Multiple constraint space-time direct data domain approach using nonlinear arrays

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)


Electrical Engineering and Computer Science


Tapan K. Sarkar

Second Advisor

Magdalena Salazar-Palma

Third Advisor

Michael C. Wicks


Space-time, Direct data domain, Nonlinear arrays, Signal processing

Subject Categories

Electrical and Computer Engineering | Engineering


This dissertation is a study of two existing array signal processing techniques. The first technique is the space-time direct data domain least squares (D 3 LS) algorithm. This adaptive signal processing algorithm relies on knowledge of characteristics of the signal of interest (SOI), such as the angle of arrival and frequency, to extract the signal from an interference environment. A powerful feature of this technique is its ability to operate effectively against coherent and non-coherent interference, without the use of secondary data. This algorithm operates on a snapshot by snapshot basis. In this work we show that the performance of this algorithm is severely degraded when the characteristics of the SOI vary slightly from its assumed values. We propose a novel multiple constraint solution to overcome this degradation, and demonstrate the performance improvement using simulated radar data.

The second technique is an array data transformation technique. This transformation approach has been used to convert voltages of signals received on a non-uniform, nonlinear array to those received by a virtual uniform linear array. This has been done using one dimensional, spatial domain data, with and without mutual coupling. In this work we extend this technique to space-time data, and demonstrate the utility of this approach using several examples. We then combine the array data transformation approach and the multiple constraint space-time D 3 LS approach and study their performance using simulated airborne radar data. In this analysis the performance of the modified techniques are evaluated using the adaptive processor's input and output signal-to-interference-plus-noise ratio (SINR) and the pattern of the output weight vector.


Surface provides description only. Full text is available to ProQuest subscribers. Ask your Librarian for assistance.