|
| |
Fields
of interest
SVMs, Sensor Networks, Machine Learning, Neural Network, Pattern Recognition, Signal Processing
and Control. Most of my papers are in the field of Machine Learning
(Publication List).
Research Interests
Click Here for SVMheavy Homepage
My research interests and some
my publications (Full List) are given below:
Modelling
and Control:
-
PALANISWAMI, M. and
Feng, G. - Digital Estimation and Control with a
New Discrete time Operator, in Proc. 30th IEEE Conference on Decision and
Control, Brighton, U.K., pp. 1631-1632 (1992).
-
Fok, C.S., DRANSFIELD,
P. and Palaniswami, M. - 'Model Reference
Velocity Control of an Air Powered Servo drive, Intl.Conf on Fluid Power,
Control and Robotics, Chengdu, China, (1990).
-
Palaniswami, M.
- A New Discrete-Time Operator for Digital Estimation and Control, Research
Report No. 1-1989, Dept. Of Electrical and Electronic Engineering, The
University of Melbourne
Disturbance Rejection:
-
Palaniswami, M.
- Adaptive Internal Model for Disturbance Rejection and Control, in Proc. IEE
- Part D, Vol. 140, No:1, pp. 51-59, (1993). ). (JIF - 0.47)
-
Feng, G. and Palaniswami, M.
- Continuous Time Indirect Adaptive Control with Disturbance Rejection,
Systems and Control Letters, Vol. 18, No: 3, pp. 211-215 (1992). (JIF - 0.69)
-
Feng, G. and Palaniswami, M. -
n Adaptive Control Algorithm for Disturbance Rejection -
Continuous Time Case, IEE Proceedings-D, Vol. 139, No. 2, March, pp. 167-171,
(1992). (JIF - 0.47)
-
Palaniswami, M. and Feng, G
- Adaptive Control Algorithms for Disturbance Rejection, Int. Journal.
Computers and Electrical Engineering, Vol. 17, No. 1, pp. 31-37, (1991). ).
(JIF - 0.07
-
Manzie, C., Palaniswami, M. and Watson, H.
- model predictive approach to disturbance rejection in idle speed control,
in proc. of the Asian Control Conference, Singapore, September, 2002.
-
Palaniswami, M. and Goodwin, G. C.
- Disturbance Rejection in Adaptive Control of Industrial
Processes, Proc. 14th Annual Conference of IEEE Industrial Electronics
Society, October, Singapore, 290-296 (1988).
-
Palaniswami, M. and Feng, G.-
Adaptive Control Algorithms for Disturbance Rejection, Proc.
4th I.E. Conference on Control Engineering, Gold Coast, Qld, pp. 95-99,
August (1990).
-
Palaniswami, M.,
- Adaptive Internal Model for Disturbance Rejection and Control, Research
Report No. 2-1989, Dept. of Electrical and Electronic Engineering, The
University of Melbourne.
-
Palaniswami, M.
- Adaptive Internal Model for Disturbance Rejection and Control, Research
Report No. 2-1989, Dept. Of Electrical and Electronic Engineering, The
University of Melbourne.
Neural Networks Architectures:
-
CHANDRASEKARAN, V., PALANISWAMI, M., AND CAELLI, T - Range
Image Segmentation by a Dynamic Neural Network Architecture, Pattern
Recognition Journal, Vol. 29, pp. 315 - 329, Feb. 1996. (JIF - 0.82)
-
CHANDRASEKARAN, V., PALANISWAMI, M AND CAELLI, T. - Spatio-Temporal
Feature Map’s using Gated Neuronal Architecture, IEEE Trans. on Neural
Networks, Vol. 6, pp. 1119-1131, September 1995. . (JIF - 1.5)
-
CHANDRASEKARAN, V. LIU, Z., AND PALANISWAMI, M - Fuzzy Gated
Neuronal Architecture for Pattern Recognition, Proc. IEEE Int. Conf on Neural
Networks, ICNN95, pp. 1622-1627, November, 1995.
-
Chandrasekaran, V., Palaniswami, M. and Caelli, T.
- Invariant Property of Spatio-Temporal Feature Maps using Gated Neuronal
Architecture, in Proc. IEEE International Conference on Acoustics, Speech and
Signal Processing, ICASSP94, pp. 609 - 612, Adelaide, 1994.
