Neural Network Applications, The Antenna And Filter

This dissertation focuses on the harmonic suppression of antenna by spur-line, two SRR filters and the neural models of them. The research covers the analysis of the harmonic suppression features of spur-line, spur-line antenna design, the analysis of SRR, two SRR filters design and the neural models design of them. It is proved that the harmonics of antenna can be suppressed by spur-line very well, and the radiation efficiency of antenna is improved. The proposed filters loaded with SRR exhibits compact size and the out put performance of the filters are improved. However the relationship between the dimension of the components and the output characteristic is nonlinear and very complex. Neural model will be a good choice to complete nonlinear model.The electromagnetic interference (EMI) is an important problem in microwave system which is caused by harmonic radiations in most cases. In order to solve this problem, it's quite often to add a harmonic suppression filter between the antenna and the power amplifier which increases the size of the antenna system, moreover, the extra interconnections between the antenna and the power amplifier will disturb the match of the system and increase loss at high frequencies. In order to solve this problem, a harmonic suppression method using spur-line is introduced. Moreover spur-line will not increase the design complexity. At last we degin a neural model of spur-line antenna.SRR is an artificial structure proposed by Pendry. SRR can exhibt left handed transmission line feature. So it is used to design many microwave devices.SRR is a very simple structure and can be used in many planar circuits. We have designed two SRR filters and the equivalent circuits of them are also given in this dissertation. The analysis of SRR filter is very difficuit and complex. So neural models of the two filters are used and the nonlinear relationship between the dimension of the filter and the output characteristic is expressed by it clearly.This dissertation only discusses some new phenomena and questions on the basis of an applied fundamental research. Thus there are still many problems existing and need to be explored in future works.