Renee Ka Yin Chin

Particle Swarm Optimization Based Resource Allocation in Orthogonal Frequency-division Multiplexing

Current cellular wireless data communications require wide availability and satisfying quality of service to keep up with the rising number of wireless devices. It is challenging for the service provider to cater to different kinds of upcoming new services and applications as data rates vary apart in that respect. For that reason, it is of paramount importance to adaptively utilize the radio resources which are limited and valuable. The ability of Orthogonal Frequency-division Multiplexing Access (OFDMA) to split the entire bandwidth into smaller sub-carriers allows adaptive resource allocation to be performed in downlink communications. However, due to the time-varying nature of the wireless channel, not all links experience the same channel response. Therefore, this situation is akin to a strategic combination problem to find the best allocation out of the huge combinations according the current channel characteristics. This paper presents an evolutionary method, particle swarm optimization (PSO), to optimally allocate the rates and sub carriers to multiple users. The results display good convergences towards the optimal solution for a minimal transmit power allocation. At transmission intervals, the cellular base station will be able to use the best allocation relating to channel conditions of any point in time.

Trajectory Clustering for Behavioral Pattern Learning in Transportation Surveillance

The development of an efficient traffic flow monitoring system has been the main focus for many researchers working in the field. Due to the rapid development in urbanization, the complexity of traffic intersections provides challenges for researchers to detect the underlying traffic scenes. With the emerging video based surveillance system, vehicle trajectory can be extracted for observation and prediction via behavioral pattern learning. Prior to the learning, clustering of the extracted vehicle trajectory data is performed to group the data based on similarity measures. In this paper, the implementation of clustering algorithm on the trajectory data is analyzed and issues concerning the trajectory clustering are discussed.

Optimization of urban traffic network signalization using genetic algorithm

This work aims to minimize average delay for an urban signalized intersection under oversaturated condition using genetic algorithm (GA). Relieving urban traffic congestion is an urgent call for traffic engineering. The effectiveness of traffic signalization is one of the key solutions to reduce congestion, but regrettably the current traffic signal control system is not fully optimized for handling oversaturated condition. Therefore, this work proposes GA to optimize traffic signals for reducing average delay at a signalized crossed intersection under oversaturated condition. A comprehensive traffic model based on Public Works Department, Malaysia has been developed as the platform. The average delay experienced by vehicles to traverse the crossed intersection is used as the performance metric to evaluate performances of the proposed algorithm. Simulation results show GA is able to control the traffic signals for minimizing the average delay to 55 sec/veh or equivalent to level of service (LOS) D.

Progressing toward urban topology and mobility trace for Vehicular Ad Hoc Network (VANET)

Mobile communication and intelligent transportation system (ITS) show a promising technology as it could enhance the traffic efficiency and promote on-road traffic safety. Vehicle Ad Hoc Network (VANET) is one of a notable example of successful wireless networking paradigm emerges from the combination of mobile communication and ITS. It allows wireless communication between vehicles and remote application. Many VANET related researches have been done in simulation rather than in real-world due to its inexpensive, repeatable and easily controlled experiment. Simulation of VANET can be divided into two types: network and mobility simulation. In order to develop a vehicular network service, a realistic mobility environment must be generated beforehand as it could greatly affect the performances of developed VANET protocol. In this paper, traffic simulator (SUMO) is used to generate realistic traffic of the city Kota Kinabalu, Sabah, Malaysia. Different scenarios are generated to evaluate the performance of VANET on such realistic developed topologies.

Clustering Algorithm in Normalised Cuts Based Image Segmentation

Normalised cut method has been effectively used for image segmentation by representing an image as weighted graph in global view. It does segmentation via partitioning the graphs into sub-graphs. Clustering algorithm is implemented such that sub-graphs with common similarities are grouped together into one cluster and separates sub-graphs that are dissimilar into distinctive clusters. Clustered segments from the normalised cuts are then produced. As the clusters initialisation gives influence to the segmentation result, optimisation of the clustering algorithm is implemented to achieve better segmentation. With the approach applied in the normalised cuts based image segmentation, the constraint of using normalised cuts algorithm in image segmentation can be alleviated. In this paper, evaluation of the clustering algorithm with the normalised cuts image segmentation on images has been carried out and the effect of different image complexity towards normalised cuts segmentation process is presented.

