Date of Award

Fall 2013

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Electrical and Computer Engineering

First Advisor

Michael D. Zoltowski

Committee Chair

Michael D. Zoltowski

Committee Member 1

Mark R. Bell

Committee Member 2

David J. Love

Committee Member 3

Chih- Chun Wang

Abstract

In this dissertation, we deal with radar systems where multiple radar waveforms are simultaneously transmitted from different antennas. The goal is to process the returns in a way that the overall ambiguity function is a sum of individual ambiguity functions, such that the sum better approximates the ideal thumbtack shape. The transmitted waveforms are based on complementary sequences, which have this property that their individual autocorrelation functions sum to a delta function. A unitary design for the illustrative 4x4 example prescribes the scheduling of the complementary sequence based waveforms over four transmit antennas over four PRIs. Further, it dictates how the matched filtering of the returns over four PRIs are combined in such a way so as to achieve both perfect separation (of the superimposed returns) AND perfect reconstruction. Perfect reconstruction implies that the sum of the time-autocorrelations associated with each of the four waveforms is a delta function. The net result of the processing of four antennas over four PRIs yields 16 cross-correlations all of which ideally exhibit a sharp peak at each target delay. Conditions for both perfect separation and perfect reconstruction are developed, and a variety of unitary waveform sets satisfying both are proposed. Since complementary sequences are very sensitive to Doppler, a scheme for Doppler compensation is proposed based on a data dependent weighting of the different PRI matched-filtered outputs prior to summing. The scheme is further improved to provide a reasonable estimate of the Doppler phase shift using the MUSIC algorithm. A new Doppler estimation and compensation scheme based on a clever application of DFT is also developed, which provides an improvement in both in detection performance, and processing times. We have also developed a channel estimation scheme for MIMO-OFDM systems based on our proposed designs. We show that our designs are not limited to symmetric MIMO systems, and that they work for arbitrary transmit and receive antenna dimensions. Proof-of-concept simulations are presented verifying the efficacy of the proposed unitary waveform matrix designs in conjunction with the proposed Doppler compensation and estimation techniques.

Share

COinS