Large-Array Signal Processing for Deep-Space Applications

This article develops the mathematical models needed to describe the key issues in using an array of antennas for receiving spacecraft signals for DSN applications. The detrimental effects of nearby interfering sources, such as other spacecraft transmissions or natural radio sources within the array’s field of view, on signal-to noise ratio (SNR) are determined, atmospheric effects relevant to the arraying problem developed, and two classes of algorithms (multiple signal classification (MUSIC) plus beam forming, and an eigen-based solution) capable of phasing up the array with maximized SNR in the presence of realistic disturbances are evaluated. It is shown that, when convolutionally encoded binary-phase shift keying (BPSK) data modulation is employed on the spacecraft signal, previously developed data pre-processing techniques that partially reconstruct the carrier can be of great benefit to array performance, particularly when strong interfering sources are present. Since this article is concerned mainly with demonstrating the required capabilities for operation under realistic conditions, no attempt has been made to reduce algorithm complexity; the design and evaluation of less complex algorithms with similar capabilities will be addressed in a future article. The performances of the candidate algorithms discussed in this article have been evaluated in terms of the number of symbols needed to achieve a given level of combining loss for different numbers of array elements, and compared on this common basis. It is shown that even the best algorithm requires approximately 25,000 symbols to achieve a combining loss of less than 0.5 dB when 128 antenna elements are employed, but generally 50,000 or more symbols are needed. This is not a serious impediment to successful arraying with high data-rate transmission, but may be of some concern with missions exploring near the edge of our solar system or beyond, where lower data rates may be required.