Abstract
An l 0 -norm constraint Lorentzian (L 0 -CL) algorithm is proposed for adaptive sparse system identification to combat impulsive noise. The L 0 -CL algorithm is derived via exerting an l 0 -norm penalty on the coefficients in the cost function, which is equivalent to add a zero-attractor in the iterations. The zero-attractor attracts the coefficients to zero during the iterations. By the way, the L 0 -CL algorithm can achieve lower mean square error (MSE) for estimating the sparse systems. The simulation results presented in this paper demonstrate that the proposed algorithm has superior performance in both convergence rate and steady-state behavior by identifying the sparse systems in the impulsive noise environment.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
