Passive Fetal Movement Signal Detection System Based on Intelligent Sensing Technology

Algorithm 2

FM signal sparse detection.

	Input: the preprocessed FM signal Si, , dictionary Dk, k = 1, 2, and sparsity T0.
	Output: identify the class Li, .
	Sparse FM recognition using the OMP algorithm.
	for
	for
	while
	Initialise the number of iterations t = 1. Residual r0 = Si. Index set U0 = Ø. D = Dk.
	Find the index λt corresponding to the maximum value of the product within each atom of the residue rt−1 and the dictionary D, denoted as .
	Update the index set Ut = Ut−1∪λt. Reconstruct the dictionary .
	is obtained using least-squares calculations.
	Update the residuals .
	end while
	Buffered sparse residuals ek = rt.
	end for
	If e1 < e2, identify as FM Li = 1; otherwise, identify as non-FM Li = 0.
	end for