Bimonthly Newsletter

Volume 1, Issue 2

June 2007


Dear fellow researchers,


Its time again for update with our bimonthly newsletter. This issue features a write-up by ISSNIP researcher Daniel Lai about the application of wireless sensor networks and computational intelligence to the analysis of human gait and movement. As usual, we have a couple of news, events and seminars to announce. Hope you enjoy this issue. Thank you.


Best regards,



A pdf of this newsletter can be downloaded from here.

Table of Contents


Feature Article: Gait and Human Movement Analysis: The Need for Computational Intelligence

by Daniel T.H. Lai


Our societys growing demand for a better quality of life is set to affect every aspect of the global economy, more so the healthcare sector since health is increasingly linked to our quality of life. Current worldwide shortages in medical personnel have immensely strained present healthcare systems in the midst of a growing population [1]. In view of this, immediate plans for new technologies and paradigms must be actively pursued to provide efficient and effective avenues to sustain the anticipated increase in healthcare demand. 


The area of gait and human movement science deals primarily with our locomotion ability. Locomotion is an extremely complex phenomenon, considering the fact that our body spends 80% of its time being supported by a single foot during walking. Gait analysis is the study of the biomechanics involved in human movement such as walking and running. This study is important for comprehension of the factors governing the well being of the lower extremities as locomotion is vital to ensuring the smooth execution of daily activities. Research in this area has been shown to be extremely useful in the quantification of movement function while dealing with many clinical and health-related problems. Current studies in this field encompass applications such as the detection of movement disorders, identification of biomechanical factors associated with ageing and balance, and the assessment of clinical interventions and rehabilitation programs.


Clinical gait disorders afflict all age groups and can arise from pathologies in the lower extremities such as patellofemoral pain syndrome, knee osteoarthritis and tendon rapture which result in antalgic gait. Pathologies such as Cerebral Palsy, Parkinsons disease, dementia and stroke lead to abnormal gait patterns. These diseases are more severe as they disrupt human locomotion control which is important for balance and safe negotiation of the environment. Aging effects on the other hand, give rise to a different set of gait disorders. Studies in this area are fast gaining prominence because the elderly population (above 65 years) is projected to increase by 55% between 2015 and 2030 due to improving healthcare. This age group is highly susceptible to balance impairments that cause slipping and tripping falls which incur high medical costs and alarming mortality rates.


Initial gait studies have focused on identifying suitable gait variables that best characterize a particular disorder. Contemporary gait variables using kinematics information (such as joint angle measurements and time-distance variables), and kinetics (e.g., ground reaction forces obtained from force platforms) have been frequently investigated to identify gait disorders, assess the effectiveness of rehabilitation programs and monitor post-surgery recovery.  These variables have been commonly analysed using statistical and linear analysis methods such as linear discriminant analysis. More recently, gait variables such as the minimum toe clearance during gait have been modelled as a time series giving rise to application of digital signal processing techniques such as Fourier and wavelet analysis. The advantage of signal modelling is that nonlinear dependencies characteristic of a disorder may be extracted from a sequence of gait cycles as opposed to single statistical quantities such as the mean or standard deviation.


As with other biomedical areas, statistical and signal processing information from gait studies are indirectly used to describe the outputs of complex biomechanical systems which are difficult to explicitly model.  System modelling, however, is important in gait analysis as it facilitates the understanding of the underlying aetiology of the gait disorder allowing more accurate diagnosis and focused treatment. Gait analysis studies usually yield large volumes of data, making analysis a difficult if not dreaded task for the clinician. A current promising technology is computational intelligence (CI) which is a fusion of artificial intelligence and computing, specifically suited for the design of powerful decision systems capable of interpreting and processing large volumes of data. CI is gaining rapid popularity in several areas of healthcare such as intelligent detection systems for aiding the diagnoses of diseases, the study of protein structures for drug design and more pervasive patient monitoring systems for tracking patient recovery.


