非准确目标先验知识MIMO雷达波形优化研究

发布时间：2018-05-19 18:45

本文选题：MIMO雷达 + 目标检测　；参考：《西安电子科技大学》2015年博士论文

【摘要】：近年来,多输入多输出(multiple-input multiple-output, MIMO)雷达由于其良好的性能受到了学者的广泛关注。MIMO雷达具有提高目标检测性能、提高角度估计准确率及能检测速率更慢的动目标等诸多优势,在军事和民事应用中具有广阔的前景。MIMO雷达通常基于目标和环境的先验知识设计波形,而由估计得到的先验知识不可避免存在误差,基于此先验知识得到的优化波形的检测和参数估计等性能严重下降,本文对非准确目标先验知识MIMO雷达波形优化进行了系统的研究,所取得的主要研究成果为：1.目前MIMO雷达中的波形优化通常基于某一种而不是整体性能,针对该问题,本文提出一种提高MIMO雷达检测和参数估计性能的波形优化方法。该方法综合考虑提高检测概率、降低参数估计方差及抑制旁瓣三种性能约束优化MIMO雷达的发射波形相关矩阵(waveform covariance matrix, WCM)。首先,本文推导了检测概率及克拉美-罗界(Cramer-Rao bound, CRB)的等价表达式,然后,联合最大化主旁瓣差约束,并对各约束条件分别加权,进而可通过灵活调整加权系数以解决实际应用中不同需求的波形优化问题。该波形优化问题可描述为线性规划问题,因而可进行高效求解。仿真实验中,详细分析了三种约束条件对雷达性能的影响,实验结果验证了所提方法是有效的。2. MIMO雷达通常基于目标和环境的先验知识设计波形,而由估计得到的先验知识不可避免存在误差,基于此先验知识得到的优化波形的检测和参数估计等性能下降严重。针对此问题,本文提出两种稳健波形设计方法,分别提高MIMO雷达的检测性能和参数估计性能。在目标位置误差和通道误差有界的条件下,分别构造以信噪比(Signal Noise Ratio, SNR)口CRB为代价函数的优化问题。为最大化SNR和最小化CRB,分别给出迭代算法,交替以发射波形相关矩阵和通道矩阵误差为优化变量求解,将迭代的每一步转化为凸优化问题,从而在最差情况下提高MIMO雷达系统的检测性能和参数估计性能。仿真实验结果验证了所提方法能有效改进MIMO雷达的检测和参数估计性能。3.传统MIMO雷达发射波形设计方法对传播矩阵误差敏感,最优匹配波形难以得到,进而造成系统的检测性能严重下降。针对该问题,在概率约束条件下,本文提出一种稳健的MIMO雷达发射波形设计方法。该方法考虑最差情况的发生为小概率事件,基于输出信噪比低于可接受水平的概率小于中断概率的约束条件,通过最大化输出信噪比设计最优波形。利用传播矩阵误差的概率分布特性,将概率约束转化为凸约束,从而将统计优化问题转化为确定性优化问题。该方法在传播矩阵存在误差情况下以高概率实现系统性能最优化。仿真结果表明所提方法能够提高输出信噪比,具有较好的检测性能。4.均匀圆阵(uniform circular array, UCA)与MIMO雷达的结合可在不牺牲UCA雷达主要优点的同时兼有MIMO雷达的优势。然而,UCA-MIMO雷达的波束形成与距离无关,该特点限制了其抑制特定距离干扰的性能。本文提出频率分集(frequency diversity, FD)UCA-MIMO雷达距离依赖波束形成方法。该方法中UCA的每个阵元发射具有微小频率步进的不同频率,从而远场的信号积累与距离有关,进而波束形成依赖于距离。仿真实验表明了该方法的有效性。
[Abstract]:In recent years, multiple-input multiple-output (MIMO) radar has attracted wide attention from scholars because of its good performance..MIMO radar has many advantages, such as improving target detection performance, improving angle estimation accuracy and moving targets with slower detection rate, and has a broad prospect in military and civil applications,.MIM The O radar usually designs the waveform based on the prior knowledge of the target and the environment, and the estimated prior knowledge inevitably has errors. The performance of the optimized waveform detection and parameter estimation based on this prior knowledge is seriously reduced. This paper systematically studies the waveform optimization of the inaccurate target prior knowledge MIMO radar. The main research results are as follows: 1. at present, waveform optimization in MIMO radar is usually based on one kind but not the whole performance. In this paper, a waveform optimization method to improve the performance of MIMO radar detection and parameter estimation is proposed. This method comprehensively considers the detection probability, reduces the variance of parameter estimation and inhibits the three performance of the sidelobe. Constrained optimization of the emission waveform correlation matrix (waveform covariance matrix, WCM) for MIMO radar. First, this paper derives the equivalent expression of the detection probability and the kratho bound (CRB). Then, the joint maximizes the main side lobe difference constraint and weights the constraints respectively, and then the weighted coefficients can be adjusted flexibly to solve the problem. The waveform optimization problem of different requirements in practical applications. The waveform optimization problem can be described as a linear programming problem and can be efficiently solved. In the simulation experiment, the effects of three constraints on the radar performance are analyzed in detail. The experimental results verify that the proposed method is an effective.2. MIMO radar, which is usually based on the target and the environment. According to the knowledge design waveform, the estimated prior knowledge inevitably has errors. The performance of the optimized waveform detection and parameter estimation based on this prior knowledge is reduced seriously. In this paper, two robust waveform design methods are proposed to improve the detection performance and parameter estimation performance of MIMO Rada respectively. Under the conditions of bounded error and channel error, the optimization problem of the cost function of the Signal Noise Ratio, SNR port CRB is constructed respectively. In order to maximize SNR and minimize CRB, an iterative algorithm is given respectively, which alternately solves the waveform correlation matrix and the channel matrix error as the optimal variable, and transforms each step of the iteration into a convex optimization. In the worst case, the detection performance and parameter estimation performance of the MIMO radar system are improved. The simulation experiment results show that the proposed method can effectively improve the detection and parameter estimation performance of the MIMO radar..3. traditional MIMO radar transmission waveform design method is sensitive to the propagation matrix error, and the optimal matching waveform is difficult to obtain, and then it is built. In this paper, a robust MIMO radar emission waveform design method is proposed under the condition of probability constraints. This method considers the worst case occurs as a small probability event, and is based on the constraint condition that the probability of the output signal to noise ratio is lower than the acceptable level and the maximum loss. The optimal waveform of signal-to-noise ratio is designed. By using the probability distribution characteristic of the propagation matrix error, the probability constraint is converted into a convex constraint, and the statistical optimization problem is transformed into a deterministic optimization problem. The method realizes the optimal performance of the system with high probability in the presence of error of the propagation matrix. The simulation results show that the proposed method can improve the transmission capacity. The combination of.4. uniform circular array (uniform circular array, UCA) with MIMO radar, which has good detection performance, has the advantage of MIMO radar without sacrificing the main advantages of UCA radar. However, the beamforming of the UCA-MIMO radar is independent of the distance. This feature limits its performance to suppress specific distance interference. This paper proposes frequency. Frequency diversity (FD) UCA-MIMO radar range dependent beamforming method. In this method, each element of the UCA has a different frequency with a small frequency step, thus the signal accumulation in the far field is related to the distance, and then the beam formation depends on the distance. The simulation experiment shows the effectiveness of the method.
【学位授予单位】：西安电子科技大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：TN958

【相似文献】