Pattern null steering using the cross entropy method by controlling the current phases

To eliminate interference from certain direction, array antenna null steering become an important technology. A Cross-Entropy method based on importance sampling strategies is presented in this paper. This proposed algorithm is also applied to null steering of linear array by controlling only the current phases. In this method, the population samples are sampled from a parameterized probability distribution and are updated during the optimization procedure, so the diversity of the samples can be kept, and the difference between samples set and optimization results is smaller. This method not only effectively overcomes the premature phenomenon but also accelerates calculation speed. Numerical simulation results indicate that nulls can impose at three locations: in the neighborhood of the main beam and sidelobe, and at symmetric places with respect to the main beam. In addition, the region of null steering by using the method is larger than that of other methods. When zero power width of mainlobe is 12° and the maximal side lobe is -12dB, average null steering level is -192.12dB which is three times more than immune algorithm's level. At the same time, the selected samples are smaller, and rate of convergence is two times faster than other ways. It provides more effective method for array antenna optimum design.