【作者】 赵鹏；

【导师】 余小游；

【作者基本信息】 湖南大学 ， 信息与通信工程， 2014， 硕士

【摘要】 车联网是互联网在车载环境下的延伸,智能交通是车联网的重要内容。智能交通要求路边设施、车载设备具有无线通信和环境感知能力,车辆利用通信信号同时实现无线通信和雷达功能能够减小对通信网络的依耐性,增加车辆行驶的安全性。OFDM信号用于通信和雷达融合系统是其中一项关键技术,本文旨在研究车载通信环境下OFDM信号的雷达特性及不同参数情况下的雷达信号处理算法。具体研究内容包括:(1)详细分析了具有代表性的24G频点和5.9G频点两种OFDM信号,前者具有大时宽和大带宽特性,但信号功率弱,具有足够的分辨率,噪声对雷达测距测速影响较大,后者时宽和带宽不足,具有较大的信号功率,分辨率是该信号用于雷达测距测速时的主要问题。(2)分析了 OFDM信号的多载波和脉冲串对距离和速度分辨率的影响,得到了两种OFDM信号对应的距离和速度分辨率,理论证明多载波能够提高距离分辨率,脉冲串能够提高速度分辨率。建立了 OFDM雷达信号处理的指标,包括分辨率、精度、主瓣-旁瓣比和动态范围。(3)对于24G频点OFDM信号提出了最大似然加窗算法用于降低旁瓣的干扰,数值分析表明在一定的信噪比条件下,该算法以损失分辨率为代价,有效地降低了噪声引起的旁瓣干扰,并且其性能接近最优。(4)对于5.9G频点OFDM信号,为了提高分辨率,同时对目标的距离和速度进行联合估计,本文从空间谱估计中的时空联合估计方法中得到启发,提出了距离-速度联合估计算法,详细推导了其处理步骤,数值分析表明该算法能够同时估计出各个目标的距离和速度值,在不同信噪比情况下对四种算法进行了对比仿真,分析证明距离速度联合估计算法具有最好的抗噪声能力。更多还原

【Abstract】 Mobility internet extends the normal internet to vehicle environment in which intelligent transportation system is the most key concept.The infrastructure and vehicle have the ability of both wire communication and environment sensing.The technology of fusion signal can reduce the dependency of wireless network and increase the safety of vehicle.OFDM signal has a great advantage in application of wireless communication and radar fusion system.This paper analysis the radar characteristics of two kind of OFDM signal used in communication scene,and propose corresponding radar signal processing algorithm for each OFDM signal.First,this paper analysis two kind of signal named 24G and 5.9GHz frequency point OFDM signal.The former signal have a large time and band width but weak power,noise is the principal factor.The latter signal have a small time and band width but high power,range and velocity resolution is the principal factor.Second,.It is proved that multiple carry can improve resolution of range and multiple pulse can improve resolution of velocity in theory and numerical simulation.To measure the performance of each kind of algorithm,some indicators,include resolution、precision、main-side lobe and dynamic range has been established.Third,some algorithm for 24GHz frequency point signal are analyzed,then a method named maximum likelihood windowing is proposed to solve the influence of side lobe.Numerical simulation proves that this method can eliminate the side lobe to a large extent in cost of reducing resolution and better anti-nose property,which is suitable for the signal.Forth,since normal algorithm are faced with the poor resolution in deal with the 5.9GHz frequency point signal,some super resolution algorithm,include Prony method,MUSIC method and ESPRIT method are introduced to the signal,which acquire target’s information by match each range and velocity after estimate range and velocity independently.Then an algorithm named joint estimation of range and velocity is proposed and the process is given up.Numerical simulation proves this method has best anti-noise property in the four algorithms.更多还原