These techniques which include artificial neural networks, support vector machines, fuzzy classifiers and genetic algorithms offer newer and more powerful modelling techniques by learning implicit relationships between pathologies and patient data [2]. CI techniques are frequently applied to pattern recognition and function estimation problems, and have been found to be relatively cost effective to implement requiring only computer software. Implementation is not intricate as statistical and signal processing information can be seamlessly integrated with CI methods by pairing them with respective gait pathologies. These pair-wise relationships constitute labelled input data which can then be used by CI methods to learn the complex underlying biomechanical process. This integration unfortunately has not been fully utilized in human movement sciences, the primary reason being that clinicians still view these black box techniques with a certain degree of suspicion.


Today, automated gait systems employing intelligent CI technology still face several technical challenges, such as guaranteed diagnostic accuracies, robustness to patient variability, practicality of implementation, and convenience of usage before widespread deployment is possible. Future automated gait systems may also be integrated with other enabling and emerging technologies such as sensor networks (SN). SN technology opens new possibilities to gait studies, where for example, wireless sensor technology would allow online gait monitoring in natural walking environments instead of experimental restrictions imposed by current data collection methods in gait laboratories. Ultimately, the onus on researchers (thats us) is to demonstrate that these systems can be accepted and confidently utilized by clinicians and patients alike. In view of the aforementioned issues, there has never been a better time to undertake research in this sector, more so as these technologies are currently poised towards becoming an affordable and ubiquitous healthcare solution, promising vast benefits for continued improvement of the global quality of life! 


[1]  World Population Report (WHO) 2006, Online:

[2]  Schollhorn, W. I. , Applications of artificial neural nets in clinical biomechanics, Clinical Biomechanics, 19, 876-898, 2004. 

Past Seminars

Prof Tzyy-ping Jung, associate director of Schwatz Institute at the University of California, San Diego, is a specialist in biosignals. He gave a seminar entitled Identifying drowsiness for automobile drivers using wireless EEG, covering topics ranging from signal classification and biomedical engineering, to MEMS and VLSI technology. He also introduced the state-of-the-art EEG wireless electrode system that is being developed at the 0.18ᄐm 8in-wafer micro-fabrication foundry at National Chiao Tung University, Taiwan, where he is also affiliated with. The tiny wireless sensors could potentially revolutionize the application of wireless sensor networks to the measurements of biosignals.


Professor Jung giving a lecture at RMIT. 

Professor Jung giving a lecture at RMIT                         Prof Jung giving a lecture at the University of Melbourne



Upcoming Seminars

Title:            Sensor Web

Speakers:     Sutharshan Rajasegarar and Slaven Marusic

Venue:         Brown Theatre

Date:           8th June (Friday)

Time:           11am - 12pm

Abstract:      Sensor Web provides a portal for querying and visualising sensor network data in real-time using geographical interface like windows live local. This is to utilise for the dissemination of data from the Great Barrier Reef sensor network deployment. This talk will provide a basic overview of the SensorMap tutorial, which was conducted in conjunction with IPSN 2007 conference. We will also give a brief review and pointers to selected papers from the conference.


Title:            Happy Sheep - Decentralised Movement Analysis

                    through Local Relational Sensing

Speaker:      Patrick Laube, Department of Geomatics, the University of


Venue:         Brown Theatre

Date:           15th June (Friday)

Time:           11am to 12pm

Abstract:      Advances in location sensor technology offer new reliable and cost-effective means to track moving objects. Being spatio-temporal in nature, movement data tends to be voluminous and hence requires sophisticated analysis techniques to derive high level event knowledge from low level trajectory data. Conventional movement analysis bases on centralised information processing, where global systems such as spatial databases or GIS collate all information in the system and run omniscient data mining algorithms in order to detect salient events. However, when modelling and analysing movement of location-aware roaming agents, such global approaches fail due to the known energy and communication constraints in sensor systems. Hence, new decentralised models are required, where individual moving computing elements infer from locally sensed relational information only whether they are part of a movement pattern. Agents shall not only be 'location-aware' but 'spatially-aware', inferring themselves whether they are on collision course or run into traffic jam. Reviewing recent research on movement analysis, this talk makes a strong argument for the development of techniques for decentralised, in-network detection of salient movement patterns in wireless ad-hoc networks of location-aware moving agents and presents work in progress implementing such decentralised approaches. 

