Skip to main content

Über dieses Buch

This book constitutes the refereed proceedings of the 5th International Conference on Convergence and Hybrid Information Technology, ICHIT 2011, held in Daejeon, Korea, in September 2011. The 94 revised full papers were carefully selected from 323 initial submissions. The papers are organized in topical sections on communications and networking, intelligent systems and applications, sensor network and cloud systems, information retrieval and scheduling, hardware and software engineering, security systems, robotics and RFID Systems, pattern recognition, image processing and clustering, data mining, as well as human computer interaction.



Communications and Networking

An Efficient Mobile Peer to Peer Architecture in Wireless Ad Hoc Network

With the rapid development of wireless communication technologies and mobile devices, the mobile peer to peer (MP2P) network has been emerged. Since the computation ability, bandwidth, storage and the dynamic network topology of mobile devices are limited, directly applying the existing wired P2P method to wireless ad hoc network will result in poor performance. A hierarchical MP2P architecture based on super peers has been proposed to alleviate the problem. However, the management cost of the mobile network topology is still high. In this paper, we propose a hierarchical MP2P architecture using clustering mobile peers. The proposed architecture considers the maximum connection time of connected peers, the hop count with the super peer, and the number of connected peers of the super peer. The connection time between the connected peers can be determined by the location, velocity and communication range of each mobile peer. Therefore, the network topology is relatively stable, the management cost of the network topology is decreased and the success rate of contents search is increased. Experiments have shown that our proposed scheme outperforms the existing schemes.

He Li, Kyoungsoo Bok, Yonghun Park, Jaesoo Yoo

Performance Evaluation of MINTRoute Protocol at Different Scenarios

Due to natural weakness of sensor node in the point of wireless communications, reliable data delivery becomes one of the major challenges in wireless sensor networks. To handle this problem, many schemes in each layer have been proposed in the several literatures. Among those, MintRoute protocol is designed to achieve high reliability by choosing the best qualified path which is mainly measured by the delivery ratio of packets. Moreover, although some literatures provide the simulation and experimental results of performance evaluation, as far as the authors know, there is no previous research works to mention impact on performance by sensor deployment strategies, effect of failure, and the density of nodes. This paper details a simulation study that compare performance of MintRoute at different scenarios. In addition, research challenges will be given through analysis and discussion for performance.

Ki-Il Kim, Min-Jung Baek

Modified Packet Scheduling Algorithm for Real-Time Service

In this paper analyzes various scheduling algorithms applied to multi-media streaming system for providing real-time service in Internet and proposed a scheduling algorithm that can meet the QoS requirements of delay-sensitive traffic and loss-sensitive traffic for specific services. First, the concept of QoS drafted from the recommendation of ITU-T is defined and the requirements for packet multiplexer in Internet are examined. Then the scheduling algorithm to satisfy various QoS based on functional architecture of packet multiplexer for streaming system bound to real-time service and QoS levels of input source traffic is designed. Finally, the performance of the algorithm is evaluated thru experiments that measure the packet loss and average delay time of the proposed algorithm.

Ha-Sung Koo

Virtual Research Domains Management on L2 Optical Private Network

Recent advances inoptical networks lead the advent of service sensitive applications andthe community researches free from the individual research. Sensitive applications and community researches overOptical Private Network (OPN)is required to supportefficient network resource management, intelligent network switching, and sophisticated access control and monitoring. In this paper, we introduce L2 OPN architecture and describe our experience in implementing L2 OPN. Lastly, we address the function and framework ofVirtual Research Domains Management over L2 OPN.

Min-Ki Noh, Byung-Yeon Park, Dong-Kyun Kim, Joon-Min Gil

Charging Algorithm for Quality-Based VoIP Charging Mechanism

Recently, many companies are moving from PSTN to packet-switched networks as it provides easier access to voice communication anywhere in the world. VoIP Quality is one of the important problems service providers want to solve as providing QoS guarantees in VoIP applications becomes more challenging in wireless and mobile networks. In this paper, we used a methodology of Listening and Rate-Opinion Trial (LRT) in order to define the burst boundary, which enables users to identify if they are satisfied with their experiences in a short period of time. We derived the charging rate algorithm that can be used as a reference for calculating the “delivered packet amount”-based prices by observing the connection between PESQ MOS and the gap ratio for sex samples of both English and Korean. The connection of PESQ MOS and the gap ratio could be obtained by measuring the burst. The charging rate algorithm can be derived from the measured gap ratio.

