Classification of ground moving targets using bicepstrum-based features extracted from Micro-Doppler radar signatures

EURASIP Journal on Advances in Signal Processing

Table 1 Confusion matrices for the considered techniques with a data length of 512 ms and decision making time of 2 s

The confusion matrix shows the probability of declaring a class w _j if the feature set from class w _i was at the input of the classifier. The three classes examined are listed in the left column and the declared classes are listed in the top row. The values within the main diagonal illustrate the probability of the correct classification when the input class w _i is correctly declared as w _i.