Seminars are always announced at:


Event Announcements

DG/SUM'07: is an international workshop on distributed and mobile spatial computing held in conjunction with COSIT'07, the Eighth International Conference on Spatial Information Theory. The workshop will be held on 19 September 2007, the day before COSIT'07 begins, at Mount Eliza, Melbourne, Australia, the same venue as COSIT'07. More details at: 


Workshop on Optimization in Sensor Networks: 3 December 2007 (Tentative) at the Langham Hotel, Melbourne

Chairs: Adil Bagirov, Ballarat University and Daniel T.H. Lai, the University of Melbourne 

     This one day workshop will be held in conjunction with the ISSNIP 2007 conference (3-6 December 2007) and aims to bring together Australian and international researchers to discuss recent optimization problems in sensor networks that require further attention and scrutiny. It is hoped that this event will bridge the gap between engineers and mathematicians in the field, encourage collaboration efforts among researchers and students, foster better international research ties and generate more awareness of the problem diversity in this field. The workshop will cover optimization problems in the following sensor network areas (but not limited to): 

a)    Localization in sensor networks

b)    Energy efficiency schemes in wireless sensor networks

c)    Optimal sensor placement for tracking

d)    Scheduling for optimal sensor coverage 

Prospective authors are invited to submit papers of 4-6 pages in length following the ISSNIP 2007 conference format ( Papers can be submitted directly to the chairs, Adil Bagirov ( or Daniel T.H. Lai ( 

Important Dates:

Submission Closes: 31st August 2007

Notification of Acceptance: 1st October 2007

Camera Ready: 15th October 2007


ISSNIP 2007 Conference  

As you may already know, the ISSNIP 2007 conference would be held in Melbourne from 3 to 6 December. The following is a preliminary listing of symposia that will be held as part of ISSNIP 2007: 

l         Symposium on Middleware for Sensor Networks (MiSS'07).
Chair: Mohan Kumar (The University of Texas, Arlington, USA)

l         Symposium on Autonomous Configurability and Control in Dynamic Wireless Networks.
Chair: Stuart Milner (University of Maryland, USA) and Sylvie Perreau (University of South Australi

l         Symposium on Machine Learning and Applications.
Siddhi Kulkarni (University of Ballarat, Australia) and Brijesh Verma (Central Queensland University, Australia)

l         Symposium on Environmental Sensor Networks.
Chair: Stuart Kininmonth (Australian Institute of Marine Science)

l         Symposium on Sensor Fusion, Intelligent Sensors and Applications.
Chair: Danil Prokhorov (Toyota, USA) and Thomas Hanselmann (University of Melbourne, Australia)

l         Symposium on Bio-signal Processing and Networked Sensors in Healthcare.
Chair: Rezaul Begg (Victoria University, Australia) and Dinesh Kant Kumar (RMIT University, Australia)

l         Symposium on Sensor Networks.
Chair: Paul Havinga (University of Twente, The Netherlands)

l         Symposium on Information Processing in Sensor Networks.

Chair: Salim Bouzerdoum (University of Wollongong, Australia) 

l         Symposium on Computational Intelligence for Sensor Networks.

                Chair: G. K. Venayagamoorthy (University of Missouri-Rolla, USA)

Detailed information is available here:


Editorial Board:  Yee Wei LAW, Paul HAVINGA, Slaven MARUSIC, Marimuthu PALANISWAMI


ISSNIP Bimonthly Newsletter, ARC Research Network on Intelligent Sensors, Sensor Networks and Information Processing

Contact Details: M Palaniswami, Convener, ARC RN ISSNIP,,

Dept. of Electrical and Electronic Engineering, The University of Melbouren, Australia

top of page