The Design Of Adaptive Waveform Based On Ambiguity Function And Maximum Mutual Information In Cognitive Radar

With the emergence of modern high-tech military equipment as well as the progress of battle tactics, making the modern battlefield environment become increasingly complex. The development of stealth technology and the training of low altitude/super low altitude tactics cause serious influence and threats to modern air defense radar system. Traditional radar transmits fixed waveform irradiating environment and targets at the radar transmitter end and adopts adaptive filtering techniques and information processing algorithms at the radar receiver end. However such passive adaptive intelligent information processing becomes more and more difficult to deal with the challenges of modern warfare, needing for the appearance of a higher intelligence radar system--cognitive radar. Cognitive radar integrates brain science and artificial intelligence into radar system, building the closed loop system of transmitter, environment, receiver and transmitter. Cognitive radar adjusts the emission waveforms adaptively for the expectation of optimal environment matching, using the feedback information from the receiver to transmitter, improving radar performance.Different from traditional radar, cognitive radar adaptively adjusts the emission waveforms according to the environment. This dissertation focuses on target detection and target recognition problem, analyzing the existing problems, putting forward solutions, obtaining some useful conclusions:Firstly, for the effect of intensive Gaussian clutter on the weak target detection problem, this dissertation has an in-depth study on the reason that clutter enter through side lobe into the target unit to reduce the signal-cluster-noise ratio, which reduces the target detection performance. In the case that the clutter power distribution density is known, a waveform adaptive method is designed based on the ambiguity function particle swarm optimization algorithm. The simulation verifies the method can suppress clutter and improve the signal-cluster- noise ratio, improving the performance of target detection.Secondly, at present, the relationship between emission waveforms, target, noise and clutter is still unclear. This dissertation studies the waveform design method based on maximum mutual information under noise background, and derives the optimal waveform based on the maximum mutual information under the dense Gaussian clutter background. The simulation finds that, the energy spectrum of optimal waveforms will lean to area of small spectrum variance and put more energy to area of big target spectrum variance in order to get more target information.Finally, for the problem of multi-target recognition, this dissertation has designed adaptive waveform mechanism based on the maximum mutual information criterion and the maximum signal to clutter ratio standard. The rationality of adaptive waveform mechanism is verified by simulation. By comparing the two kinds criterion of waveform on the performance of multi-target recognition, it is concluded that the adaptive waveform design based on the maximum mutual information owns higher stability and the recognition efficiency when the clutter amplitude value at a higher degree.In summary, the adaptive waveform design based on ambiguity function and particle swarm algorithm improves the performance of detection, solving the traditional target detection problems in waveform design; the waveform design based on mutual information for the multi-target recognition improves target identification performance compared to the maximum SCR criterion.