Spatiotemporal array signal processing in MIMO radar for moving targets in ground clutter
File(s)
Author(s)
Ren, He
Type
Thesis
Abstract
This thesis is concerned with the parametric estimation problem in an arrayed
MIMO radar, which is defined as the radar type using antenna arrays at both its
transmitter and receiver.
Initially, the parametric multitargets estimation problem in arrayed MIMO
radar is presented in conjunction with the modelling of conventional MIMO radar.
The novel concept of the “manifold extender” is then introduced providing a huge
increase of degrees of freedom (DoF). In particular, two “Manifold Extender” are
presented, where the first approach incorporates the Doppler-STAR manifold and
the second approach utilises the spatiotemporal-virtual manifold.
Then, a comprehensive modelling for multi-targets parameter estimation is
presented and the effects of ground clutter is discussed under the arrayed MIMO
radar framework. Clutter properties of target-correlated and low-rank will be
mainly considered. Accordingly, a novel subspace is proposed to successfully isolate
the targets from the clutter and this is supported by computer simulation studies.
Finally, a novel spatiotemporal manifold vector is proposed by incorporating.
the Rx array geometry and the symbol-to-symbol Doppler frequencies, as well
as a transformation matrix. This approach is capable of isolating the clutter
effect from the received signal and accordingly, a new subspace-based method for
estimating the actual path delays, radial velocities and directions of arrival (DOA)
of all targets is presented. Using a representative example of a “base-protection”
scenario, the proposed approach demonstrates a great ability for handling the
multi-target parameter estimation problem in the presence of ground clutter.
MIMO radar, which is defined as the radar type using antenna arrays at both its
transmitter and receiver.
Initially, the parametric multitargets estimation problem in arrayed MIMO
radar is presented in conjunction with the modelling of conventional MIMO radar.
The novel concept of the “manifold extender” is then introduced providing a huge
increase of degrees of freedom (DoF). In particular, two “Manifold Extender” are
presented, where the first approach incorporates the Doppler-STAR manifold and
the second approach utilises the spatiotemporal-virtual manifold.
Then, a comprehensive modelling for multi-targets parameter estimation is
presented and the effects of ground clutter is discussed under the arrayed MIMO
radar framework. Clutter properties of target-correlated and low-rank will be
mainly considered. Accordingly, a novel subspace is proposed to successfully isolate
the targets from the clutter and this is supported by computer simulation studies.
Finally, a novel spatiotemporal manifold vector is proposed by incorporating.
the Rx array geometry and the symbol-to-symbol Doppler frequencies, as well
as a transformation matrix. This approach is capable of isolating the clutter
effect from the received signal and accordingly, a new subspace-based method for
estimating the actual path delays, radial velocities and directions of arrival (DOA)
of all targets is presented. Using a representative example of a “base-protection”
scenario, the proposed approach demonstrates a great ability for handling the
multi-target parameter estimation problem in the presence of ground clutter.
Version
Open Access
Date Issued
2017-12
Date Awarded
2018-08
Advisor
Manikas, Athanassios
Publisher Department
Electrical and Electronic Engineering
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)