Deep adaptive temporal network (DAT-Net): an effective deep learning model for parameter estimation of radar multipath interference signals

EURASIP Journal on Advances in Signal Processing

Table 2 Comparison of MAE for parameter estimation in radar multipath interference signals across various models

Set	Parameter(s)	Signal Type	XGBoost	LSTM	LSTNet	AdaRNN	DAT-Net
1	BW	LFM	0.1977	0.0275	0.2727	0.0121	0.0039
1	BW	NLFM	0.2043	0.0273	0.0414	0.0196	0.0054
1	BW	BPSK	0.2399	0.1910	0.1759	0.0171	0.0188
1	BW	QPSK	0.2164	0.1833	0.1756	0.0486	0.0202
2	PW	LFM	0.1539	0.0526	0.0685	0.0138	0.0067
2	PW	NLFM	0.1215	0.0428	0.0623	0.0149	0.0043
2	NSP	BPSK	0.2198	0.1776	0.1826	0.0914	0.0197
2	NSP	QPSK	0.2121	0.1768	0.1571	0.0323	0.0163
3	BW, PW	LFM	–	0.1407	0.1945	0.0267	0.0150
3	BW, PW	NLFM	–	0.1602	0.1926	0.0231	0.0118
3	BW, NSP	BPSK	–	0.2667	0.2611	0.0623	0.0429
3	BW, NSP	QPSK	–	0.2548	0.2429	0.0543	0.0369