Combinatorial
Optimisation Problems:
-
SMITH, K., PALANISWAMI, M and KRISHNAMOORTHY, M. -
Solving Combinatorial Optimisation Problems Using Neural Networks, IEEE
Transactions on Neural Networks, Nov. 1998. (JIF - 1.5)
-
SMITH, K. AND PALANISWAMI, M. - Static and Dynamic Channel
Assignment using Neural Networks, IEEE Journal on Selected Areas in
Communications, Vol. 15, No:2, pp.238-249, 1997.(JIF - 1.11)
-
SMITH, K., KRISHNAMOORTHY, M. AND PALANISWAMI, M. - Neural
Versus Traditional Approaches to the Location of Interacting Hub Facilities,
Location Science, Vol.4, No:3, pp. 155 -171, 1996. (JIF - )
-
SMITH, K., PALANISWAMI, M. AND KRISHNAMOORTHY, M. - Traditional
Heuristic versus Hopfield Neural Network Approaches to a Car Sequencing
Porblem, European Journal of Operations Research (Special Issue), Vol. 93,
No:2, pp. 300-316, Sept. 1996. (JIF - 0.36)
-
SMITH, K., PALANISWAMI, M. AND KRISHNAMOORTHY, M. - A Hybrid
Neural Approach to Combinatorial Optimisation Problems, Computers and
Operations Research Journal (Special Issue), Vol. 23, pp. 597-610, June 1996.
(JIF - 0.43)
Neural
Networks for Mobile Communications:
-
TISSAINAYAGAM, D., EVERITT, D. and PALANISWAMI, M. - Feedback
Neural Network Implementation of Channel Packing Algorithms in Micro-Cellular
Cordless Telephony, Australian Journal of Intelligent Information Processing
Systems, pp. 18- 23, 1998. (JIF - )
-
SMITH,
K. AND PALANISWAMI, M. - Static and Dynamic Channel Assignment using Neural
Networks, IEEE Journal on Selected Areas in Communications, Vol. 15, No:2,
pp.238-249, 1997.(JIF - 1.11)
-
Chan, P.T.C. H.,
PALANISWAMI, M., and Everitt, D. - Neural Network Based Dynamic Channel
Assignment for Cellular Mobile Communication Systems, IEEE Trans. Vehicular
Technology, Vol. 43, No.2, pp. 279-288 (1994). (JIF - 0.47)
-
TISSAINAYAGAM, D., PALANISWAMI, M. AND EVERITT, D. -
Computational Complexities of Neural Networks based Dynamic Channel Assignment
Algorithms, in Proc. of 3rd IEEE International Conference on Intelligent
Processing Systems, ICIPS, pp. 560-564, Gold Coast, Australia, August 1998.
-
TISSAINAYAGAM, D.,
Everitt, D. and Palaniswami, M. - A Performance Comparison of Neural
Network Based Dynamic Channel Assignment in Cellular Mobile Radio Systems, in
Proc. of IEEE Singapore International Conference on Networks (SICON'97), pp.
181-196, Singapore, Apr. 1997.
-
TISSAINAYAGAM, D.,
Everitt, D. and Palaniswami, M. - A Neural Network Driven Solution to
a Channel Assignment Problem in Wireless Telephony, in Proc.of the
International Conference on Neural Networks (ICNN'97), Vol.1, pp. 133-137,
Houston, June 1997.
-
TISSAINAYAGAM, D.,
Everitt, D. and Palaniswami, M. - Mosaic Learning: A New Algorithm for
Self-Organising Neural Networks to Learn Dynamic Channel Assignment Schemes,
accepted for International Conference on Neural Information Processing
(ICONIP'97), Dunedin, Nov. 1997
-
TISSAINAYAGAM, D.,
Everitt, D. and Palaniswami, M. - Self-Organising Feature Map
Implementation of the Random Channel Search Algorithm, accepted for
Asia-Pacific Conference on Communications (APCC'97), Sydney, Dec. 1997
-
Terrill, S., Everitt, D.
and Palaniswami, M. - Neural Networks for Limited Channel
Rearrangement in Cellular Mobile Communication Systems, in Proc. IEEE
Conference on Neural Networks, Orlando, USA., June 26 - July 2, pp. 3592-3596
(1994).
-
Chan, P.T.C.H.,
Palaniswami, M., and Everitt, D. E. - Feed forward Neural Networks
Applications to Dynamic Channel Assignment for Cellular Mobile Communication
Systems, Proc. 1992 International Joint Conference on Neural Networks,
IJCNN'92, Beijing, China, November 3-6, pp. 218-223 (1992).
-
Chan, PTH, Palaniswami,
M. and Everitt, D. - Dynamic channel assignment for cellular mobile
radio systems using feed forward neural networks, in Proc. International Joint
Conference on Neural Networks, Singapore, pp.1242-1247, November (1991).
-
Terrill, S.,
Palaniswami, M. and Everitt, D - Feed forward Neural Networks for
Limited Channel Rearrangement in Cellular Mobile Communication Systems, in
Proc. Fifth Australian Conference on Neural Networks, ACNN94, Brisbane, pp.
246-249, February, 1994.