Al Ann Ibanez, Younggin Hong, Younchan Jung

A Study on Dynamic Gateway System for MOST GATEWAY Scheduling Algorithm in MOST25 and MOST150 Networks

In recent years, the driver needs the convenience of their vehicles and that there is an increasing requirement. Many researches have been mainly focused on MOST Networks to provide quality of multimedia service. The MOST network has different protocols, such as MOST25, MOST50, and MOST150, requiring gateways to send and receive information. In the case of isochronous and Ethernet channel data that can be supported only with MOST150, when data is transmitted from MOST150 to MOST25, it cannot be treated by MOST25, leading to data loss and transmission delays. Thus, this study proposes a MOST gateway system to connect the MOST25 and MOST150 networks to form a single network. It aims to use the simulation tool NS-2, to analyze the FIFO, FQ, WRR, CBQ scheduling algorithm performance between the MOST25 and MOST150 networks, and to consequently propose scheduling algorithm improvement measures suitable to cars.

Seong-Jin Jang, Jong-Wook Jang

Study on Asynchronous Data Transmission of MediaLB Bus Communication

As demands increase for the Infotainment System, the Multimedia Networking technology for automobile, called MOST(Media Oriented Systems Transport), has been actively applied to the automobile industry, to meet the demands of the Infotainment System. MOST is define the protocol, H/W and S/W layers to support the transportation of synchronous, asynchronous and control data in an efficient and low cost way. The existing MOST devices relied on 12C, 12S communication mode for internal bus communication but the widening bandwidth in the MOST network and additional installations in a single device required bus communication and MediaLB(Media Local bus). Majority of the existing MOST devices use I2S or I2C bus communication mode to process synchronous data (stream data) and asynchronous data (packet data). Hence, this study will take a look at control channel and transmission mechanism of synchronous data and asynchronous data by applying MediaLB mode, which is an advance from 12S or 12C bus communication mode proposed by MOST network and currently in use, suggest an efficient way to transmit asynchronous data and analyze its performance.

Jong-Wook Jang, Chang-Young Kim, Gil-Bong Kwak, Yun-Sik Yu

TCP Throughput Improvement over Vertical Handover between 3G LTE and WLAN

We address TCP throughput degradation over vertical handover between 3G-LTE and WLAN. 3G-LTE link layer tries to recover packet loss and deliver recovered packets in-order. This may involve link layer holding packets before later arriving packets arrive, causing spurious timeout. A state-of-the-art scheme holds fast retransmit/recovery for one RTT waiting for packets arriving at TCP layer out of order. We show this wait is unnecessary and costs TCP throughput. Our scheme is also able to improve TCP throughput by loss differentiation with parameters adapted to new network on vertical handover, considering link layer characteristics of 3G LTE which supports in-order delivery. Simulation result shows improved TCP throughput under various scenarios with TCP Reno and TCP SACK.

Sujin Park, Ju-Wook Jang

An Adaptive QoS Control Scheme Based on Traffic Estimation for Efficient Multimedia Services in Wireless Access Networks

In this paper, an enhanced bandwidth allocation scheme according to traffic estimation for adaptive QoS control is proposed with the use of the IEEE 802.11e standard in the wireless access network. The interface between the core network and the access network includes not merely a translation of the QoS requirements from one network to the other but also a means of provisioning the differentiated Service (DS) considering the requirements of a particular flow. Simulation results demonstrate the effectiveness of the proposed QoS control scheme based on traffic estimation for the aspect of access delay.

Moonsik Kang, Kilhung Lee

Utilizing Intelligent Middleware for Reconfiguration of Applications on Android

The recent technological advances in mobile communication and handheld devices transform the existing mobile phones to powerful computers with variety of functionalities and capabilities. This evolution has led to development of advanced applications such as maps and navigation, social networking, enhanced games, and many more which can be utilized by thesephones. However, these smartphoneshave several limitations such as, limited energy, memory and processing power. These constraints limit the ability of thesephones for performing complex tasks, or computationally intensive tasks at a constant level of performance for thesephonesto match the users’ desires. Therefore, in this paper we develop an intelligent middleware to enhance the capabilities of these handheld devices. This middleware allows the device to switch applications from executing locally to executing remotely.

