Application of genetic algorithm to the problem of optimizing the geometric structure of an extended antenna field

Large-diameter radio telescopes (32…70 m) are a standard instrument for space exploration by radio engineering methods. Currently, research is underway to replace such antennas with an antenna field consisting of smaller diameter elements in order to reduce the cost of the system and simplify its maintenance. However, when constructing such fields from the ratio of wavelength (frequency) and the physical dimensions of the elements, it turns out that the inter-element distance is more than half the wavelength, i. e. the antenna field is extended. In extended fields, lateral maxima are observed that differ from the main ray – the side lobes. An important task is to reduce the level of the side lobes (SLL). A decrease in SLL is possible due to the amplitude-phase distribution (APD) or (and) due to a change in the geometric structure of the antenna field. In this article, the second option is considered, since all the elements are assumed to be identical. A genetic algorithm (GA) has been developed that enables to find the location of antenna elements relative to each other to minimize SLL. Due to GA, it is possible to improve the SLL by 15 % compared to a completely random array. The GA effect is equivalent to adding 20…30 elements to the array, i. e. due to GA, the cost of the system can be reduced and it can be used more efficiently for radar.