自适应技术在卫星导航接收机抗干扰中的应用
发布时间：2018-12-16 16:19
【摘要】：卫星导航接收机所处的干扰环境未知且随时间变化,接收机必须具有一定的环境适应性以消除干扰捕获信号。为使导航接收机对信号具有空间分辨能力,通常将自适应技术与阵列天线技术相结合,通过自适应调整天线阵列的权系数,使导航接收机天线阵主波束指向期望信号,在干扰方向形成强衰减,也称这种技术为自适应波束形成技术。本文将主要研究动态环境中卫星导航接收机天线阵的自适应波束形成算法及FPGA实现,包括以下三个方面:(1)对空域自适应天线阵分别在白噪声和有色噪声环境下的抗干扰能力进行定量分析。其结果是:对于具有M个天线阵的系统,在高斯白噪声环境下的信噪比改善值为10log(M);在已知信号到达角、信号及干扰的统计特性的理想条件下,有色噪声环境中的信噪比改善度与高斯白噪声环境下相当,说明自适应波束形成器能够消除有色干扰。但由于实际系统采用样本协方差近似真实协方差,降低了系统信噪比改善度。(2)对空时功率倒置(PI,Power Inversion)波束形成器的性能进行了系统评估,结果表明,空时功率倒置算法可消除多种类型干扰,但受简单约束影响,干扰与信号临近时,信干噪比(SINR,Signal to Interference and Noise Ratio)衰减严重。研究了闭环自适应权矢量计算方法,引用改进LMS算法——限定稳定裕度归一化最小均方差(SM-NLMS,Set Membership NLMS)算法基本形式,采用PI准则做约束得到限定稳定裕度归一化功率倒置(SM-NPI,Set Membership Normalized Power Inversion)算法。仿真结果表明,SM-NPI能够在保持LMS小运算量特征的基础上,改善收敛速度。因进一步提高波束形成器权矢量收敛速度的需要,研究了波束形成器的开环算法——维纳滤波方法,为解决维纳滤波矩阵运算量大问题,研究了维纳滤波器的降维方法。仿真结果表明,采用多级维纳滤波方法可以在满足滤波器性能要求下降低矩阵运算维数,采用相关相减多级维纳滤波器可以进一步提高运算速度。(3)研究了自适应算法在FPGA中的实现方法,因LMS算法存在滤波、误差计算、权值更新必须按序进行的缺陷,采用延时最小均方差算法(DLMS,Delay Least Mean Square)实现LMS的并行运算,结合SM-NPI权值更新方法,得到可在FPGA中实现的权值计算算法——D-SM-NPI。ModelSim仿真和硬件测试表明,本文设计的抗干扰算法能够在压制性干扰噪声环境中捕获卫星信号。
[Abstract]:The jamming environment of the satellite navigation receiver is unknown and changes with time. The receiver must have a certain environmental adaptability to eliminate the interference acquisition signal. In order to make the navigation receiver have spatial resolution to the signal, the adaptive technique is usually combined with the array antenna technology. By adjusting the weight coefficient of the antenna array, the main beam of the antenna array of the navigation receiver is directed to the desired signal. Strong attenuation in the direction of interference is also known as adaptive beamforming. In this paper, the adaptive beamforming algorithm and FPGA implementation of antenna array of satellite navigation receiver in dynamic environment will be studied. It includes the following three aspects: (1) quantitative analysis of anti-jamming ability of spatial adaptive antenna array under white noise and colored noise respectively. The result is: for the system with M antenna array, the SNR improvement value under Gao Si white noise environment is 10log (M);. Under the ideal condition of known signal arrival angle, signal and interference statistical characteristics, the improvement of SNR in colored noise environment is similar to that in Gao Si white noise environment, which indicates that adaptive beamformer can eliminate colored interference. But because the sample covariance is used to approximate the real covariance, the improvement of SNR is reduced. (2) the performance of space-time power inversion (PI,Power Inversion) beamformer is systematically evaluated, and the results show that, Space-time power inversion algorithm can eliminate many types of interference, but it is affected by simple constraints. When the interference is approaching to the signal, the signal-to-noise ratio (SINR,Signal to Interference and Noise Ratio) attenuation is serious. The closed-loop adaptive weight vector calculation method is studied, and the modified LMS algorithm is used to normalize the minimum mean square error (SM-NLMS,Set Membership NLMS) of the restricted stability margin, which is the basic form of the SM-NLMS,Set Membership NLMS) algorithm. The restricted stability margin normalized power inversion (SM-NPI,Set Membership Normalized Power Inversion) algorithm is obtained by using PI criterion as constraint. The simulation results show that SM-NPI can improve the convergence speed on the basis of preserving the characteristics of LMS small computation. In order to improve the speed of weight vector convergence of beamformer, the open loop algorithm of beamformer, Wiener filter, is studied. In order to solve the problem of large computation of Wiener filter matrix, the dimension reduction method of Wiener filter is studied. The simulation results show that the multistage Wiener filtering method can reduce the computational dimension of the matrix when it meets the performance requirements of the filter. Using correlation subtractive multistage Wiener filter can further improve the operation speed. (3) the realization method of adaptive algorithm in FPGA is studied. Because of the defects of LMS algorithm, such as filtering, error calculation, weight updating must be carried out in order. Using the delay minimum mean square error algorithm (DLMS,Delay Least Mean Square) to realize the parallel operation of LMS and the SM-NPI weight updating method, the D-SM-NPI.ModelSim simulation and hardware test show that the algorithm can be implemented in FPGA. The anti-jamming algorithm designed in this paper can capture satellite signal in the environment of suppression jamming noise.