-
Terrill, S.,
Palaniswami, M. and Everitt, D - Limited Channel Rearrangement in
Cellular Mobile Communication Systems using Neural Networks, in Proc. Eighth
Australian Telegraphic Research Seminar, ATRS, Melbourne, pp. 80 - 89, (1993).
-
Chan, PTH, Palaniswami,
M. and Everitt, D - Dynamic channel assignment for cellular mobile
radio systems using self organising neural networks, in Proc. 6th Australian
Telegraphic Seminar, Wollongong, November (1991).
Signal Processing for
Information Retrieval:
-
PARK,
L., PALANISWAMI, M., and KOTAGIRI,
R. – Internet Document Filtering
using Fourier Domain Scoring, in Principles of Data Mining and Knowledge
Discovery, Luc De Raedt and Arno Siebes, Eds. Pages 362-372, Lecture Notes in
Artificial Intelligence (No: 2168), Springer Verlag, 2001.
-
PARK,
L., PALANISWAMI, M., and KOTAGIRI, R - A novel Web text mining method using the
Discrete Cosine Transform, Principles of Data Mining and Knowledge Discovery, Eloma, T, Mannila, H., Toivonen, H. August 2002, No:2431 in Lecture Notes in
Artificial Intelligence, pp. 385-396 (2002).
-
PARK L.A., KOTAGIRI, R., and PALANISWAMI, M.
- Fourier Domain Scoring: a novel document ranking method, Accepted for
publication in IEEE Trans. on Knowledge and Data Engineering, 2004. (JIF - 0.62)
-
PARK, L.,
KOTAGIRI, R. and PALANISWAMI, M. – A new
Implementation Technique for Fast Spectral based Document retrieval Systems,
Proc. of the IEEE International Conference on Data Mining (ICDM 2002), Dec.
2002, Pp. 346-353.
Adaptive Support Vector
Machines:
-
Shilton, A., and
Palaniswami, M. - Modified$\nu$-SV Method for Simplified Regression, in
Proc. of the International Conference on Intelligent Sensing and Information
Processing, published by the IEEE, Jan 4-7, Chennai, 2004.
-
Lai, D., Mani, N. and
Palaniswami, M. - An Extrapolated Sequential Minimal Optimisation
Algorithm for Support Vector Machines, In Proc. of International Conference on
Intelligent Sensing and Information Processing, IEEE Press, Jan 4-7, Chennai,
2004.
-
Shilton, A.,
Palaniswami, M., Ralph, D., and Tsoi, A.C. - Incremental Training in
Support Vector Machines, in Proc. of the International Joint conference on
Neural Networks, Washington, July 2001.
-
Palaniswami, M.,
Shilton, A., Ralph, D., and Owen B.– Machine learning using support
vector machines, International conference on Artificial Intelligence in Science
and Technology, Hobart, Australia, December 2000 (Keynote Talk Paper).
-
Shilton, A.,
Palaniswami, M., Ralph, D., and Tsoi, A.C - Incremental Training in
Support Vector Machines, in Proc. of the International Joint conference on
Neural Networks, Washington, 2001.
Emission Control for
environment Safeguard
-
Manzie, C.,
Palaniswami, M., Ralph, D., Watson, H., and Yiao, X - Model Predictive
Control of a Fuel Injection System with a Radial Basis Function Network
Observer, ASME Journal of Dynamic Systems, Measurement and Control, pp648-658,
December 2002. (JIF - 0.22)
-
Manzie, C.,
Palaniswami, M., and Watson, H. - "Gaussian Networks for Fuel Injection
Control," Proceedings of the
Institute
of Mechanical Engineers: Part D Journal of Automobile Engineering,
vol. 215, pp. 1053-1068, 2001. . (JIF - 0.25)
-
Dingli, R., Watson,
H.C., Palaniswami, M. and Milkins, E - Adaptive Control of Air-Fuel Ratio
for Optimizing the Fuel Efficiency of a Lean Burn Natural Gas Engine, in Proc.
FISITA Conference, Beijing (1994).
I have employed these skills and techniques for solving cross disciplinary
problems and bringing new fields together with interdisciplinary areas.
The end
user applications that have benefited from my work include Defence,
healthcare and environment.
Currently,
I am
leading an Australian Research Network bid for
the emerging area of Sensors
and Sensor Networks to be administered by the University of
Melbourne. The network will have the participation of 40 world class
researchers from various institutions including MIT, Cambridge, University of
Melbourne, Louisiana State Univ. One of the most important goals for this
centre is to bring researchers from Electrical Engineering, Computer Science and
Mathematics to work on challenging problems with immediate benefit to society.
