Date of Award

Fall 2006

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Electrical Engineering - (Ph.D.)

Department

Electrical and Computer Engineering

First Advisor

Alexander Haimovich

Second Advisor

Yeheskel Bar-Ness

Third Advisor

Ali Abdi

Fourth Advisor

Rick S. Blum

Fifth Advisor

Leonard J. Cimini

Abstract

Motivated by recent advances in Multiple Input Multiple Output (MIMO) wireless communications, this dissertation aims at exploring the potential of MIMO approaches in the radar context. In communications, MIMO systems combat the fading effects of the multi-path channel with spatial diversity. Further, the scattering environment can be used by such systems to achieve spatial multiplexing. In radar, a complex target consisting of several scatterers takes the place of the multi-path channel of the communication problem. A target's radar cross section (RCS), which determines the amount of returned power, greatly varies with the considered aspect. Those variations significantly impair the detection and estimation performance of conventional radar employing closely spaced arrays on transmit and receive sides. In contrast, by widely separating the transmit and receive elements, MIMO radar systems observe a target simultaneously from different aspects resulting in spatial diversity. This diversity overcomes the fluctuations in received power. Similar to the multiplexing gain in communications, the simultaneous observation of a target from several perspectives enables resolving its features with an accuracy beyond the one supported by the bandwidth. The dissertation studies the MIMO concept in radar in the following manner. First, angle of arrival estimation is explored for a system applying transmit diversity on the transmit side. Due to the target's RCS fluctuations, the notion of ergodic and outage Cramer Rao bounds is introduced. Both bounds are compared with simulation results revealing the diversity potentials of MIMO radar. Afterwards, the detection of targets in white Gaussian noise is discussed including geometric considerations due to the wide separation between the system elements. The detection performance of MIMO radar is then compared to the one achieved by conventional phased array radar systems. The discussion is extended to include returns from homogeneous clutter. A Doppler processing based moving target detector for MIMO radar is developed in this context. Based on this detector, the moving target detection capabilities of MIMO radar are evaluated and compared to the ones of phased array and multi-static radar systems. It is shown, that MIMO radar is capable of reliably detecting targets moving in an arbitrary direction. The advantage of using several transmitters is illustrated and the constant false alarm rate (CFAR) property of adaptive MIMO moving target detectors is demonstrated. Finally, the high resolution capabilities of MIMO radar are explored. As noted above, the several individual scatterers constituting a target result in its fluctuating RCS. The high resolution mode is aimed at resolving those scatterers. With Cramer Rao bounds and simulation results, it is explored how observing a single isotropic scatterer from several aspects enhances the accuracy of estimating the location of this scatterer. In this context a new, two-dimensional ambiguity function is introduced. This ambiguity function is used to illustrate that several scatterers can be resolved within a conventional resolution cell defined by the bandwidth. The effect of different system parameters on this ambiguity function is discussed.

Share

COinS