S. Kami Makki, Narasimha Bharadwaj Srirangam, Venkata Susheel Aiswarya, Shui Yu

DTR-Filter: An Efficient Transmission Scheme for Real-Time Monitoring in Wireless Bulky Sensor Networks

Wireless sensor networks basically assume the low duty-cycle data to be generated. But as the area of applications become wide, bulky sensor data might be generated and need to be transmitted. The sensor devices have limited resources in CPU, battery and memory capacity. As well as data transmission and reception are major factor of energy consumption in sensor nodes, network bandwidth may be easily overrun with bulk data transmission. So the reduction of data transmission is very important in bulky sensor data transmission. We proposed a data filtering algorithm, DTR-filter, to suppress the trivial value changes in transmission while encouraging the transmission of important event values. Our DTR-filter algorithm adaptively controls the filtering criteria. DTR-filter showed 53% of average reduction in data transmission while having 86% of correctness with original unfiltered data.

Yanli Qian, Jungmuk Kang, Jookyoung Kim, Jooki Min, Youngmi Kwon

Multi-Channel MAC Protocol to Improve Network Throughput and End-to-End Delay in OFDMA-Based Wireless Network Systems

This paper proposes multi-channel MAC protocol for OFDMA Wireless LAN. This MAC protocol consists of one control channel and multiple data channels. Each data channel has multiple data sub channels and one control sub channel. Each station transmits channel negotiation control frame via control channel and data frame via data channels. Our method lets a station get through more than one communication sessions simultaneously. The basic idea is to use one sub-channel in data channel as control sub-channel. With this control sub-channel, even in the midst of one communication session, a node can set up another communication session in separate use. As a result, this mechanism improves end-to-end delay and overall network throughput.

Jookyoung Kim, Joonki Min, Youngmi Kwon

DSRC-Based Channel Allocation Algorithm for Emergency Message Dissemination in VANETs

VANET (Vehicular Ad-hoc Network) is divided into V2V (Vehicle-to-Vehicle) communication and V2I (Vehicle to Infrastructure) communication. V2V requires no infrastructure or roadside devices and vehicles communicate with each other using wireless inter-vehicle communication. V2I requires some infrastructure such as RSUs (Road Side Units). OBUs (On Board Units) installed in vehicles can access to backbone networks by using RSUs. Unlike MANET (Mobile Ad-hoc Network), VANET requires a mechanism to accommodate the environment that the moving speed of vehicles is very fast and the network topology changes frequently. VANET can use IEEE 1609.4 that supports multi-channel operation. The multi-channel approach of IEEE 1609.4 uses orthogonal channels to communicate between RSU and OBU. However, if emergency messages should be processed in high priority, the delay time will be increased because the multi-channel approach makes a fair share of available channels. Therefore, this paper proposes DMAE (DSRC-based Multi-Channel Allocation for Emergency Message Dissemination) algorithm to resolve this problem. DMAE allocates the highest bandwidth channel to the urgent message firstly, and guarantees QoS between RSU and OBU through periodic channel switching. Simulation results using ns-2 show performance improvement in terms of end-to-end delay and emergency message delivery rate.

Min-Woo Ryu, Si-Ho Cha, Kuk-Hyun Cho

Quadrature Amplitude Position Modulation Combined with Multicarrier Communication System

In this paper, we analyze performance of low power consumption modulation scheme such as Phase Shift Position Modulation(PSPM), Phase Silence Shift Keying(PSSK) and Quadrature Amplitude Position Modulation(QAPM) using OFDM transmission in AWGN channel. A PSPM, PSSK and QAPM modulation scheme are decrease in bandwidth efficiency but improved power efficiency than existing PSK and QAM. These modulation schemes can achieve power efficiency because every symbol of modulations has a zero-envelope period like PPM techniques. The OFDM system is high bandwidth efficiency than single carrier system. In this paper, we compare PSPM, PSSK and QAPM modulation using OFDM transmission regard to BER performance, throughput and PAPR.