Collaboration
The
research has made special links with Australian industry providing innovative
solutions to many practical problems such as smart sensor scheduling algorithms
for target tracking in Defence, content based search engines for security
analysis, electronic monitoring of fish species, automotive engine control,
automatic target recognition, auto scaling of ionograms, option pricing and
financial models, portfolio management, weather monitoring, and modelling of
blast furnaces. I have extensive collaboration with over 100 researchers
(including my students) from many countries including USA, U.K, Japan, Canada,
India and eight different Australian industries. As a result of taking much of
my theory to practice, a commercial software product intelligent warehousing
(ICS) is currently used as part of an overall package by the organisation and in
addition, there is a provisional patent in search engines for internet
technology industry. Much of the research work has also been found to be
valuable by the Australian Industry and other research organisations. This has
been demonstrated by the continuous financial support from these bodies and by
the establishment of active collaborative links.
The specific industries and
institutions that have worked as partners for doing joint work include
-
BHP Research Laboratories
-
Tenix Ltd
-
Melbourne
IT
-
ANZ Bank
-
National Australia Bank
-
Signal Processing Associates
-
Integrated Control Systems
-
Compumedics
-
Cardiac Dynamics
-
Intellirad solutions
The
active research collaboration with different educational institution include
-
California Institute of Technology, USA
-
Cambridge
University, UK
-
University of
California, Berkely,
USA
-
Stanford
University, USA
-
CSIRO Division
of Mathematics and Statistics,
Melbourne
-
CSIRO Division
of Manufacturing Technology, Melbourne
-
Florida
International University, USA
-
Nanyang
Technological University,
Singapore
-
Indian
Institute of Science,
Bangalore, India
-
Indian
Institute of Technology,
Chennai, India
-
Bureau
of Meteorology, Melbourne
-
Communications Research Laboratory, Japan
-
Defence Science and Technology Organisation, Australia
-
Monash
University, Melbourne
More than 10
international visitors have made collaborative visits resulting in excess of 13
research publications. The joint publication is evidence of the success of the
collaborative visits.
Sensor
Networks (www.sensornetworks.net.au)
The
convergence of the internet, communications, and information technologies with
techniques for miniaturized sensors has placed the field of sensor information
technology at the threshold of a period of major growth. There is increasing
evidence that the new emerging technologies can decrease the size, cost of
sensors and sensor arrays by orders of magnitude and increase their spatial and
temporal resolution. In this regard, it is important to observe the current
international trend: The National Science Foundation (NSF) has committed for a
fundamental research program of over US$33million p.a in sensors and sensor
networks, the Taiwanese government has committed $750 million for sensor
technologies and US government has released a figure of $4.5 billion commitment
for works related to sensors and applications including defence. The compelling
question is: What can we do with these sensors in order to exploit their
potential fully. The current challenges are developing highly innovative
techniques to make sensors smart, to scale well and to apply for problems and
areas that were previously not considered. In order to meet the future demands,
it is necessary for sensor systems to leverage and incorporate projected
advances in adjacent technologies such as nanofabrication, bio-systems,
massively distributed networks, wireless communication technologies, control and
signal processing, ubiquitous computing, information and decision systems. The
end user applications, in defence, healthcare and environment, are highly
exciting and valuable to the community, and include projects such as bush fire
detection (tiny networked sensors activated at bush fire to communicate location
and spread), sensor monitoring of vast agricultural fields (bio-degradable
networked sensors used to determine the nutrition and health levels for healthy
crops), unmanned aerial vehicles for aerial surveillance and monitoring,
intelligent transportation system (autonomous vehicles equipped with smart
networked sensors navigating through GPS support), biodegradable sensors
monitoring ocean health (ref. world bank project for protecting the health of
Great Barrier Reef), smart monitoring of body function (with liquid sensors
mapping the body dynamics) and integrated sensor management and monitoring for
healthcare (ability to provide diagnostic services via communication of
important information to central nodes for processing).
In order to
fully achieve the benefit of these challenging applications, there is a serious
need to solve many fundamental scientific problems in the areas of intelligent
sensors, biomimetic systems, nonlinear control, machine learning, stochastic
scheduling, sensor fusion and tracking. Many of the scientific challenges call
for skills from interdisciplinary areas for providing viable practical
solutions. Having worked on cross disciplinary and inter disciplinary areas at
the highest levels, I have the vision to lead and steer the field with active
contributions from some of the best academics and industries and from my
contributions in the core areas of control engineering, optimization, adaptive
learning and signal processing. Our fundamental research projects that
contribute to the long term visionary areas are
listed
below:
-
Sensor
Scheduling for Sensor Networks
-
Adaptive
Support Vector Machines
-
Bio
inspired Fusion and Tracking
-
Sensor
Information Retrieval
-
Bioinformatics and Biomedical Engineering
|
