校正精度较高,稳定性好且易于工程实现。Nonlinearity correction technology for VCO is one of the key techniques to realize high precision and high resolution for Linear Frequency Modulation Continuous Wave(LFMCW)radar.Aiming at addressing the problems of the existing VCO nonlinear correction methods in the aspects of precision,cost and realization,a VCO nonlinear correction method based on BP(Back Propagation)neural network was proposed.In this method,back-training BP(Back Propagation)network is used to get the FM voltage needed to control the LFM signal,and a timing correction mechanism is introduced to track the tiny nonlinearity caused by the aging of the components and the change of the environment.Simulation results show that compared with the existing methods,this method has higher accuracy,better stability and easier implementation.

关 键 词：非线性度校正 VCO BP网络 

分 类 号：TP202[自动化与计算机技术—检测技术与自动化装置] TN958[自动化与计算机技术—控制科学与工程]