Jae-Hoon Choi, Heung-Gyoon Ryu

Wireless Multimedia Acoustic Transmission with MIMO-OFDM

Acoustic communications are able to transmit multimedia data and sound simultaneously. But, there is a problem about low data throughput because of narrow frequency band of audible sound in acoustic communication system In this paper, we propose an acoustic multimedia data transmission system with multiple-input multiple-output orthogonal frequency division multiplexing (MIMO-OFDM) for high data rate in acoustic communication system. We compare the receive performance of the proposed system with the existing system with single-input single-output (SISO) by experiment. The data rates of SISO and MIMO system are 4.9 kb/s and 9.1kb/s respectively, with QPSK modulation, no channel coding, and about 3.5 KHz data frequency bandwidth. We ensure that the data throughput of the proposed system is increasing approximately 80% than SISO system. But the proposed systems need more Pilot SNR about 2dB than SISO system to satisfy the BER performance of 10e-3. In other words, we can dramatically increase data throughput of acoustic communication system with MIMO but receive performance is a little decrease.

Junyeong Bok, Heung-Gyoon Ryu

The Impact of Finite Rate Feedback on Adaptive Double Space-Time Transmit Diversity in Rayleigh Fading Channels

We investigate the impact of feedback signaling on the performance of adaptive double space-time transmit diversity (STTD) systems in wireless fading channels. We consider four transmit antennas with two space-time block encoders, where at least two receive antennas are required for detection. In adaptive double STTD systems, transmit power weights are adaptively adjusted relative to channel fading based on the feedback information. In our work, the optimum power weights are determined by a numerical search, which are functions of the number of feedback bits and the modulation size employed. We apply symmetric and asymmetric modulation methods in each space-time block encoder in cooperation with adaptive antenna weights. It is shown that the performance of adaptive double STTD with asymmetric modulations depends on the number of available feedback bits, while the same system with symmetric modulations has a relatively robust performance to variable feedback bits. Our results also show that in adaptive double STTD with asymmetric modulations, the number of feedback bits can be reduced with little performance degradations.

Ye-Hoon Lee, Dong-Ho Kim

Viterbi Algorithm Based Noncoherent Detection Scheme for Differential Space-Time Block Code

The Viterbi algorithm based noncoherent detection schemes are considered for the detection of differential space-time block code in fast fading environments. In order to overcome the channel variation due to fading, we consider exploiting short observation interval and applying the Viterbi algorithm to mitigate the performance degradation due to this short interval. Our simulation results shows that the proposed detection scheme outperforms not only multiple symbol differential detection but also coherent detection with channel estimation under fast fading conditions. In addition, we easily extend the proposed detection scheme to an iterative version by simply incorporating a priori information from the channel decoder. It is also shown that this iterative method improves the performance as the number of iterations increases.

Ye Hoon Lee, Dong Ho Kim

Transmission Power Based Source Routing Protocol for MANET with Unidirectional Links

In mobile ad-hoc networks, a communication link between two mobile computers is not always bi-directional, i.e., uni-directional, since the transmission power of mobile computers are not the same. Though some ad-hoc routing protocols, e.g., extended DSR (Dynamic Source Routing), support routing with unidirectional links, multiple flooding RREP are used and communication overhead is high. In this paper, we propose a new ad-hoc routing protocol based on DSR that supports unidirectional links in mobile ad-hoc networks. We use the neighborhood list of high transmission power node to detect the existence of unidirectional links and to transmit single unicast RREP. Our protocol achieves better performance than the extended DSR in such environments as having unidirectional links in terms of the number of required messages.

Jun-Jie Piao, Tae-Mu Chang

ERID: Edge Router Identification for Fast Forwarding Packet in BGP Domain

In this paper, serial edge router ID based fast forwarding scheme proposed to enhance the performance of high-speed backbone router for fast packet forwarding. Proposed scheme takes advantage by serializing the identity of edge routers on high speed backbone that is used for indexing the memory location of next hope information to enhance the packet forwarding speed. In this Scheme, Each edge router that connects the two Autonomous System domains is assigned serial edge router ID (ERID) by central authority and each edge router maintain the ERID list and corresponding next hop information in ERID-FIB table generated through current routing information base. When two peers establishes the session between them using current forwarding Information base, initially they exchange the ERID of edge router that immediate to it in own AS. In subsequent communication, ERID is included in packet and next hop information is determined from ERID-FIB table at memory location corresponding to ERID. Through proposed scheme, next hop information retrieve in O(1) time complexity and in single memory access time.

