Research Of Radar Adaptive Waveform Optimization

The radar transmitting waveform influences the amount of information included in the echoes. It has a direct relationship with the radar performance with respect to detection, estimation, imaging, tracking, and interference mitigation. Choosing transmitting waveforms adaptively according to the operating environment is favorable for extracting more information from the echoes in the best way. This enjoys important theoretical significance and practical value in improving the sensing capability, adaptability, and anti-jamming ability for radars. Adaptive optimizing and designing radar transmitting waveforms has become a hot research topic internationally in recent years.This thesis centers on adaptive optimizing and designing radar transmitting waveforms for improving the detection, anti-interference, and imaging performances. The main works are illustrated as follows:(1) To solve the problem of improving the detection performance in additive noise for Single Input Single Output (SISO) radar based on the principle of matched illumination, the method for adaptively designing the transmitting waveform is studied. The vector expression in time-domain of the output signal-to-noise ratio (SNR) of SISO radar according to the matched illumination theory is derived. Considering Orthogonal Frequency Division Multiplexing (OFDM) radar operating in complex Gaussian white noise, an adaptive waveform optimization method is given to improve the output SNR under the constraint of low peak-to-average power ratio (PAPR). This method employs convex optimization to acquire the sub-optimal coefficient vector for modulating the subcarriers of OFDM signal. The detection performance achieved by the optimized waveform approximates that realized by the optimum matched illumination modulation OFDM signal. The PAPR of optimized waveform approaches to0dB. The given method can make a comprehensive improvement in both detection and PAPR performances. Regarding the SISO radar performing in the additive colored noise, an adaptive waveform design method of weighted target signal subspace is provided. This method realizes improvements in the SISO radar output SNR as well as reductions in PAPR and range sidelobes. The optimized waveform can enhance the ability for suppressing the colored noise. Simultaneously, it is favorable to reduce the complexity of SISO radar system, make full use of transmitting power, and cut down the probability of weak targets being covered.(2) To improve the detection performance of Multiple Input Multiple Output (MIMO) radar in clutter combined with complex Gaussian white noise, studies are conducted on adaptive waveform optimization based upon matched illumination. Taking advantages of the frequency diversity, OFDM signal is applied to MIMO radar with each transmitting antenna emitting one subcarrier. Under the instruction of matched illumination principle, a method adopting an alternate iterative algorithm is presented to jointly optimize subcarrier modulation coefficient vector and receiving filter vector. This method has a low computational complexity and fast convergence. The optimized waveform can improve the OFDM-MIMO radar detection performance effectively and has a good robustness against clutter. Since the OFDM signal usually has a high PAPR, not suitable for practical engineering application, a method is further proposed to adaptively design the random phase-coded transmitting waveforms. To optimize the phase-coded signal with regard to receiving filter, the proposed method transforms the optimization problem into a convex problem through semi-definite relaxation, and then recurs to the classical bisection procedure and Gaussian randomization. A sub-optimal solution can be obtained, and can make a further improvement in MIMO radar output signal-to-clutter-plus-noise ratio (SCNR), and improve the detection performance. In addition, the optimized waveform has a constant envelope, which has more adaptability in engineering.(3) To deal with the problem of the narrow-band interference mitigation faced by wide band and ultra wide band radar, studies on the adaptive sparse frequency waveform (SFW) design are carried out. A method to design a phase-coded SFW with low range sidelobes is proposed. Through making the power spectrum density (PSD) of the waveform match the expected PSD, the anti-interference performance is optimized. The range sidelobes are reduced by minimizing the integrated sidelobe level (ISL) of the waveform autocorrelation. Then according to the Pareto theory, a target function jointly optimizing the PSD and ISL performances is constructed. A cyclic iterative algorithm (CIA) based on Fast Fourier Transformation is proposed to solve the optimization problem. CIA is computationally efficient, easy to be realized in engineering, and flexible. Further improvements can be brought into the performances of PSD and range sidelobes, which can be balanced by means of adjusting the value of Pareto weight. As the narrow band interferences also disturb MIMO radar, the two-stage alternate projection method is presented to design a set of orthogonal SFWs. This method decomposes the optimization problem into two subproblems, which are intended for solving the optimized spectrum and waveform synthesis. A set of phase-code orthogonal SFWs can be obtained, in addition, a set of orthogonal SFWs with low PAPR can also be acquired through parameter settings. The proposed method has good applicability, low computational complexity, and fast convergence.(4) Aiming at increasing the range-angle imaging accuracy for MIMO radar, investigations of adaptive waveform design are made. A method combing waveform design and Radar Cross Section (RCS) estimation is proposed. The transmitting waveforms and the corresponding filters are optimized against each range-angle bin based on the criterion of MMSE. The RCS of the scatter in the corresponding range-angle bin is estimated with the optimized waveforms. The above process is conducted in a recursion way. When all the range-angles bins in the target scene are covered, a new round of updating is carried on. The proposed method takes the advantages of waveform design, and can improve range-angle imaging accuracy and robustness against Gaussian white noise for MIMO radar.