A framework for integrating project-based learning into the curriculum for outcome based education

Project-based learning (PBL) has been shown to be an effective tool in engineering education. Thus, in this paper a framework for integrating PBL into the curriculum to enable the delivery and the assessment of the necessary knowledge, skills, and attitudes required by three generic program outcomes required for accreditation of engineering programs in countries that signatories to the Washington Accord. The breadth and depth required by these outcomes and how it is implemented in this framework is also highlighted. It is believed that this framework is general enough to be implemented in any engineering program.

Model cart control using EMG signal

This paper describes the development of EMG signal acquisition system application for model cart control. The main application of this type of control is to assist disabled people or frail people to control a cart or device via their healthy muscle. In this project, the circuit was simulated, developed and troubleshoots. The aim of model cart control was to control four basic directions which is forward, backward, turn left and turn right. Threshold voltage level comparison method is used in this project for model cart control. By considering the user friendly issue for the application system, EMG signal acquisition method was decided to be used and obtained with surface electrode. There are two stages to be done for this system, in which the first stage is the signal acquisition system and the second stage is to control the action of model cart. First stage included pre-amplifier, rectification and filtering, which were done with instrumentation amplifier, active low pass and high pass filter and precision full wave rectifier. Second stage included the microcontroller and motor driver in which the microcontroller functions to read the two channel EMG signal and give signal for the motor driver to control the DC geared motor. The result of the hardware testing is compared with the software simulation where the same output is obtained. There were also some limitation and problem which have not been solved. There was wire connected to the model cart where it is not very convenient for the model cart to move around or further.

Performance Analysis of Enhanced Genetic Algorithm Based Network Coding in Wireless Networks

Network coding combines several packets from different sources and broadcasts the combined packet to several destinations in single transmission time slot. Each destination is capable to extract the intended information by decoding from a common packet. In short, network coding improves the throughput for wireless and wired networks but also causes side effects such as complexity of packets management and increased delay for coding opportunity. Hence, genetic algorithm is used to optimize the resources for network coding. Genetic algorithm will search for suitable routes to the destination according to the desired throughput with a desired multicast rate. In this paper, genetic algorithm is further enhanced in searching of optimum route for a packet. The simulation results show that enhanced genetic algorithm can adapt to various situations with different topologies, providing an optimum route for a packet in the network.

Characterization Study of Using Acoustic Signals for Underwater Non-Destructive Testing

Non-destructive testing in structural inspection is a list of techniques that is used to evaluate the condition of a structure in order to determine whether cracks, flaws defects or damages occur in structural build without causing any damage to the test object. This work presents the characterization study of using acoustic signals for underwater non-destructive testing. The optimum operating frequency for sensors, transmission angle between the transmitter and receiver and operating distance between the sensors with reflecting medium are determined. Early results can be implied in exploring the possibility of using acoustic signals in underwater structural inspection with non-destructive testing methods.

Vehicle Tracking Using Particle Filter for Parking Management System

Increment of on-road vehicles has urged public venues to provide visitors with a larger area of parking space. As the parking area grew larger for example in a hyper mall, a well-organized parking management system is necessary to assist drivers in locating parking position. Besides, it can also help the management team to monitor vehicle flow in the parking lot. Vehicle tracking plays an important role to the parking management system, as accurate tracking result will lead to a more efficient management system. Among commercially available sensors, video sensor has been commonly deployed in the parking area due to its ability in obtaining a wide range of vehicle information. However, images captured using video sensors are limited under situations where vehicles are undergoing occlusion and maneuvering incidents. This will cause tracking error therefore affecting the performance of the parking management system. Particle filter has been proven as one of the promising techniques to track vehicle under disturbances. Therefore, particle filter is proposed to track vehicle under occlusion and maneuvering incidents in this study. Experimental results show that the particle filter is able to track a target vehicle under different disturbances.