Gohel Bakul, Madhusudan Singh, Dhananjay Singh

The Design of Reliable Routing Protocols in Wireless Mesh Networks

Wireless Mesh Networks provide much flexibility in establishing a network compared to existing wire networks. A higher level of security is required for a wireless environment that makes the network accessible by anyone. Compared to MANET, Wireless Mesh Networks is less portable, and have less sensitivity to electric power. This enables the adoption of a higher level of security. This study takes advantage of the PKI (Public Key Infrastructure) security algorithm to optimize a Wireless Mesh Network, so that it is applicable to a Wireless Mesh Network environment. A Pre-Shared Key, which is dynamically created and utilizes the hash function, is adopted as the routing authentication method of the mesh router and for the distribution of Keys necessary for data encryption. Each router node of a Wireless Mesh Network encrypts a random number data for electronic signature. The techniques for supervision, detection and avoidance of malicious nodes between router nodes on a Wireless Mesh Network are presented. The security technique is applied in this environment.

Chang-Pyo Yoon, Jae-Wan Lim, Hyun-Jong Cha, Ho-Kyung Yang, Hwang-Bin Ryou

Smooth Device Handover System for Seamless Audio Service

Device handover based on user’s movement without relying on an infrastructure enables various services in a localized personal area. We present an audio speaker handover system based on wireless distance estimation in a Bluetooth environment. It is designed to enable smooth device handover at walking speed without any help from the infrastructure. For smooth device handover, we studied the distance estimation method which satisfies a fast reaction and high resolution. We collected received signal strength indicator (RSSI) values and transmission power level (TPL) values and determined a relative distance using the model function obtained by statistical analysis method. With these methods, we experimented and analyzed the performance of device handover. Based on these experiments, we proposed a modified RSS based distance estimation that limit the transmission power control function. We also showed that the proposed system achieves good performance in the aspects of handover delay time and handover failure rate.

Jiyeon Son, Jihyun Lee, Junhee Park, Kyongdeok Moon, Younghee Lee

Efficient Common Radio Resource Management Techniques for Cloud Systems

This paper proposes resource management method for effieicnt wirless network resource management by applying Common Radio Resource Management(CRRM), which is efficient in managing overlapped heterogeneous wireless network, to cloud computing environment. When the congestion increases, the suggested method utilizes CRRM and forces users who are connected to it to move to other mobile network to solve its own congestion.

Jae-Won Ji, Hyun-woo Cheon, Chun-Myoung Park, Geuk Lee

An Effective Method for Fairness in the Case of Missing CTS

The contention-based DCF of IEEE 802.11 adopts the 4-way handshaking mechanism for reasonable transmission with reduced packet collisions. But, expected CTS is not successfully received by the transmitter, the NAV of the transmitter’s neighbors leads to two problems: degrading channel utilization and unbalancing of fairness. We propose BN-MAC to improve the performance of WLAN using BNACK packet to reset NAV to zero. The neighbors that receive BNACK defer only SIFS for fairness of overall network. The performance of our approach is compared with the IEEE 802.11 DCF and optional DCF by simulation. The results show that our proposed approach reduces end to end delay and improves the channel utilization.

Inhye Park, Hyungkeun Lee

Improved TCP Performance during the Upward Vertical Handover Using an IEEE 802.21 MIHF Framework

In Upward Vertical Handover(UVHO) from a fast WLAN link to a slow 3G cellular link, the TCP sender may encounter precocious timeout since the RTT of the 3G link is too large for the RTO(Retransmission Time Out) which is tuned to the small RTT of the WLAN. Although IEEE 802.21 MIHF framework provides the vertical handover procedure, it does not deliver RTT of the new link since this does not belong to static information to be delivered by IEEE 802.21 MIHF. Conventional TCP such as TCP Reno, even with IEEE 802.21 MIHF, may encounter timeout after UVHO, degrading TCP throughput considerably during UVHO. In this article, we propose a modified TCP scheme in which we estimate the RTT of the new link using timeout instant, the ACK traveling through the new link just after UVHO and the old RTT. We also estimate appropriate CWND size for the new link. Simulation results show that our proposed scheme improves TCP performance during the UVHO compared with previous schemes as well as TCP Reno.

