Department Publications

The biological world is an ideal place for seeking inspiration for developing mathematical optimization algorithms. In this paper we propose two hybrid stochastic optimization algorithms that bear resemblance to the sexual reproduction cycle of Jellyfish and asexual reproductive cycle of species of Hydra. The performance of these two algorithms are investigated against other common optimization algorithms on a set of benchmark optimization problems. The results show that the proposed algorithms perform well.

Brain-Computer Interaction (BCI) is a technology developed with the purpose of building a pathway between the brain and computer which is independent of neuromuscular functions. Potential applications in rehabilitation of patients with motor disabilities and video gaming make BCI an important field of research. A task like controlling a prosthetic limb using BCI is challenging. Performing this with readily available consumer grade EEG devices complicates the matters further due to lower accuracies. This paper presents the work related to an online classifier for imagined hand movement implemented using Emotiv Epoc for EEG data capturing. The system attempts to discriminate between left and right hand movement imagination by analysing the recordings of two electrodes placed over the motor-cortex. Auto-Regression (AR) based signal processing techniques are employed to derive features that enable …

Listening to the heart sounds is a common practice in identifying cardiac malfunctions. Since this method has many limitations, tools that aid physicians in their diagnosis of heart diseases are very useful. This paper presents a software tool to predict cardiac abnormalities which can be identified using heart sounds. Both heart sound information and symptoms are used in disease prediction. First audio inputs at four clinically important locations on the chest are acquired using an electronic stethoscope and entered to a database with symptoms for each patient. After de-noising, prominent features and statistical parameters needed for disease detection are extracted from the heart sound samples using several algorithms. Then the disease classification is performed to find out possible disease and murmur types. The software tool reported in this paper is capable of identifying normal heart sounds and abnormal heart …

The natural sciences have been transformed with the incorporation of advanced technologies, and the current technological wave of change is also revolutionizing biological science. With the wide use of herbal medicines in traditional medical systems, the demand for medicinal plants has increased enormously. Though the current trend has made the use of medicinal plants more popular, we still require better methods to distinguish between medicinal plants and plants which do not possess medicinal value. Manual methods of plant recognition rely on a plants flower information. However, these manual processes of flower recognition are not straightforward for lay persons unless expert guidance is provided. Motivated by this reasoning, an intelligent flower analyzing system has been developed to recognize medicinal plants. Various tests have been performed on 160 images taken from four types of flowers, to recognize flowers based on their colour and shape features. The experiment was carried out using Support Vector Machine (SVM) and Principal Component Analysis (PCA) methodologies respectively for colour and shape extraction. Acquiring an average of 65% recognition rate implies that the applicability PCA and SVM in the specified domain is a valuable step forward.

We present a novel method where smart phone sensors are used to sense the prevailing traffic situation and to share extracted information among a community. Instead of using the power-hungry GPS receiver in the mobile device for detecting congestion through vehicular motion, we use the accelerometer along with a novel algorithm. We show that traffic congestion situations can be identified with 78% accuracy with our technique. Vehicular traffic has become a severe problem, particularly in the less developed countries. Therefore pre awareness of congestion situations is very important for an efficient lifestyle.

Human activity recognition finds many applications in areas such as surveillance, and sports. Such a system classifies a spatio-temporal feature descriptor of a human figure in a video, based on training examples. However many classifiers face the constraints of the long training time, and the large size of the feature vector. Our method, due to the use of an Support Vector Machine (SVM) classifier, on an existing spatio-temporal feature descriptor resolves these problems in human activity recognition. Comparison of our system with existing classifiers using two standard datasets shows that our system is much superior in terms of the computational time, and either it surpasses or is on par with the existing recognition rates. It performs on par or marginally inferior to existing systems, when the number of training examples are a few due to the imbalance, although consistently better in terms of computation time.

