Detection and Extraction of Weak Pulse Signals in Chaotic Noise with PTAR and DLTAR Models

Algorithm 1

	Inputs: observed signal data with chaotic noise and parameters , etc.
Outputs: the optimal value of the parameters and and the weak pulse signal .
Begin
Standardize raw data and reconstruct phase space
Initialization of parameters such as .
Initialize the distribution, such as ; is discrete uniform distribution; ; is the inverse gamma distribution.
While
Check for the presence of according to Section 2;
Calculate according to the DLTAR model constructed according to (13);
Updating the parameter with the profile least squares algorithm;
Bring the estimated into the pulse signal and calculate ;
End while
Output the optimal parameter value, and calculate the weak pulse signal.
End
Note: (1) The delay time and embedding dimension remain unchanged. Since is a weak signal, the delay time and embedding dimension in phase space reconstruction are basically not affected. (2) When calculating the derivative, a differential algorithm is used because all data is discretized.