| Frequency, 2-D AOA (two-dimensional angles of arrival), polarizations and other parameters are important characteristic of the electromagnetic wave propagating in the space. By measuring these various characteristics of received waveforms, we can draw conclusions about the events that caused them. Thereby, it is the precondition of knowing the situation around us and taking intelligent action in electronic war to estimating Frequency, 2-D AOA, polarizations and other parameters using arrays. And finds extensive applications in electronic countennine such as inducting disturb, active concealment and electromagnetism camouflage and so on. Along with the development of the science and technology, multi-parameter estimation plays more and more important roles in other applications such as communication, navigation, geologic reconnoiter etc.Today, the signal circumstances become more and more deterioration, the signals distribute more and more widely and densely. Then it is necessary to estimate the frequency, 2-I) AOA and polarizations and other parameters of multiple signals in broad-band and in wide space.Using commonly used polarization sensitivity antenna, three dipoles, based on some basic supposes, The dissertation constructs the polarization sensitivity array signal model of joint frequency, 2-D AOA and polarization estimation. Then, two novel joint frequency, 2-I) AOA and polarization estimation methods of incoming multiple independent spatial narrow-band signals using rectangular and L-shape array respectively are proposed. And take the conclusion that the two methods have similar estimating precision and the L-shape array has simpler configuration and less redundancy.Those two methods above are both valid foe only white Gaussian noise environment, so their applications are restricted in narrow limits, because the practical noise circumstances usually color Gaussian. The application of high-order cumulant can resolve this problem. After introduced the application of high-order cumulant on parameter estimation, the dissertation proposes a new method of joint frequency, 2-D AOA and polarization estimation methods of incoming multiple independent spatial narrow-band non-Gaussian signals in arbitrary Gaussian noise environment using forth-order cumulants and ESPRIIT. This method can be used not only in arbitrary Gaussian noise environment but also for arbitrary array geometries. But the method has two disadvantages, one is more computation quantity, the other is less estimating precision than the method based on correlation function.To fit the broad-band (2.- 18G) application of receivers, using the technology of severe undersampling and nonuniformly L-shape array with large aperture, a new method of joint frequency~ 2-I) AOA and polarization unambiguous estimation of incoming multiple independent spatial narrow-band signals is proposed in chapter 7. Analyses the unambiguous estimating algorithm of joint frequency, 2-D AOA and polarizations. Reduced the request of the receiver antenna and hardware by increase the complexity of algorithm. The method is more applied than mentioned before. The results of computer simulation confirm its availability.Compare with other parameter estimating methods, ESPR1T are less computation quantity and higher precision. But to estimate a parameter using ESPR1T, a pair of identical subarrays are needed, therefore it has high request on the consistency of array and its channels. As a result, it is very important to analyze the affect of array error. In chapter 8, the dissertation constructs the array error model of joint frequency, 2-D AOA and polarization estimation, analyzes the parameter estimation performance of the algorithm based on theoretical conditions under array error, and deduces the mean-square error (MSE) of frequency, 2-D AOA and the polarizations. From the computer simulation we can see that the results of theoretical and the simulation are similar.In conclusion, the innovations of this dissertation concentrate in a few aspects as follows.
|
PDF Full Text Request