Pattern Synthesis Of Smart Antennas Based On A Improved Genetic Algorithm

In the early stage of the issued fourth generation mobile communication license,China’s independent research and development of the fourth generation mobilecommunication technology standard strive for further improvement based on the thirdgeneration mobile communication. The three major operators will contend for thefourth generation mobile communication operation licence. In the fierce marketenvironment, the key is to obtain the appropriate licences for competition in themarket. The smart antenna is one of the key technology for the fourth generationmobile communication technology,which can improve not only the call quality, butalso the capacity of the whole communication system. A top priority is how to designthe antenna in mobile communication. The three major indicators of smart antenna,namely the main lobe position, low side lobe and zero notch characteristics,are themain problem in smart antenna design。The main lobe position determines whetherthe mobile user can obtain clear call quality。The low side lobe can restrain thedisturbance to the surrounding environment noise。The zero notch can inhibit theinterference between users and users effectively。In this paper， three index will befocused on the study of smart antenna。This paper introduces the background of mobile communication and situation ofsmart antenna development. Pattern synthesis of antenna array is the key technologyof smart antenna, and array antenna optimize the radiation field mainly by changingarray element distance, the excitation current size and the number of array element。This paper mainly discusses the effect of the radiation field by changing the size ofthe excitation current, and the main lobe position is controled by changing the phasesize, so that the main lobe position is satisfied with the design requirements. The sidelobe and generating zero notch are controled by changing the current amplitude. Thispaper also introduces the based comprehensive method for linear array as well as theclassic array antenna。Based on biological "survival of the fittest" principle, genetic algorithm breedsnext generation by some methods such as choose, crossover and mutation。 Thegenetic algorithm has advantages in solving the multiparameter, nonlinear andcomplicated problem。 And it is s widely used in the area of antenna design in recentyear。In this paper, some disadvantages of ordinary genetic algorithms are improved, which takes four methods。this is the single point crossover, arithmetic crossover,extrapolation and operator migration to produces the next generation。In the processof variation in the genetic algorithm, improvements that the original fixed geneticprobability increases with the increasing in the number of iterations are put forward tothe genetic algorithm the mutation based on the analysis of early and late on thevariation in different requirements. The iteration of genetic algorithm is greatlyreduced in terms of a variety of methods, which saves time and reduces therequirement for computer hardware. The simulation of the linear array shows thecorrectness of the improved genetic algorithm.