面向低功耗广域网的DSSS系统关键技术研究及实现

发布时间：2018-02-25 16:14

本文关键词： LPWAN DSSS LECIM PHY 捕获频偏估计低信噪比　出处：《安徽大学》2017年硕士论文　论文类型：学位论文

【摘要】：低功耗广域网(LPWAN:Low Power Wide Area Network),是专门为实现低功耗、远距离、低速率、大连接的物联网应用而设计。电气和电子工程师协会IEEE(Institute of Electrical and Electronics Engineers)针对这一需求在低速率无线个域网上制定了 EEE802.15.4k标准。该标准主要适用于低功耗关键基础设施监控(LECIM:Low Energy Critical Infrastructure Monitoring)网络的应用。本文基于IEEE 802.15.4k标准的直接序列扩展频谱系统的低功耗关键基础设施监控网络(DSSS LECIM:Direct Sequence Spread Spectrum LECIM),对其物理层(Physical Layer)收发链路各模块的关键技术进行了深入研究,设计并搭建了整个基带收发链路。实际工程应用中由于低成本的限制,一般采用普通的廉价的晶振,晶振的频率漂移高达40PPM(parts per million)以上。晶振的频率漂移将引起接收性能的恶化,这对接收端在低信噪比环境下正确恢复出发送的数据造成了一定的难度,因此对物理层接收机链路中同步技术的研究具有非常重要的学术价值以及实际应用意义。针对上述问题,论文首先针对基带接收机在低信噪比环境下存在随机频偏的长码字捕获技术进行研究。主要以基于快速傅里叶变换(FFT:Fast Fourier Transformation)运算的时频域二维搜索捕获和基于差分处理捕获算法两个方面进行研究,传统的捕获算法都不能满足LECIM网络的应用需求。本论文提出一种基于差分处理的M阶自相关伪码同步改进算法。先将接收信号和伪码相乘以消除伪码信息,然后进行M阶差分自相关消除载波频偏信息同时对噪声有一定的抑制。最后和传统的几种差分捕获算法在相同条件下进行仿真验证。实验结果表明本论文提出改进算法的抗噪性能更加优越,其捕获性能有4-5dB的改善,很大程度上改善了信噪比,提高了基带系统的接收灵敏度,故更适应于低信噪比条件下工作。其次,对实现超低信噪比环境下的高精度频偏估计技术,主要研究了利用FFT运算以及基于时域上的经典Fitz算法。基于经典Fitz算法,提出了对信号先进行预处理再进行相关运算的策略。该项处理在不缩小频偏估计范围的前提下,能通过有偏自相关并迭代的方法抑制噪声,从而实现了在低信噪比环境下的频偏估计问题。该算法可以大大的提高估计性能,并且随着预处理次数增加而提高估计精度。该方法稳定可靠,非常适合应用于IEEE802.15.4k标准下超长码字、低'信噪比、远距离通信、超高接收灵敏的应用需求。最后,基于MATLAB平台搭建整个基带系统,分别对接收端的频偏纠正模块以及信道编译码模块进行相应现场可编程门阵列FPGA(Field Programmable Gate Array)功能的实现。通过分析结果验证了模块实现的正确性。
[Abstract]:Low power WAN: low Power Wide Area Network is designed to achieve low power consumption, long range, low speed. The association of electrical and electronic engineers, IEEE(Institute of Electrical and Electronics engineers, has developed a EEE802.15.4k standard for this demand on low-rate wireless personal area networks. The standard is mainly suitable for low-power critical infrastructure. In this paper, the low power critical infrastructure of direct sequence spread spectrum system based on IEEE 802.15. 4k standard, DSSS LECIM:Direct Sequence Spread Spectrum LECIMIMM, is used to monitor the physical layer of the physical layer. An in-depth study was carried out. The whole base-band transceiver link is designed and built. Due to the limitation of low cost in practical engineering application, the ordinary cheap crystal oscillator is generally used, and the frequency drift of the crystal oscillator is up to 40 PPM parts per million. The frequency drift of the crystal oscillator will cause the deterioration of the receiving performance. This makes it difficult for the receiver to recover the transmitted data correctly in the environment of low SNR. Therefore, the research of synchronization technology in physical layer receiver link has very important academic value and practical application significance. In this paper, the long codeword acquisition technology of baseband receiver with random frequency offset in low signal-to-noise ratio (SNR) environment is studied, mainly based on time-frequency domain two-dimensional search acquisition and difference based on fast Fourier transform (FFT) operation. Two aspects of processing capture algorithm are studied. The traditional acquisition algorithm can not meet the application requirements of LECIM network. In this paper, an improved M-order autocorrelation pseudo-code synchronization algorithm based on differential processing is proposed. The received signal and pseudo-code are multiplied to eliminate the pseudo-code information. Then M-order differential autocorrelation is carried out to eliminate carrier frequency offset information and the noise is suppressed. Finally, simulation results are carried out under the same conditions as some traditional differential acquisition algorithms. The experimental results show that this paper proposes a new method. The anti-noise performance of the feedforward algorithm is more superior. The acquisition performance is improved by 4-5dB, which greatly improves the signal-to-noise ratio (SNR) and the reception sensitivity of the baseband system, so it is more suitable for working under the condition of low SNR. Secondly, the high precision frequency offset estimation technology in the ultra-low SNR environment is realized. This paper mainly studies the use of FFT operation and the classical Fitz algorithm based on time domain. Based on the classical Fitz algorithm, the strategy of preprocessing the signal first and then performing the correlation operation is put forward. This process does not reduce the range of frequency offset estimation. The frequency offset estimation problem in low SNR environment is realized by using biased autocorrelation and iterative method to suppress noise. The proposed algorithm can greatly improve the estimation performance. And the estimation accuracy is improved with the increase of preprocessing times. This method is stable and reliable, which is very suitable for the application of IEEE802.15.4k standard, such as ultra-long codewords, low 'signal-to-noise ratio (SNR), long distance communication, super high receiver sensitivity. Finally, Based on the MATLAB platform, the whole baseband system is built, and the frequency offset correction module and the channel coding and decoding module are implemented respectively. The implementation of the corresponding FPGA(Field Programmable Gate array function is verified by the analysis results.
【学位授予单位】：安徽大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP393.2

【相似文献】