EMRS DTC >
Research Programme > Transducer
Embedded Processing
Transducer Embedded Processing
SIGNAL PROCESSING BASED SENSOR IMPROVEMENTS
SUMMARY
Many hardware systems are operating
close to their physics limits of performance. Where
this is not the case then they are operating on the
limits of affordability and cannot be improved greatly
without significant extra cost. In this environment
there is strong motivation to improve signal processing
methods in order to achieve greater performance from
current hardware. Alternatively signal-processing methods
may be used to allow lower performance hardware to be
deployed in order to achieve existing performance at
a lower cost. Currently these signal-processing enhancements
are carried out on an ad-hoc basis to solve specific
programme problems as they arise. The objective of this
proposal is to carry out a more coherent programme of
signal-processing-based improvements.
MILITARY BENEFITS
If sensor functionality or affordability
can be improved by software additions to existing hardware
then clearly the resultant capability will benefit the
military. Also the hidden benefit of having a flexible
software-based system capable of rapid update and evolution
should not be overlooked.
RESEARCH OBJECTIVE
To address sensor resolution limits, by looking
at superresolution methods to determine whether
smaller apertures and narrow bandwidth systems
can replace existing sensors. The robustness of
the improvement factors will be researched.
To address a key limitation of existing systems,
the inability adequately to suppress clutter.
A number of signal processing methods have emerged
that need further refinement. These include both
slow time and fast time space-time adaptive processing
(STAP) methods.
To address the problem of the influence of multipath
on tracking performance. Signal processing methods,
such as blind signal separation, superresolution
and fast time STAP are emerging as techniques
for overcoming these problems.
RESEARCH OUTLINE
This programme will include an over-arching
study to determine the current limits on system performance
and the state of the art in terms of mitigation methods.
Both hardware and signal processing based solutions
will be reviewed, a full loss budget for the signal
processing chain will be developed in order to identify
inefficient algorithms etc. Where significant losses
are identified, research will be carried out to determine
the potential for reducing them. This could lead to
proposals for substantial work in following years of
the DTC.
The first year activities will contain
a study into the use of superresolution techniques and
their robustness in the real world. The study will determine
the impact on radar antenna size and cost as well as
performance improvements that may be possible by utilising
this signal processing method.
Multipath mitigation methods will
be modelled and, in particular, the effect of applying
these techniques on a tracker will be determined. This
programme will benefit from the ability to access real
data from MESAR2.
Overall during year one the programme
will develop a list of further research topics that
will need to be addressed alongside a list of the potential
gains that can be achieved from pursuing these projects.
These gains will be measured in both performance improvements
and cost savings in hardware.
CO-ORDINATION WITH EXISTING /
PREVIOUS RESEARCH
This research will make maximum use
of internal funded programmes in this area. It will
also endeavour to align itself with related CRP programmes.
Clearly if a DTC is established in the area of signal
processing then significant alignment could occur but
it is the intention of this programme to remain very
close to the problem of transducer enhancement via signal
processing rather than do much fundamental algorithm
development.