The short-term electricity demand forecasting has become one of the major research area in power system engineering. By combining the smart metering to the short-term demand forecasting techniques, new features can be added to save on demand and electricity bill. This paper illustrates the methodology used to forecast electricity demand over short period of time which can be used with smart meters. Polynomial fitting with interpolation is used to forecast the demand by taking the apparent power sample points from smart meters. The outcome of this work will be beneficial to the residential or industrial electricity consumers to control the demand side loads. It will help the industrial consumers to save on maximum demand charge with the introduction of warning message or residential consumers to reduce their electricity bill by cutting down non-essential loads in peak hours.

Online learning is commonly accepted as a support tool for educators as well as a medium of delivery of any-time, any-where content of a wide range of study programs to a widely dispersed learner community. Web-based learning environments are a relatively new medium of learning to Sri Lankan universities. Characteristics of learners of such environments vary widely, from technology geniuses to technological novices, from high bandwidth access to slow Internet connections etc. There are predictions of a near future boom of digital learning in Asian educational context, challenging the conventional face to face learning environments. Like any learning process, digital learning depends on effective communication of human knowledge, whether this occurs in a face-to-face classroom or across the Internet. The effectiveness of digital learning also depends on establishing two-way communication between facilitators and learners, and among learners themselves. Within the state university settings, learners experience varied emotions and interest towards learning. Although both emotions and interest can increase learners’ likelihood to engage in traditional learning, little is known about the influence of emotions and interest in learning activities in a digital environment. Previous research have identified that emotions occur while individuals assess events in their environment that are related to the needs, goals and well-being. Moreover, recent research on the emotional response to online learning has focused on the importance of learners’ feelings in relation to the community of learning. Major objective of this research study is to introduce a …

This paper presents a methodology to diagnose some heart diseases which can be identified using heart sounds. As the first stage, audio files in . wav format are acquired using an electronic stethoscope and entered to a database with patient symptoms. Prominent features and statistical parameters needed for disease detection are extracted from the heart sound samples using some algorithm. Patient's heart sound fde and symptoms are given as inputs to the developed software for analysis to find out the diseases

This paper proposes the use of an intrusion detection system (IDS) tailored to counter the threats to an IEC61850-automated substation based upon simulated attacks on intelligent electronic devices (IEDs). Intrusion detection (ID) is the process of detecting a malicious attacker. It is an effective and mature security mechanism. However, it is not harnessed when securing IEC61850-automated substations. The IDS of this paper is developed by using data collected by launching simulated attacks on IEDs and launching packet sniffing attacks using forged address resolution protocol (ARP) packets. The detection capability of the system is then tested by simulating attacks and through genuine user activity. A new method for evaluating the temporal risk of an intrusion for an electric substation based upon the statistical analysis of known attacks is also proposed.

In elephant management and conservation, it is vital to have non-invasive methods to track elephants. Image based recognition is a non-invasive mechanism for elephant tracking, albeit the inefficiency in manual method due to difficulties in handling large amount of data from multiple sources. To mitigate the drawbacks in manual method, we have proposed a computer vision based, automated, elephant recognition mechanism, which mainly relies on appearance based recognition algorithms. We have tested feasibility of the system running on a web based interface, which can facilitate researchers and conservationists all around the world to actively participate in elephant conservation.

Network traffic has been shown on numerous occasions to be self similar under normal conditions. This self similar property is however, lost during anomalous conditions such as device failure, congestion and malicious intrusions. Therefore, this loss of self similarity can be used to detect such events. The Hurst parameter (H) is the most widely accepted parameter for determining self similarity. However, an accurate estimate is data and computationally expensive. This paper discusses the potential of using efficient classifier and soft computing based approaches for determining self similarity. Traffic data is obtained for various user activities from genuine browsing to malicious attacks. This data is then analysed for self similarity. The logarithmic normalized histogram of the packet interarrival time is used to obtain a feature set for classification. Various techniques are used to analyse and reduce the feature set …

