Research On Techniques For Sparse Array MIMO Radar Imaging

The emergence of radar imaging is an important milestone in the history of radar. Imag-ing radar can work all-weather and full time, and can detect target by penetrating obstructions such as vegetation, clothes and the surface, which has broad application prospects. MIMO radar is a new radar system, which introduced from communication in recent years. Starting from the date of birth, it has drawn the wide attention from academia and industry both at home and abroad. It has many advantages in radar imaging, target angle estimation and an-ti-stealthy targets. The use of MIMO radar imaging can overcome the deficiency of the tradi-tional radar imaging technology. MIMO radar imaging does not need a long integration time, and can reduce the requirement of target motion state. MIMO radar can improve resolution at the same time, which providing radar imaging a new train of thought. On this basis, thinning the MIMO radar array, can effectively reduce the amount of equipment, reduce the burden of signal processing, improve the system reliability, and improve the flexibility of the application. These advantages can further highlight the value of the MIMO radar imaging, expand its ap-plication space.Sparse array MIMO radar imaging is studied along two technical routes, which are single snapshot imaging and multiple snapshots imaging. For each technology route, the cases of regular and wide-angle are discussed. Main work includes the following five parts:For the foundation of sparse array MIMO radar imaging problems, the first part summa-rizes the advantages and disadvantages of the conventional method for thinning arrays. With the help of ambiguity function, the resolution of sparse array MIMO radar is studied. The parabolic fitting is used to calculate the mainlobe width of the horizontal and vertical cut of the ambiguity function. Then the mathematical expression for the horizontal and vertical res-olution of distributed MIMO radar is deduced. By the mathematical expression, the factors that affect the resolution of distributed MIMO radar are investigated, and some guidelines on system design and array design are derived.The second part studies the sparse array MIMO radar imaging with single snapshot. There are two kinds of MIMO array that can be used for2D imaging, which are transmitter array and receiver array with parallel and perpendicular allocation, their corresponding virtual array are the line array and planer array. For parallel situation, an approach for thinning MIMO radar arrays using difference set theory is presented. Array design methods are studied for the cases of both computational sensitive and computational insensitive. For the former case, a fast array elements locating algorithm, based on difference set theory, is given. It can yield results that have lower peak sidelobe levels than those in random arrays which are commonly used. For the latter case, a method which combines the difference set theory and genetic algorithm is offered. Compared with conventional algorithm, using this method better performance and faster convergence could be obtained. For perpendicular situation, an ap-proach based on the difference bases and difference sets to construct the minimum redundan-cy arrays is proposed. Through the spatial convolution of the transmit and receive phase cen-ters, MIMO radar with two perpendicular arrays (PA-MIMO) is able to form a virtual planar array with two-dimensional resolution capability. PA-MIMO, designed in accordance with the minimum redundancy criteria, have the advantage of increasing the baseline length and re-ducing the number of elements. In this paper, element placement based on the line combina-tions of difference bases and difference sets is proposed. Analysis shows that the results yield by the approach forming the virtual array is minimum redundancy in the area of design re-quirements. At last, the limit of array redundancy is derived.The third part studies the sparse array MIMO radar wide-angle imaging with single snapshot. Because target can be thought in the near field when wide-angle imaging, the first thing is to deduce MIMO radar boundary conditions between far field and near field. Then aiming at the condition of the target in the near field, the maximum projection pattern is cho-sen to represent the PSF (Point Spread Function), which is used to measure imaging perfor-mance of different sparse array MIMO radar. Then, how to optimize MIMO radar arrays in the near field using Genetic Algorithm (GA) is presented. GA determines which elements are turned on in two periodic arrays (transmitting and receiving arrays) to yield the lowest maxi-mum relative sidelobe level. The fitness function of GA for MIMO radar near field pattern is designed."Pair Crossover Strategy" is taken to make a constant thinned ratio.The fourth part studies on using inverse synthetic aperture radar technology to compen-sate spatial sampling missing, which caused by sparse array of MIMO radar. MIMO radar imaging using ISAR technique combines the space sampling of MIMO radar and the time sampling of ISAR techniques, which can save many antenna elements and improve image resolution. However, there will be the nonequivalent relationship between the space sampling and time sampling, when parameters fail to be matched. By analyzing the effect of nonequiv-alent relationship between space and time in detail, this paper obtains the number and loca-tions of false targets. Analytical formulas to calculate the Peek to False Ratio is deduced, and amplitude attenuation of targets is quantitatively analyzed. On this basis, further discussion gives some characteristics of images for targets in the special locations. Rotational motion estimation is essential for MIMO radar imaging with ISAR technique. According to the esti-mation, the echo data can be rearranged and interpolated, and the cross-range scaling can be implemented for range-Doppler imaging. For an object rotating with a constant acceleration, a method is proposed to jointly estimate the initial rotating velocity and the rotating accelera-tion. It estimates the phase factors of the difference signal by exploiting the phase difference between the echo signals from two different channels of MIMO radar. Based on this, the er-rors induced by trigonometric approximation in the derivation of the method are analyzed, and then the influencing factors causing these errors are obtained. Meanwhile, the motion pa-rameter estimation resolutions are assessed quantitatively.In the fifth part, the system configuration and signal model of MIMO radar using ISAR technique is analyzed. Through discussing the distribution characteristics of wavenumber ob-servations using two-step dimensionality reduction, a new effective polar format algorithm is presented, and some guidelines for designing MIMO imaging radar arrays are obtained. The algorithm transforms the echo data from four-dimensional to two-dimensional through alter-nately reducing dimensionality and sorting. The1-D interpolations in range and azimuth are used during converting polar coordinate to Cartesian coordinate. Analysis of the error pro-vides limits on the imaged scene size.