An new strategy for online evaluation of analog circuit performance based adaptive least squares support vector regression with double kernel RBF tuning

A novel strategy for online evaluation of analog circuit performance based on adaptive least squares support vector regression machine is proposed. Regarding reducing the computation, simultaneously, employing double kernel RBF to interfuse more flexibility to the kernel function online such as the bandwidths. And the design idea and constructed steps based on adaptive least square support vector regression with double kernel RBF tuning are introduced. Experiment adopted the typical circuit Sallen-Key low pass filter to prove the proposed evaluation strategy via the performance eight indexes. Simulation results show that the evaluation performance and the testing speed, especially the testing speed of the proposed is superior to that of the traditional LSSVR and ε-SVR, which is suit for applying online.