The bag of visual words model has seen immense success in addressing the problem of image classification. Algorithms using this model generate the codebook of visual words by vector quantizing the features (such as SIFT) of the images to be classified. However, a codebook so formed tends to get biased by the nature of the dataset. In this paper we propose an alternative method to create the codebook for the dataset of images to be classified. Instead of directly using the dataset itself we first create a visual word dictionary by studying the SIFT features of a universal set of images. The codebook for the images to be classified is derived from this dictionary. To assess the effectiveness of the codebook thus derived, we classify the images using Probabilistic Latent Semantic Analysis in an unsupervised setting and Naive Bayes' classification in a supervised setting. The use of a dictionary achieves results …

This paper is about the development of a framework for the formal certification of software for masquerade detection based upon natural mimicry. A masquerade is when an intruder mimics a genuine user in order to compromise the security of a computer system. The framework consists of looking at mimicry in the natural world. Based on this, a model for mimicry is developed based on the principle of indiscernibility. This is followed by a look into the requirements for masquerade detection software. A formal model for an agent for masquerade detection is then derived which discerns the static attributes of the user as well as detects anomalies in the dynamic user behavior via a statistical classifier. This agent is then used for masquerade detection in the Human Machine Interface (HMI) of an industrial automation network.

Industrial automation is undergoing an increased use of wireless networks due to high flexibility and ease of deployment. However, despite the benefits, wireless networks have their inherent problems and vulnerabilities. This paper investigates the feasibility of using anomaly detection using possibility theory for network traffic. This is then used as a lightweight hostbased intrusion detection system for single board computer or embedded devices of an IEEE 802.11 based wireless industrial automation network. Traffic data is collected for genuine browsing and simulated attacks. It is then subjected to cluster analysis and tested using standard classifiers. The logarithmic histogram of the interpacket delay is used as the feature for classification. Subsequently it is used for training and testing a possiblisitic anomaly detector. The performance is then compared with a statistical outlier detector.

This paper proposes a new biologically inspired algorithm for optimization. The algorithm, called the Paddy Field Algorithm (PFA) operates by initially scattering seeds at random in the parameter space. The number of seeds of each plant depend on the function value such that a plant closer to the optimum solution produces the most seeds. Out of these, depending on the number of neighbors of the plant, only a percentage will become viable due to pollination. In order to prevent getting stuck in local minima, the seeds of each plant are dispersed. This algorithm is tested on four sample functions along side other conventional algorithms. The effect of various parameters on the performance of the algorithm is also investigated. Its performance is also tested with a hybrid algorithm. The results show that the algorithm performs well.

This paper details the use of possibilistic decision trees for a lightweight Intrusion Detection System (IDS) to be used in Intelligent Electronic Devices (IEDs) of IEC61850 automated electric substations. Traffic data is captured by performing simulated attacks on IEDs. Data is obtained for two types of genuine user activity and two types of common malicious attacks on IEDs. The genuine user activity includes, casual browsing of IED data and downloading of IED data while a Ping flood Denial of Service (DoS) and password crack attack are performed for malicious attacks. Classification is done using possibilistic decision trees for the logarithmic histogram of the time difference between the arrival of two consecutive packets. The main contribution of this paper is the use of non-specificity for obtaining a continuous valued possibilistic decision tree and its cut points. It also includes the use of mean distance metrics to …

Safety of critically ill patients in intensive care units is an important aspect of medical care. There are many factors contributing to shortcomings and errors in patient care in the intensive care setting, such as long working hours, high levels of stress, lack of enough people, may cause human errors and affecting the effectiveness of the decisions of the physician. Several attempts have been made to increase the effectiveness of such decisions by issuing early alerts on adverse patient conditions. However, such alerts are based on single parameter variations but not on the relationship between multiple parameter variations. Thus, inability to provide an effective communication model causes a considerable bottleneck in intensive care unit (ICU) operations. The proposed model is an integrated solution which identifies the adverse patient conditions on multiple parameter variations and then provides predictive treatment …

