Detection And Parameter Estimation Of Complex Modulated LPI Radar Signal

Low probability of intercept(LPI) radar is an important milestone in the history of radardevelopment. The term LPI is a property of a radar that, because of its low peak power, largetime-bandwidth, or other design attributes, makes it difficult to be detected by a reconnaissancereceiver. The use of complex signal modulation also makes the performance of the traditionalsignal parameter estimation algorithms degradation or even failure. For the above, this paperinvestigates the low interception principle of LPI radar, expounds the relationship between thelow interception performance of the radar and target characteristics, radar characteristics, as wellas the sensitivity of reconnaissance receiver, and analyses the low interception performance oftwo complex modulation LPI radar signals from time-frequency domain and ambiguity function.Then the LPI radar signal detection and parameter estimation methods for complex modulationLPI radar signals are researched. The primary work of this paper is as follows.1. The relationship between the low interception performance of the radar and targetcharacteristics, radar characteristics, as well as the sensitivity of reconnaissance receiver areanalyzed, and the relationship between the detection range of the LPI radar and thetime-bandwidth of the signal are deduced. Two kinds of complex modulation LPI radar signal,pseudo-random binary phase code and linear frequency modulation(PRBC-LFM) signal and thehybrid FSK/PSK signal, are researched. PRBC-LFM signal not only has the advantages of boththe PRBC signal and LFM signal, but also has higher speed and range resolution compared withsimply modulation signal; the hybrid FSK/PSK signal has a more uniform ambiguity basecompared with FSK signal. Phase modulation at each frequency eliminates the power sharppoints of FSK signal in the frequency domain. Simulation verifies the low interceptionperformance of the two complex modulation LPI radar signals.2. To overcome the difficulty that peak power detection can’t effectively detect LPI radarsignal, a novel method based on poly-phase filter bank is proposed. In each sub-channel, avariety of detection methods process in parallel, such as peak power detection, long timenon-coherent accumulation detection, frequency domain detection, Radon-Ambiguity transform(RAT) detection and so on. It makes the radar signal be intercepted all probability, and improvesthe detection probability obviously. Simulation verifies the validity of the method in detection ofLPI signals.3. The performance of the traditional parameter estimation algorithms would inevitablycorrupt even failure at low SNR, a novel method based on poly-phase filter bank and high-ordercumulant is proposed. Firstly, the signals are carved up rapidly and uniformly in the frequencydomain by poly-phase filter bank. Then the Gaussian noise can be suppressed by using theshort-time estimate of the third-order cumulant on each narrow frequency band. The estimationof the whole time-frequency matrixes can be obtained through envelope detection. Finally, thenumber of code, code-width, chirp rate, carrier wave frequency, band width, initial frenquencyand ending frequency of compound signal can be estimated by the Time-Frequency images fromRadon transform and extracting the frequency curves. Simulation results demonstrate that the improved algorithm can give high accuracy, low arithmetic complexity and estimate theparameters effectively of PRBC-LFM at low SNR.4. A novel algorithm to estimate the parameters of hybrid FSK/PSK signal at low SNR isproposed. The FSK signal is separated from hybrid FSK/PSK signal by using nonlineartransform to eliminate spread spectrum caused by pseudo code phase modulation. Thetime-frequency image of FSK signal at low SNR is obtained by combining treatment ofpolyphase filter bank and higher order cumulants. The hopping frequencies, hop duration andhop code of hybrid signals can be estimated by extracting the frequency curves. Then PSK signalis separated from the original signal by using the estimated parameters of FSK modulation.Because the accumulation of real and imaginary parts of PSK baseband signal at the phase jumppoints will produce the inflection points, PSK signal code width and the code sequence can beestimated. The simulation results show that the algorithm can estimate the parameters of hybridFSK/PSK signal effectively at low SNR.