表4.HEX2的信道分配.最小干扰值为17
表3和4列出了由基因算法产生的实际的的两个问题的信道分配方案.KUNZ1,HEX2的结论中:结果”0”代表无干扰分配。我们可以看出对于HEX2和KUNZ1我们获得了比其带爬坡的Hopfield神经网络算法(the hill-climbing Hopfield network (HCHN) )[8]中更好的数据. 在仿真过程中,一些参数,例如交叉操作机率,变异操作机率和族群大小都需要去设定.我们是通过反复试验来设定这些参数的.到目前为止,许多研究者已经研究了在保证无干扰情况下最小化所需信道数的问题。而本论文则是针对那些实际可用信道数少于无干扰所需信道数的实际问题,研究在有限的信道的条件下来最小化生成干扰的的可行性方案,这将会很有实际应用价值.基因算法是一个有趣的方法,它是从点到点的全局搜索,在解决优化组和问题时,可快速获取更优的解。基准问题的仿真结果表明基因算法可得到比其它方法更理想的结果,即在满足需求限制的条件下,使得信道分配带来更少的干扰的解决方案. 更高级的基因算法诸如并行基因算法(parallel GA)和微基因算法(micro GA)可以在短时间内解决信道分配问题2,得到更好的结果. 基因算法(GA)特别适合于在高速并行计算机上运算.目标函数和限制条件可同时执行,对整个族群操作运算,通过交叉和变异操作生成选取新一代适应度更高的子族群参数。 因此对硬件性能要求高,直接关系到运行时间长短,效率问题.在一台高速并行机上, 基因算法预计能以几K倍的速度处理很多问题,K是入口尺寸大小。即使要并行的评估的个别问题功能有效性,也可在最短时间内获得最佳解决办法。REFERENCES参考文献 1 K. Smith, “Solving combinatorial optimization problems using neural networks,” Ph.D. dimerfation, University of Melboume, Australi4 1996. 2 D. Kunz, ‘‘Suboptid solutibni obtained by the Hopfield-Tank neural network algorithm”, Biologicnl Cybernetics, vol.65, pp. l29-133,1991. 3 F. BOX,~‘‘A heuristic technique for issigning frequencies to mobile:radio nets,” IEEE Trans. Veh. Techno/., vol. VT-27,no.2,pp..57-64,1978. - ~ 4 M.:Duque&to& D. Kunz and B. Ruber, “Static and dynamic channel assignment using simulated annealing,”Neural Nehvorkr in Telecommunications. B. Yuhas and N.&sari, E&. Boston, MA:Kluwer, 1994. 5M. Sengokq “ Telephone traffic in a mobile radio comunication system using dynamic frequency assignments,’’IEEE Trans. Veh. Technol.. vo1.29, no. 2, pp. 270-278,1980. 6 A. Camst, “Homogeneous distribution of frequencies in a regular hexagonal cell system,” IEEE Trans. Veh. Technol.,vol. 31. no. 3,pp. 132-144,1982. 7 A. Gamst, “Some lower bounds for a class of frequency assignment problems,’’ IEEE Trans. Veh. Technol., vo1.35, no.I ,pp. 8- 14,1986. 8 K. Smith and M. Palaniswami, “Static ind Dynamic Channel Assignment using Neural Networks”, IEEE Jouml on Selected Areas in Communications, vol. 15, no. 2,pp. 238-249,1997. 9 E. Falkenauer, Genetic algorithms and grouping problems.Chichester, England: Wiley, 1998. 10 R. Matbar and 1. Mattfeldt, ” Channel assignment in cellular radio networks”, IEEE Trans. Veh. Technoi., Vo1.42,pp.1421, Feb 1993. 11.S.Kitq S. H. Park, P. W. Dowd, and N. M. Nasrabadi,“Channel assignment in cellular radio using genetic algorithm”, Wireless Persona: Commun, vo1.3, 110.3, pp.273-286, Aug.1996. 12 D. Beckmann and U. Killat, “A new strategy for the application of genetic algorithms to the channel assignment problem”, IEEE Trans. Veh.Technol., vol. 48, no. 4, pp.1261-1269, July, 1999.13 E. David Goldberg, Genetic algorithms in search.optimization, and machine learning. Reading, Mass.: Addison-Wesley Pub. Co., 1989. 14 K. Deb, “Multi-objective Optimization Using Evolutionary Algorithms”, John Wiley & Sons, 2001.15 Lawrence Davis, Handbook of Genetic Algorithms. New York VanNosbandReinhold, 1991. 16 K. A. Smith, “A genetic algorithm for the channel assignment problem.” IEEE Global Technoha Conference,vol. 4, 1998. 17 Donald E. Knuth, The Art of computer programming: Fundnmental Algorithms. n i r d Edition. Reading, Mass: Addison-Welsey Pub. Co., 1997 I8 T. Kohonen, “Self-organized formation of topologically correct feature maps, ”Biol.Cybern., vol. 43, pp. 59-69, 1982. 19 A. Thavarajah and W.H. Lam, “Heuristic approach for optimal channel assignment in cellular mobile systems,” IEE Proceedings Communications, vol. 146 3,pp. 196-200, June, 1999. 20 G. Chahborty and B. ChaLborty, “A genetic algorithm approach to solve channel assignment problem ~in cellular radio networks,” Proc. I999 IEEE Midnight-Sun Workshop on Soft Computing Methods in Industrial Applications, pp.3439, 1999. 21 M. Williams, “Making the best use of the airways:an important requirement for militaty communications,”Electronics & Communication Engineering Joumal.v01.12, no.2, pp.75-83, April, 2000. 22 F.J. Jaimes-Romero, D. Munoz-Rodriguez, and S.Tekinay, “Channel assignment in cellular systems using genetic algorithms,” IEEE 46th Vehicular Technology Conference, vol. 2, pp.741 -745, 1996. 23 W. K. Lai and G. G. Coghill, “Channel assignment through evolutionary optimization,” IEEE Transactions on Vehicular Technology, vo1.45, no.1, pp.91 -96, Feb.,1996. 24 C.Y. Ngo and V.0.K Li, “Fixed channel assignment in cellular radio networks using a modifiedgenetic algorithm,” IEEE Trans. Vehicular Technology,vol. 47, no. 1, pp. 163-172, Feb., 1998.