Seung Su Kim, Ju Wook Jang

Performance Evaluation of Uncompressed High-Definition Content Transmission on Parallelled Digital Subscriber Lines

High-Definition (HD) information communication is now emerged as a very interesting topic in today’s communication world. There are many ways to improve the performance of transmitting the large amount of information through the network, such as compressing the HD information, enhancing the HD encoder and decoder, or economically increasing the bandwidth of the communication channel. This paper introduces the deployment of parallel cores to increase the bandwidth of the communication channel in order to support the transmission of HD information with very less compression and decoding computational complexity. This paper also simulates the application of the paralleled Digital Subscriber Lines (DSLs) to convey the HD information, and evaluates the performance of the communication in terms of throughputs, utilization, and queue performance matrices.

Piboonlit Viriyaphol

Intelligent Systems and Applications

Smart Pulse Wave Detection System Using Intelligence

The pulse rate in the human species is a significant factor in oriental medicine. Pulse wave is a potential energy which has been generated by main artery expansion from the left ventricle, transforms into kinetic energy and passes through the vessel wall to the peripheral nervous system. In this paper, observations of patterns will be used to determine the blood circulation pulse wave which has been differentiated twice through the process of APG (Accelerated Photoplethysmo Graph). Moreover in this paper, it proposed the algorithm which diagnoses the patient’s health condition using intelligence method.

You-Sik Hong, Sang-Suk Lee, Dong-Hyun Nam, Woo-Beom Lee, Hongkyun Kim, Young Jun Song, Kwang-Baek Kim, Geuk Lee, Jeong-Jin Kang, Rethina Kumar, HyunSoo Jin, Chanwoo Moon, Youngdae Lee, ChangDuk Chung

Mean-Risk Model for Hybrid Portfolio Selection with Fuzziness and Randomness

This paper discusses the hybrid portfolio selection problem in the situation where only some security returns can be well reflected by their past data and are suitable to be described by random variables, but the other security returns can hardly be predicted through the historical data and are suitable to be described by fuzzy variables. By using chance theory, this paper extends the risk curve to hybrid portfolio selection and develops a hybrid mean-risk model. In addition, the way for computing the expected value and the risk curve of the hybrid portfolio return is provided and a genetic algorithm is presented for finding the optimal solution. As an illustration, an example is also provided.

Xiaoxia Huang

Investigation into the Effects of an Individual Amino Acid on Protein Function by Means of a Resonant Recognition Model

Upon identification of a new protein it is important to single out these amino acids responsible for the structural classification of the protein as well as the amino acids contributing to the protein’s specific biological characterisation. A novel approach is presented to identify and quantify this cause and effect relationship between amino acid and protein. This exploits the Common Frequency Peak (CFP) that it extracts from the Resonant Recognition Model (RRM). Applicability and robustness of the method are shown on a case study where five different protein families of the influenza A virus Neuraminidase (NA) genes are studied. They include H1N1, H1N2, H2N2, H3N2 and H5N1. The analyses identified five segments, namely three between H1N1 and H5N1 and two between H1N2, H2N2 and H3N2 and suggested that they play a key role in Influenza A NA gene functionality and can potentially be considered as target areas for future antiviral drugs and vaccines such as neuraminidase inhibitors.

Charalambos Chrysostomou, Huseyin Seker, Nizamettin Aydin

Optimizing a Personalized Cellphone Keypad

Current layouts for alphabetic input on mobile phone keypads are inefficient. We propose a genetic algorithm (GA) to find a suitable keypad layout for each user, based on their personal text history. It incorporates codes for frequent multigrams.

We optimize for two-thumb use, minimizing the number of strokes, and consecutive use of the same key or the same hand. Using these criteria, the algorithm re-arranges the characters on a 10-key pad. We demonstrate that this arrangement can generate a more effective layout, especially for SMS-style messages. Substantial savings are verified by computational analysis.

Joonseok Lee, Bob McKay

Hidden Node Pruning of Multilayer Perceptrons Based on Redundancy Reduction

Among many approaches to choosing the proper size of neural networks, one popular approach is to start with an oversized network and then prune it to a smaller size so as to attain better performance with less computational complexity. In this paper, a new hidden node pruning method is proposed based on the redundancy reduction among hidden nodes. The redundancy information is given by correlation coefficients among hidden nodes and this can save computational complexity. Experimental results demonstrate the effectiveness of the proposed method.

