Parameter estimation and residual generation for (CARARMA) algorithm model depend on D-RGELS

a parameter estimation problem for the input linear controlled autoregressive autoregressive moving average (CARARMA) models is derived in the paper. Firstly, the system decompose into two subsystems, first subsystem, the parameter and the information vectors of the system, the other is the parameter and information vectors of the noise. As a result, the computational efficiency of the model is improved. The estimated parameter vector of the systems cannot generate depend on a standard RLS algorithm. The decomposition based on the recursive generalized extended least squares algorithm is used to solve this problem. Finally, the algorithm efficiency is indicated in simulation results.