Jung-Chun Liu

Genetic Algorithm for Shortest Driving Time in Intelligent Transportation Systems

The route guidance system, which provides driving advice based on traffic information about an origin and a destination, has become very popular along with the advancement of handheld devices and the global position system. Since the accuracy and efficiency of route guidance depend on the accuracy of the traffic conditions, the route guidance system needs to include more variables in calculation, such as real time traffic flows and allowable vehicle speeds. As variables considered by the route guidance system increase, the cost to compute multiplies. As handheld devices have limited resources, it is not feasible to use them to compute the exact optimal solutions by some well- known algorithm, such as the Dijkstras algorithm, which is usually used to find the shortest path with a map of reasonable numbers of vertices. To solve this problem, we propose to use the genetic algorithm to alleviate the rising computational cost. We use the genetic algorithm to find the shortest time in driving with diverse scenarios of real traffic conditions and varying vehicle speeds. The effectiveness of the genetic algorithm is clearly demonstrated when applied on a real map of modern city with very large vertex numbers.

Construction and Application of an Intelligent Air Quality Monitoring System for Healthcare Environment

Indoor air quality monitoring in healthcare environment has become a critical part of hospital management and policy. Manual air sampling and analysis are cost-inhibitive and do not provide real-time air quality data and response measures. In this month-long study over 14 sampling locations in a public hospital in Taiwan, we observed a positive correlation between CO2 concentration and population, total bacteria, and particulate matter concentrations, thus monitoring CO2 concentration as a general indicator for air quality could be a viable option. Consequently, an intelligent environmental monitoring system consisting of a CO2/temperature/humidity sensor, a digital plug, and a ZigBee Router and Coordinator was developed and tested. The system also included a backend server that received and analyzed data, as well as activating ventilation and air purifiers when CO2 concentration exceeded a pre-set value. Alert messages can also be delivered to offsite users through mobile devices.

Anomaly Detection Using LibSVM Training Tools

Intrusion detection is the means to identify the intrusive behaviors and provides useful information to intruded systems to respond fast and to avoid or reduce damages. In recent years, learning machine technology is often used as a detection method in anomaly detection. In this research, we use support vector machine as a learning method for anomaly detection, and use LibSVM as the support vector machine tool. By using this tool, we get rid of numerous and complex operation and do not have to use external tools for finding parameters as need by using other algorithms such as the genetic algorithm. Experimental results show that high average detection rates and low average false positive rates in anomaly detection are achieved by our proposed approach.

Energy Analysis of Multimedia Video Decoding on Mobile Handheld Devices

Video applications on mobile handheld devices are very challenging mainly due to the complex computational needs of video decoding and the limited capacity of batteries of handheld devices. The increasing number of codecs makes the situation even more perplexing for users trying to find the optimal video codes for playing on handheld devices. In this paper, we try to find the optimal video codecs from energy analyses of them on handheld devices. Various experiments have been done to investigate the effects from choices of codecs, file formats, and codec parameters.

A method for managing green power of a virtual machine cluster in cloud

Abstract A green power management scheme is proposed to determine how many physical machines should be run or turned off based on the gross occupied resource weight ratio of the virtual machine cluster. The gross occupied resource weight ratio is defined as the ratio of the sum of resource weights of all virtual machines over the sum of available resource weights of all running physical machines. When the gross occupied resource weight ratio is greater than the maximum tolerant occupied resource weight ratio, preset to ensure quality of service, a standby physical machine in the non-running physical machines is selected and wakened up to join as one of the running physical machines. On the other hand, when the gross occupied resource weight ratio is less than the minimum critical occupied resource weight ratio, preset to trigger energy saving algorithms, one of the running physical machines, selected as a migration physical machine with the virtual machines therein removed after live migration, is moved from other running physical machines, and then turned off. As a result, a resource allocation process is realized to distribute loads of the running physical machines such that the total number of the running physical machines can be flexibly dispatched to achieve the objective of green power management.

Implementation of Smart Power Management and Service System on Cloud Computing

This paper plays a role for electricity providers, who are actively developing digital meters and power management systems, and develops an energy management system by using service-oriented architecture. In addition to a power management system, the policy has changed from the traditional way in Taiwan by using energy-saving appliances. Therefore, the solution is to design devices that are more intelligent and energy saving at the same time. Moreover, computers can control the mechanism of the smart devices. The system uses a service-oriented architecture approach to construct services and utilize common virtualization technologies on cloud infrastructure. This work also applies distributed storage technologies to store and distribute data in order to make sure the system can meet massive loads of demands for services. Finally, we not only show provided services and hardware control in the implementation but also conduct performance tests of the distributed database.

Using Adaptive Bandwidth Allocation Approach to Defend DDoS Attacks

Denial of service attacks occur when the attacks are from a single host, whereas distributed denial of service attacks occur when multiple affected systems flood the bandwidth or resources of a targeted system. Although it is not possible to exempt entirely from denial of service or distributed denial of service attacks, we can limit the malicious user by controlling the traffic flow. In the paper, we propose to monitor the traffic pattern in order to alleviate distributed denial of service attacks. A bandwidth allocation policy will be adopted to assign normal users to a high priority queue and suspected attackers to a low priority queue. Simulations conducted in network simulator of our proposed priority queue-based scheme shows its effectiveness in blocking attack traffic while maintaining constant flows for legitimate traffic.

On improvement of cloud virtual machine availability with virtualization fault tolerance mechanism

Virtualization, particularly in the field of cloud computing, is a common strategy to improve existing computing resources. Hadoop, one of the Apache projects, is designed to scale up from single servers to thousands of machines, each offering local computation and storage capabilities. However, how to guarantee both stability and reliability of virtualization have become important topics. In this article, to reach this goal we used current open-source software and platforms, for instance, the Xen-Hypervisor virtualization technology, and the OpenNebula virtual machines management tool. After extending components capabilities, we developed a mechanism to support our ideas and reached high availability with Hadoop that is also called as virtualization fault tolerance (VFT). We considered a practical problem, i.e., the single-point-of-failure issue that occurs frequently in virtualization systems, and the experimental results confirm that the downtime interval can be greatly shortened even if failure occurred. As a result, VFT is useful not only for Hadoop applications, but also for more areas in cluster-based systems.

Energy analysis of multimedia video decoding on mobile handheld devices

Video applications on mobile handheld devices are very challenging mainly due to the complex computational needs of video decoding and the limited capacity of batteries of handheld devices. The increasing number of codecs makes the situation even more perplexing for users trying to find the optimal video codes for playing on handheld devices. In this paper, we try to find the optimal video codecs from energy analyses of them on handheld devices. Various experiments have been done to investigate the effects from choices of codecs, file formats, and codec parameters.

A Robust Full-Band Image Watermarking Scheme

Distributed discrete wavelet transformation (DDWT) method was proposed in 2006 for copyright protection. In this paper, we propose a watermarking scheme for digital image based on the DDWT method with some modifications to improve its robustness. Our scheme transforms original image data from the spatial domain into the frequency domain by using the DDWT technique, and then embeds watermark information in the four sub-bands in the frequency domain. In this way, we not only keep intact the salient feature of the DDWT method, which distributes hidden watermark information in the spatial coefficients and is robust against the cropping attack, but also greatly improve its performance so that the new scheme is robust against geometric attacks such as rotation or scaling and non-geometric attacks such as Gaussian noise, sharpening, and contrast adjustment, too. Our experiment results show that the stego-image is superior and the watermark is robust against a variety of geometric and non-geometric attacks mentioned above