Multi-Innovation Stochastic Gradient Identification Methods

The stochastic gradient (SG) identification algorithm has a poor convergence rate. We extend the SG algorithm from the viewpoint of innovation modification and present multi-innovation stochastic gradient (MISG) identification algorithms. Since the multi-innovation stochastic gradient algorithms use not only the current data but also the past data at each iteration, parameter estimation accuracy can be improved. Further, we study the performance of the SG and MISG algorithms and show that the MISG algorithms have faster convergence rates and better tracking performance than their corresponding SG algorithms by simulation results