A delay-independent approach to robust Hinfin filtering is proposed for discrete-time system with multiple delays in the state and nonlinear inequality constraints. The uncertain parameters are supposed to reside in a polytope and a method for designing the robust Hinfin filters for the filtering error system with sector nonlinearities is derived. The designed filters can guarantee robustly asymptotically stable and a prescribed Hinfin noise attenuation level for the filtering error system. A unified framework of filter can be obtained from the solution of convex optimization problem via linear matrix inequality (LMI), which can be solved via efficient interior-point algorithm. Finally, an example is given to illustrate the effectiveness of the proposed methods.