The safety of critically ill patients in intensive care units is an important aspect of medical care. Many human factors contribute to deficiencies and errors in patient care in the intensive care setting, such as long working hours, high levels of stress, lack of enough people, may cause human errors and affecting the effectiveness of the decisions of the physician. Several attempts have been made to increase the effectiveness of such decisions by issuing early alerts on adverse patient conditions. However, such alerts are based on single parameter variations, and not on the relationship between multiple parameter variations. We developed a computer-based model is an integrated solution which identifies adverse patient events based on multiple parameter variations, and then provides predictive treatment suggestions based on the likely clinical conditions which result in the parameter variations. The proposed system follows an interactive communication cycle in order to properly notify the responsible treating physicians at different tiers of responsibility. Our model is capable of early identification of adverse conditions and providing suitable treatment suggestions, thus acting as a decision support system to assist the treating physician.

This paper details the use of decision fusion from the outputs of different types of classifiers used for analyzing network traffic to detect masquerading within an IEC61850 automated substation. Data on network traffic is collected under simulated scenarios of a genuine user casually browsing data and an attacker rapidly downloading restricted data. The logarithm of the time difference between two successive packets is calculated and the histogram of this is used for classification using support vector machines and nearest neighbor classifiers. Decision fusion of the outputs of the best classifiers is done using evidence theory and possibility theory in order to reduce the number of input data and increase the reliability.

The vision of this research paper is that the mobile phone is aware of its user's motion state and surroundings and modifies its behaviour especially the characteristics of Location-Based-Services based on this information. In the research it is implemented, and evaluated a methodology which can identify individual user, states. This learning is expected to occur online and not expected the requirement of any external supervision. The proposed system relies on hidden markov modelling and Log Likelihood Method. The implementation of the methodology is performed by first training the hidden markov model for the required number of states by an intended network trace. The log likelihood value of the data for each hidden markov model in the set is computed and the motion state is identified by choosing the hidden markov model that produced the highest value. The results of simulations carried out in WCDMA environment indicate that the proposed method is able to assist to create a meaningful user context model at various propagation conditions defined by both 3rd Generation Partnership Project (3GPP) and Wireless World Initiative New Radio (WINNER) propagation scenarios while only requiring a network trace without having an integrated sensor onboard cellular phone or any other wearable sensor device.

A novel method of individual level adaptive mutation rate control called the rank-scaled mutation rate for genetic algorithms is introduced. The rank-scaled mutation rate controlled genetic algorithm varies the mutation parameters based on the rank of each individual within the population. Thereby the distribution of the fitness of the papulation is taken into consideration in forming the new mutation rates. The best fit mutate at the lowest rate and the least fit mutate at the highest rate. The complexity of the algorithm is of the order of an individual adaptation scheme and is lower than that of a self-adaptation scheme. The proposed algorithm is tested on two common problems, namely, numerical optimization of a function and the traveling salesman problem. The results show that the proposed algorithm outperforms both the fixed and deterministic mutation rate schemes. It is best suited for problems with several local optimum solutions without a high demand for excessive mutation rates.

A data-driven adaptive neurofuzzy controller is presented for the water-level control of U-tube steam generators in nuclear power plants. This neurofuzzy controller is capable of learning the control action principles from the data obtained using other methods of automatic or manual control. There are four inputs in the neurofuzzy system, yet only eighty fuzzy rules involved. Therefore, the fuzzy system is versatile and moderately compact. The versatility is due to the higher input space dimension that helps to learn more control principles. The compactness is due to the number of rules being not too many. A 10-h evaluation trial of the trained fuzzy controller demonstrated its capability in regulating the water level under random disturbances and reference level changes.

This Paper describes the implementation of a verification system that uses palm geometry to verify the identity of an individual. Palm geometry is a biometric scheme that uses the shape of the hand to confirm the identity of an individual.