ISBN: 978-981-11-0008-6 DOI: 10.18178/wcse.2016.06.105
An Improved Correlation Identification Algorithm in Frequency Domain
Abstract— An improved correlation identification algorithm in frequency domain is proposed by us. Using
this method to calculate the impulse of system, we added a small constant to the result of the Fouriertransform
autocorrelation of the input signal to eliminate the bad effects and to increase the identification
accuracy of the system, because there is a zero in divisors. When the sampling rate is greater than the clock
frequency (we called more-times speed), there will be a zero problem, and the zero will be aggrandized.
When the zeros exist in divisors, this will produce a larger error. Formula deduction and simulations shows
that this method can be used to restrain noise and improve the accuracy of identification. Compared with the
traditional identification, the identification gets a lot of improvement. The degree of exactitude by using
proposed method is about 100 times as much as the traditional method.
Index Terms— correlation identification, more times speed, zero problem, improvement of accuracy.
Qiu Lishan, Hao Kaixue, Li Mei
Information Engineering, China University of Geosciences, CHINA
Cite: Qiu Lishan, Hao Kaixue, Li Mei, "An Improved Correlation Identification Algorithm in Frequency Domain," Proceedings of 2016 6th International Workshop on Computer Science and Engineering, pp. 607-612, Tokyo, 17-19 June, 2016.