Phase Error Calibration And Phase Code Signal Design For Distributed VHF Radar

VHF radar(also known as meter-wave radar)has the prominent advantages of anti-stealth,but some distinct disadvantages such as lower angular resolution,lower localization precision and worse transportability are also subsistent,while the distributed VHF radar system can solve these disadvantages by well.The above radar system consists of some flexible sub-arrays,which not only can detect targets alone,but also are able to reach the effect of large aperture radar through coherent synthesis processing,overcoming the contradictions of the detection precision and the transportability.In order to carry out the coherent synthesis processing for the distributed VHF radar system,the consistency of T/R channels and the applicable transmitting radar wave are necessary.Thus,this dissertation addresses some Basic research for coherent synthesis processing of the distributed VHF radar experimental system(experimental system is called for short)with the research task of the given distributed VHF radar.The research concentrates on the phase-error calibration of T/R channels and the radar phase code signals design,laying the foundation to verify the key technology such as high precision angle and distance measurement.The main content is summarized as follows:1.For the problem of large spacing between distributed sub-arrays of the experimental system,resulting in much difficulty in using an auxiliary signal source to eliminate the phase inconsistency of T/R channels,an effective method is proposed to estimate the phase errors of two sub-arrays in a unified way by connecting the two calibration networks of the two sub-arrays with a coaxial cable.Firstly,the dedicated calibration channel transmits or receives a single-frequency test signal.Secondly,the calibration network couples the test signal into the receiving end for pre-processing in real time.Finally,we estimate the phase errors of each channel with the obtained I/Q(Inphase/Quadrature)data after signal pre-processing.Moreover,a target signal is simulated to test the calibration results.The real data show that the system works as expected,and the phase errors of each channel at each frequency are within ±5°.2.As the multiple input multiple output(MIMO)working mode of the experimental system can fundamentally improve detecting performance by using orthogonal polyphase code radar signal with low correlation zone(LCZ)in some conditions,a novel approach based on prediction optimizing model is proposed to design such orthogonal signals more efficiently and more effectively than the traditional optimization method.An initial polyphase code sets with short code-length is randomly generated as a basis at first,then minimizing the weighed energy of autocorrelation sidelobes and cross-correlation in the LCZ to predict the next code of each signal continuously,until reaching any appointed code-length.Compared with the common optimization methods,the computational complexity is lower due to the recursive calculation of the correlation function,and especially,the obtained signals have lower autocorrelation peak sidelobe and cross-correlation peak,meeting the deduced theoretical bound of the ratio of peak sidelobe to mainlobe.3.As the experimental system belongs to the pulse radar system,a novel design method of radar phase code signals truncated with low sidelobes(TLS sequence)for avoiding range eclipse is proposed to solve the range eclipse problem of pulse radars that is caused by the incompletely received pulse echo signals from far or near.Firstly,a nonlinear code mapping relation is established to minimize the autocorrelation sidelobe peak or energy of the updated sequence.Then,a phase code signal with short code-length is randomly generated as an initial sequence.Finally,the established mapping relation is used to obtain the TLS sequence with any given code-length from the initial sequence,and the cross-correlation peak of the resulting sequence set is estimated.Simulation results and comparisons with m sequence and other well-known sequences show that the average of autocorrelation sidelobe peak of the whole effective sub-signals truncated from a TLS sequence after pulse compression reduces about 5d B and the resulting sequence set has a low cross-correlation peak,which can be used as an orthogonal waveform.