On the Optimization of Mobile Terminal Antenna Isolation using the Genetic Algorithm Technique

In this paper the optimization of the antenna isolation using the genetic algorithm (GA) technique is discussed. The goal is to find a suitable fitness function for measuring both the impedance matching and the isolation levels of two mobile terminal antennas. Three different approaches for measuring the isolation are tested and compared. In the first two a direct condition to the level of the isolation is set, when the effects of the matching levels of the antennas are included or excluded in the detected isolation. In the third approach a 'penalty' condition is included in the antenna impedance matching condition for operating at the frequency band of the other antenna. The approaches are tested statistically by optimizing a planar antenna pattern on top of a metal-coated printed wiring board. The area for the antenna pattern is equipped with two fixed feed pins and the short pins are left for the GA to optimize. The test results show that it is possible to design antenna patterns with at least 6 dB return loss and 10 dB isolation using any of the three approaches. However, with the selected GA parameters the first approach is considered to be the most reliable one. (6 pages)