The Research Of Frame Theory For Signa Coding Erasure

In recent years, recurrent neural networks have been extensively broadly researched because of their application in many fields, such as signal processing, pattern recognition, combination optimization, robotic control and so on. The recurrent neural networks can be classified as static neural networks and local field networks. At present, a lot of competing sufficient stability condition for local field networks have been accumulated, but these two type of recurrent neural networks cannot always be equivalently transformed to one another, so that the aforementioned results for the local field type are not effective for the static one and new stability criteria need to be explored.The stability of a new type static neural networks is investigated in this paper, mainly considered the problem of globally exponential stability and dissipativity analysis of static neural networks with time-varying delays, and giving the existing research results to improve and generalize. The main contributions of this paper are as follows:Firstly, the problem of globally exponential stability for a new type static neural networks with time-varying delay. By constructing the proper Lyapunov-Krasovslii function, based on the thoery of Lyapunov stability, take advantage of the delay partitioning technique, free-weighting matrices, integral inequality, matrix inequality and convex approach, to obtain the global exponential stability conditions. The numerical example is given to show the effectiveness of the proposed method.Secondly, the problem of globally exponential stability and dissipativity for static neural networks with time-varying delay and external disturbance. Based on the concept of unilateral control ?Q, S,R? ?? ? dissipative presented by our predecessor, we give the definition of bilateral control strictly ?Q, S,R? ?? ?? ? dissipative, By constructing the proper Lyapunov-Krasovslii function and use inequality analysis technique, deducing the sufficient condition of globally exponential stability for static neural networks without external disturbance and strictly ?Q, S,R? ?? ?? ? dissipative for static neural networks under zero initial condition.