Research on Neural Network Based Dynamic Channel Assignment Systems in Cellular Mobfle Communication Systems

点击翻译按钮获取中文摘要

Because demand for mobile communication services rapidly increases, the efficient use of frequency bands is an important issue. For the efficient use of the frequency bands, mobile communication networks should be highly optimized. Also, it is neciessary to apply intelligent control methods to mobile communication networks. To optimize the mobile networks, we must solve some NP complete problems quiclfly. For this purpose,neural networks and generic algorithms can be used, and it is very important to develop quick dynamic channel assignment systems using neural networks and genetic algorithms. The objective of this project is to develop dynamic channel assignment systems using neural networks and genetic algorithms, and to evaluate performance of the systems. It is considered that we can apply parallel computation systems consisting of some CPUs to dynamic channel assignment systems. Hence, this project develops dynamic channel assignment systems based on neural networks or genetic algorithms with parallel computation. This project has developed such systems for dynamic channel assignment. Through the evaluation process, we confirmed performance improvement by using the developed method. Also, we examined channel assignment methods to adapt to traffic fluctuation caused by vehicles' traffic fluctuation. Also, this project developed an approximate method to theoretically analyze communication traffic characteristics of dynamic channel assignment methods considering mobility. Other optimization problems and fundamental problems to improve mobile communication systems were also considered.