Font Size: a A A

Cross-coupled Bandpass Filter And Applied Research, Neural Network Optimization

Posted on:2010-07-30Degree:MasterType:Thesis
Country:ChinaCandidate:M T TanFull Text:PDF
GTID:2208360275483382Subject:Electromagnetic field and microwave technology
Abstract/Summary:PDF Full Text Request
With the rapid development of mobile communications systems and microwave communication technology, the frequency spectrum resource becomes more and more crowded, specifications for filters have tended to become much more severe, which are commonly met only by those filters with stringent frequency-selectivity, smaller size, lower insertion-loss in passband. These filters are designed only by using cross-coupled resonators to generate finite transmission zeros to increase sharp cutoff skirts and improve the performance of RF filters. The synthesis and design of cross coupled resonator filters has arouse many people's research interest due to their importance in the modern wireless communication systems.The synthesis theory of cross-coupled resonator filters and neural networks for filter design applications have been systematically studied in this dissertation. At first, tradition theory of filter design and cross coupled filter's basic theory are systematically presented. Using a hybrid optimization arithmetic technique based on that theory, I can get the coupling matrix and loaded impedance of the cross coupled filter, which has been proved by numerical examples. Secondly, the stepped-impedance resonators are investigated, and the resonance formulas of the resonator which is made up of three different impedance lines are deduced. Resonator like this has been used to construct cross-coupled filters. And then, the basic theory of neural networks and its applications for microwave circuits are introduced. In the filter design process, use neural networks to extract relationships between coupling coefficient and position of two resonators as well as relationships between resonator's external quality-factor and the position of tapped line, which has improved efficiency of filter design. Then, given the experimental results that are compatible with the simulation results, which proved the feasibility of this novel method of filter design. Lastly, triangle open loop resonator filter using loaded structures is introduced, the loaded structures can improve the electromagnetic coupling between the resonators, which create a lower insertion loss in the passband.
Keywords/Search Tags:coupling matrix, resonator, filter, neural networks, stepped-impedance
PDF Full Text Request
Related items