Hidden Markov Model based target detection

Standard tracking filters perform target detection process by comparing the sensor output signal with a predefined threshold. However, selecting the detection threshold is of great importance and a wrongly selected threshold causes two major problems. The first problem occurs when the selected threshold is too low which results in increased false alarm rate. The second problem arises when the selected threshold too high resulting in missed detection. Track-before-detect (TBD) techniques eliminate the need for a detection threshold and provide detecting and tracking targets with lower signal-to-noise ratios than standard methods. Although TBD techniques eliminate the need for detection threshold at sensor´s signal processing stage, they often use tuning thresholds at the output of the filtering stage. This paper presents a Hidden Markov Model (HMM) based target detection method for employing with TBD techniques which does not employ any thresholding.