Compressed Sensing Based Fingerprint Identification for Wireless Transmitters

Table 3

Performance comparison for the envelope extraction and feature extraction.

	Envelope extraction		Feature extraction
	STFT slice	Wavelet	Mathematical statistics	Fitting	Compressed sensing
Advantage	Eliminate the effect of the noise	Envelope is smoother and precise	Simple; the feature is easy to extract	Simple; the features are easy to extract	The features are of high representativeness
Disadvantage	Envelope exhibits a distortion	Parameters are determined by experience	Some features are not representative	Fitting function is determined by experience	Complex; need a long time
Complexity	𝑂𝑛^2	𝑂𝑛^2	𝑂(𝑛)	𝑂(𝑛)	𝑂𝑛^2
Time for calculation (s)	2.251245	3.568356	1.324417	2.419341	4.626052