Study on Numbers of Array Elements of Radio Interferometry

There are practical needs to recover surface brightness distributions of objects from their interferometric patterns of radio bands. Good quality of a generated image is of great importance, but usually needs to be achieved with quite a cost in instruments and operation. Radio astronomers have thus been concerned about the problem of how to obtain the best images with an interferometric array at a limited cost. Most observers with interferometric arrays set optimized numbers of array elements. The optimization of number of array elements is still an unsettled problem and this paper discusses it in details. We design simulations of synthetic-aperture imaging to help the decision of numbers of array elements. The simulated interferometric arrays constitute of antennas numbering from 20 to 100. The variations of array performances measured in SL, ECR, and RMSE are obtained by optimizing with the Genetic Algorithm, and are fit with functions. The decision of number of array element is based on methods such as effective cutoff and IAEE. From the analysis of simulations, we conclude that building antennas from 50 to 60 will be appropriate for a low-frequency solar array. Using about 50 antennas is proposed as the approach to minimize the cost. However, if the performances are to be improved, 60 or even more antennas should be used. In practice additional factors such as the budget and topography need to be considered to determine the optimized numbers of array elements.