A tracking radar consists of an RF/IF signal processing part, a display part(PPI), a range tracking part, a motor control part, and a fire control part. The range tracking part processes data from the display part and signal from the RF/IF signal processor part using a constant false alarm rate(CFAR) algorithm. The motor control part plays a role in angle tracking.
In this thesis work, we propose a range tracking algorithm and implement the range tracking algorithm using a digital signal processor(DSP). Detection is the process of deciding whether a target is present or not. It is accomplished by comparing the signal-plus-noise level to a threshold. The threshold is adaptively chosen according to the tracking state. In our proposed algorithm, the first threshold is decided by a modified trimmed mean(TM)-CFAR algorithm. The modified TM-CFAR adjusts the width of a reference window depending on the tracking state. we can observe the tracking states through a stored state queue. When the signal-plus-noise level is larger than the second threshold, a target detection information is stored in the state queue. The second threshold is decided by the prior threshold and the prior signal power. A narrow reference window is preferable in an interference target environment. On the other hand, in a wide reference window, we can detect a missing target again. The proposed range tracking algorithm is implemented on an ADSP-21020 DSP chip.
