面向弱信号的GPS软件接收机技术研究

发布时间：2018-04-30 11:05

本文选题：高灵敏度 + FFT鉴频　；参考：《武汉大学》2017年硕士论文

【摘要】：随着智能设备的普及,导航应用的需求逐渐被挖掘,基于用户位置的服务也慢慢地成为了人们的生活的一部分。同时,应用场景多变、复杂的特性也对导航设备的性能提出了挑战,GNSS作为最重要的室外导航方式,其性能的提升对导航的发展起着至关重要的作用。城市楼群、峡谷、森林等环境下,信号的透射、折射、反射等都会导致能量衰减,严重影响着GNSS接收机的捕获、跟踪性能,导致接收机定位精度、连续性严重恶化,因此高灵敏度接收机应运而生。捕获作为信号同步的第一步,直接影响着接收机的性能。针对捕获参数配置与优化问题,在分析各因素对弱捕获性能影响后,设计了一种基于蒙特卡洛仿真的捕获性能评估方法;针对弱信号环境下互相关干扰严重的问题,基于理论分析和测试验证对其特性进行了定量分析,为互相关干扰的抑制提供了依据;针对信号能量衰减导致常规比特同步方法不适用的问题,通过仿真对比分析了直方图法和比特能量法的性能,同时分析了不同比特跳变概率下比特能量法的同步效果。跟踪是对信号参数进行精细估计的过程,对弱信号的跟踪水平很大程度上决定了接收机对复杂环境的适应能力。针对常规鉴频器积分时长受比特跳变限制问题,设计了一种适用于弱信号的FFT鉴频器;针对动态与灵敏度互相矛盾的问题,在环路中引入了惯性辅助,提高了动态环境下的跟踪灵敏度;晶振误差是影响跟踪灵敏度的重要因素,为此提出了一种基于锁相环分析晶振误差的方法,并阐明了晶振误差反馈至NCO的必要条件;针对弱信号环境下子帧无法同步、伪距(发射时间)无法提取的问题,本文提出了辅助信息下基于信号匹配的伪距提取的方法。最后,设计实现了一套完整的软件接收机系统。仿真器定量测试表明:在多普勒和码相位的辅助条件下,750ms积分对22dB-Hz信号具有90%的检测率;当定位误差达到100m(或60m)时,静态的跟踪灵敏度为11dB-Hz(或14dB-Hz),低于0.75g的动态跟踪灵敏度为15dB-Hz(或16dB-Hz),惯性辅助的跟踪灵敏度为12dB-Hz(或14dB-Hz)。车载实测表明:在开阔环境、复杂环境下,本文研制的软件接收机均展现了鲁棒性强的捕获、跟踪能力,接收机的整体定位效果与Ublox LEA-6T 的接近。
[Abstract]:With the popularity of intelligent devices, the needs of navigation applications have been gradually mined, and user-location-based services have gradually become a part of people's lives. At the same time, the application scene is changeable, the complex characteristic also proposed the challenge to the navigation equipment performance. GNSS is the most important outdoor navigation way, its performance enhancement plays the vital role to the navigation development. In urban buildings, canyons, forests and other environments, the transmission, refraction and reflection of signals will lead to energy attenuation, which will seriously affect the acquisition and tracking performance of the GNSS receiver, lead to the positioning accuracy of the receiver, and the continuity of the receiver will deteriorate seriously. Therefore, the high sensitivity receiver emerges as the times require. As the first step of signal synchronization, acquisition directly affects the performance of receiver. Aiming at the problem of configuration and optimization of acquisition parameters, after analyzing the influence of various factors on the performance of weak acquisition, a method of performance evaluation based on Monte Carlo simulation is designed, and the problem of serious cross-correlation interference in weak signal environment is discussed. Based on the theoretical analysis and test verification, the characteristics are quantitatively analyzed, which provides a basis for the suppression of cross-correlation interference, and aims at the problem that the conventional bit synchronization method is not applicable due to the attenuation of signal energy. The performance of the histogram method and the bit energy method are compared by simulation. The synchronization effect of the bit energy method with different bit hopping probability is also analyzed. Tracking is the process of fine estimation of signal parameters. The level of weak signal tracking determines the adaptability of the receiver to complex environment to a great extent. Aiming at the problem that the integration time of conventional frequency discriminator is limited by bit jump, a FFT frequency discriminator suitable for weak signals is designed, and an inertial assistant is introduced into the loop to solve the contradiction between dynamic and sensitivity. The tracking sensitivity in dynamic environment is improved, and the crystal oscillator error is an important factor affecting the tracking sensitivity. A method for analyzing the crystal oscillator error based on PLL is proposed, and the necessary condition for the crystal oscillator error to be fed back to NCO is expounded. In order to solve the problem that subframe can not be synchronized and pseudo-range (transmit time) can not be extracted in weak signal environment, a method of pseudo-range extraction based on signal matching under auxiliary information is proposed in this paper. Finally, a complete software receiver system is designed and implemented. The quantitative test of the simulator shows that under the auxiliary condition of Doppler and code phase, the integral of 750ms has 90% detection rate for 22dB-Hz signal, and when the positioning error reaches 100m (or 60m), The static tracking sensitivity is 11dB-Hz (or 14dB-HzN), the dynamic tracking sensitivity less than 0.75g is 15dB-Hzn (or 16dB-HzN), and the inertial assistant tracking sensitivity is 12dB-Hz( or 14dB-HzN). The in-vehicle measurement shows that in the open and complex environment, the software receivers developed in this paper show robust capture and tracking ability, and the overall positioning effect of the receiver is close to that of Ublox LEA-6T.
【学位授予单位】：武汉大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：P228.4

【参考文献】