{"title":"Evaluation of the ANN Based Nonlinear System Models in the MSE and CRLB Senses","authors":"M.V Rajesh, Archana R, A Unnikrishnan, R Gopikakumari, Jeevamma Jacob","country":null,"institution":"","volume":24,"journal":"International Journal of Electronics and Communication Engineering","pagesStart":2681,"pagesEnd":2686,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/9958","abstract":"The System Identification problem looks for a\r\nsuitably parameterized model, representing a given process. The\r\nparameters of the model are adjusted to optimize a performance\r\nfunction based on error between the given process output and\r\nidentified process output. The linear system identification field is\r\nwell established with many classical approaches whereas most of\r\nthose methods cannot be applied for nonlinear systems. The problem\r\nbecomes tougher if the system is completely unknown with only the\r\noutput time series is available. It has been reported that the\r\ncapability of Artificial Neural Network to approximate all linear and\r\nnonlinear input-output maps makes it predominantly suitable for the\r\nidentification of nonlinear systems, where only the output time series\r\nis available. [1][2][4][5]. The work reported here is an attempt to\r\nimplement few of the well known algorithms in the context of\r\nmodeling of nonlinear systems, and to make a performance\r\ncomparison to establish the relative merits and demerits.","references":null,"publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 24, 2008"}