Study On The Parameter Estimation Algorithms Of MIMO Radar Signals

Benefitting from waveform diversity of emitted signals and space diversity of antennas, multiple-input multiple-output(MIMO) radar systems outperform the conventional single base radars, which poses a more serious threat to detected targets. Meanwhile, the received MIMO radar signal multiple components may overlap in time, frequency, or spatial domain. As a result, the electronic reconnaissance technology would face new challenge. The researches about electronic reconnaissance to MIMO radar signals should focus on not only modulation recognition and parameters estimation algorithms of single channel receiver, but also reconnaissance receiver with array or netted antennas, and the corresponding signal processing algorithms. Firstly, according to the overlapping in time and spatial domains of co-located antennas MIMO radar signals, we discuss the modulation recognition and parameters estimation algorithms for the signals from co-located MIMO radar with uniform linear array(ULA). Then, under the condition of array antennas receiving, detection and parameters estimation algorithms are proposed for the spatial filtered signals from MIMO radar with widely separated antennas. Final, based on netted radar reconnaissance systems, the processing procedure and algorithms of the signals from widely separated antennas MIMO radars are analyzed.The main researches and contributions of this dissertation are:1) Under the single channel receiving condition, modulation recognition algorithms based on ambiguity function(AF) and high order statistics are presented respectively. The former uses Hough transform(HT) to extract characteristic parameters, while the later uses the diversities of different signals’ high order statistics, and classify with the estimated signal to noise ratio(SNR). Both algorithms are able to classify orthogonal frequency division multiplexing monopulse(OFDM-MP), OFDM linear frequency modulation(OFDM-LFM), and orthogonal poly phase code(OPPC) signals, meanwhile distinguish them from single carrier signals of traditional radars.2) For the classified OFDM-MP, OFDM-LFM, and OPPC signals of MIMO radar with co-located antennas, we study parameters estimation algorithms respectively. For OFDM-MP signals, a multicomponent united estimation algorithm is presented, which estimates carrier frequencies accurately, and is almost not affected by the increase of components number. Multicomponent Wigner-Hough transform(MWHT) is proposed to estimate the initial frequency, chirp-rate, and frequency interval of OFDM-LFM signals, and MWHT outperforms the classical WHT in detection and parameters estimation. A coarse estimation method is introduced to reduce the computation cost of MWHT. Based on the deduction of OPPC signals’ cyclic autocorrelation function and cyclic spectrum density function, we propose the carrier frequency and coding rate estimation algorithm.3) Based on array antennas receiver, the spatial separation of signal components from MIMO radars with widely separated antennas is analyzed, and the spatially filtered OFDM signals are modeled to be the combination of multicomponent signals with unknown number and different amplitudes. The signals detection and parameters estimation algorithms are proposed as follow: we improve the detection variable and decision threshold for frequency detection of OFDM-MP signals, a higher probability of detection is acquired; based on coarse estimation of chip-rate, the fractional Fourier transform(Fr FT) of OFDM-LFM signals are calculated, detection and parameters estimation for multiple LFM components are accomplished.4) The signal receiving and processing of netted radar reconnaissance systems for MIMO radars with widely separated antennas are studied, and a cooperation signal processing method is proposed. Firstly, each separated receiver estimates the pulses’ time of arrival(TOA). Then, using the information of time differences of arrival(TDOA), the main reconnaissance receiver accomplishes signal sorting, and assigns signal processing duties to each separated receiver. Finally, separated receivers extract specific MIMO radar signal components through dynamic filters. Meanwhile, modulation recognition and parameters estimation algorithms for single component are carried out, which accomplishes optimal allocation of systems resources and ensures real-time signal processing.