【学位授予单位】：西安建筑科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN965.5
本文编号：2382684
[Abstract]:The jamming environment of the satellite navigation receiver is unknown and changes with time. The receiver must have a certain environmental adaptability to eliminate the interference acquisition signal. In order to make the navigation receiver have spatial resolution to the signal, the adaptive technique is usually combined with the array antenna technology. By adjusting the weight coefficient of the antenna array, the main beam of the antenna array of the navigation receiver is directed to the desired signal. Strong attenuation in the direction of interference is also known as adaptive beamforming. In this paper, the adaptive beamforming algorithm and FPGA implementation of antenna array of satellite navigation receiver in dynamic environment will be studied. It includes the following three aspects: (1) quantitative analysis of anti-jamming ability of spatial adaptive antenna array under white noise and colored noise respectively. The result is: for the system with M antenna array, the SNR improvement value under Gao Si white noise environment is 10log (M);. Under the ideal condition of known signal arrival angle, signal and interference statistical characteristics, the improvement of SNR in colored noise environment is similar to that in Gao Si white noise environment, which indicates that adaptive beamformer can eliminate colored interference. But because the sample covariance is used to approximate the real covariance, the improvement of SNR is reduced. (2) the performance of space-time power inversion (PI,Power Inversion) beamformer is systematically evaluated, and the results show that, Space-time power inversion algorithm can eliminate many types of interference, but it is affected by simple constraints. When the interference is approaching to the signal, the signal-to-noise ratio (SINR,Signal to Interference and Noise Ratio) attenuation is serious. The closed-loop adaptive weight vector calculation method is studied, and the modified LMS algorithm is used to normalize the minimum mean square error (SM-NLMS,Set Membership NLMS) of the restricted stability margin, which is the basic form of the SM-NLMS,Set Membership NLMS) algorithm. The restricted stability margin normalized power inversion (SM-NPI,Set Membership Normalized Power Inversion) algorithm is obtained by using PI criterion as constraint. The simulation results show that SM-NPI can improve the convergence speed on the basis of preserving the characteristics of LMS small computation. In order to improve the speed of weight vector convergence of beamformer, the open loop algorithm of beamformer, Wiener filter, is studied. In order to solve the problem of large computation of Wiener filter matrix, the dimension reduction method of Wiener filter is studied. The simulation results show that the multistage Wiener filtering method can reduce the computational dimension of the matrix when it meets the performance requirements of the filter. Using correlation subtractive multistage Wiener filter can further improve the operation speed. (3) the realization method of adaptive algorithm in FPGA is studied. Because of the defects of LMS algorithm, such as filtering, error calculation, weight updating must be carried out in order. Using the delay minimum mean square error algorithm (DLMS,Delay Least Mean Square) to realize the parallel operation of LMS and the SM-NPI weight updating method, the D-SM-NPI.ModelSim simulation and hardware test show that the algorithm can be implemented in FPGA. The anti-jamming algorithm designed in this paper can capture satellite signal in the environment of suppression jamming noise.
【学位授予单位】：西安建筑科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN965.5
【参考文献】
相关期刊论文 前1条
1 曾清平,万山虎,亓强;GPS制导系统与雷达对抗问题讨论[J];空军雷达学院学报;2004年02期
相关博士学位论文 前1条
1 张琳;卫星导航系统接收机抗干扰关键技术研究[D];哈尔滨工业大学;2007年
相关硕士学位论文 前1条
1 严岳品;基于FPGA的GPS相关器设计[D];南京理工大学;2009年
，本文编号：2382684
本文链接：https://www.wllwen.com/kejilunwen/wltx/2382684.html
