A Study On Detection Methods Of High Velocity Targets

Detecting moving targets and estimating parameters are always one of radar’s important tasks. When radar is far to the target or the target is too small to detect, the signal to noise radio(SNR) is very low, the signal is much smaller than the noise, so it’s difficult to detect. Accumulating the echo and developing the SNR, can makes radar detect longer, which is one of the important method to detect low SNR target. As to low speed objection or still objection, the neighbored pulses can be accumulated directly, because the migration is smaller than one range resolution cell. The target echo migration usually exceeds a range resolution cell, the phase between neighbored pulses lies in00-3600, when the target is moving with a high velocity. This causes Doppler frequency spectrum aliasing, and makes it troublesome to accumulate the pulses.The thesis focuses on the high speed target detection problem, and specifically discusses and analysis the accumulation of echo. The main contributions are as follows:1. The thesis discuses the basic theory of PD Radar. All kinds of launched signals are analyzed and compared. Three kinds of LFM signal’s parameters are introduced and compared, including:maximum likelihood estimator, phase difference algorithm, and wavelet algorithm.2. The reason why it’s difficult to detect the high velocity target compared to lower speed target is introduced, including migration exceeds range resolution cell and large Doppler frequency change, which makes it difficult to accumulating.3. As to low speed or non-migration targets, pulses can be directly accumulated. The method using velocity estimation can also be used. However, this method needs to know the velocity before accumulation. Another method called Keystone transform is proposed, which can compensate the migration and Doppler frequency change without estimating the velocity. The simulation shows that the method above can develop the SNR.