Sang-Hoon Oh

Attribute Grammar Genetic Programming Algorithm for Automatic Code Parallelization

A method is presented for evolving individuals that use an Attribute Grammar (AG) in a generative way. AGs are considerably more flexible and powerful than the


, context free grammars normally employed by GP. Rather than evolving derivation trees as in most approaches, we employ a two step process that first generates a vector of real numbers using standard GP, before using the vector to produce a parse tree. As the parse tree is being produced, the choices in the grammar depend on the attributes being input to the current node of the parse tree. The motivation is automatic parallelization or the discovery of a re-factoring of a sequential code or equivalent parallel code that satisfies certain performance gains when implemented on a target parallel computing platform such as a multicore processor. An illustrative and a computed example demonstrate this methodology.

Daniel Howard, Conor Ryan, J. J. Collins

Genetic Algorithms for Automatic Object Movement Classification

This paper presents an integrated approach, combining a state-of-the-art commercial object detection system and genetic algorithms (GA)-based learning for automatic object classification. Specifically, the approach is based on applying weighted nearest neighbor classification to feature vectors extracted from the detected objects, where the weights are evolved due to GA-based learning. Our results demonstrate that this GA-based approach is considerably superior to other standard classification methods.

Omid David, Nathan S. Netanyahu, Yoav Rosenberg

Optimizing for Change through Shades

The reliance of Evolutionary Algorithms on haploid genotypes has proved a difficult area for non-stationary function optimization. While it is generally accepted that various approaches involving diploidy can better cope with these kinds of problems, none of these paradigms have gained wide acceptance in the GA community. We describe Shades, a new haploid system which uses Polygenic Inheritance. Polygenic inheritance differs from most implementations of GAs in that several genes contribute to each phenotypic trait. A Knapsack non-stationary function optimization problems from the literature is described, and it is shown how Shades outperforms diploidy for this task.

Conor Ryan, J. J. Collins, Daniel Howard

A Simple Improvement Heuristic for Attributed Grammatical Evolution with Lookahead to Solve the Multiple Knapsack Problem

In this paper, we introduce a simple improvement heuristic to be used with Attribute Grammar with Lookahead approach (AG+LA), a recently proposed mapping approach for Grammatical Evolution (GE) using an attribute grammar (AG) to solve the Multiple Knapsack Problem (MKP). The results presented in this paper show that the proposed improvement heuristic can improve the quality of solutions obtained by AG+LA with little computational effort.

Muhammad Rezaul Karim, Conor Ryan

Validation Sets for Evolutionary Curtailment with Improved Generalisation

This paper investigates the leveraging of a validation data set with Genetic Programming (GP) to counteract over-fitting. It considers fitness on both training and validation fitness, combined with with an early stopping mechanism to improve generalisation while significantly reducing run times.

The method is tested on six benchmark binary classification data sets. Results of this preliminary investigation suggest that the strategy can deliver equivalent or improved results on test data.

Jeannie Fitzgerald, Conor Ryan

Sensor Network and Cloud Systems

The Design of Personal Virtualization Rule Based on Context-Awareness in Environment of Cloud Computing

Cloud services are possible to consisting of personal service using service virtualization for user. However, this process set up a group of users, and has offers a group service of common structure by organized group than a service of personalized cloud. Therefore, this paper proposes a rule of virtualization to provide personalized service with optimal resources in cloud computing. Proposed rule constitute personalized service to fit the user’s status by analyzing user’s situation. A model for personalized service configuration is based on MLP(Multi-Layer Perceptron). And, it should ensure the connectivity of service using connection weights for link of each layer. A history of Matched service with served DR(Direct Relationship) reconstruct the user’s context information by feedback. Thus, proposed rule provides personalized service automatically configured the information and application on user’s situation.

Hyogun Yoon, Hanku Lee

Private Cloud Computing Techniques for Inter-processing Bioinformatics Tools

Cloud computing is a very attractive alternative to get advantages of high performance data processing and easy management of the complex tools in a bioinformatics area. We present a method, in which computing resources such as existing PCs and small cluster systems can be utilized as private cloud computing infrastructure for an inter-query style of bioinformatics tools. We proposed a private cloud computing environment for bioinformatics applications of inter-processing tasks. Furthermore, we present a metadata repository schema and 6 query routing algorithms on it. We apply proposed algorithms to ClustalW, a multiple sequence alignment tool. Experimental result shows remarkable benefits in a proposed private cloud system in terms of performance and various user requirements.

Tae-Kyung Kim, Bo-Kyeng Hou, Wan-Sup Cho

Data Dissemination Protocol Based on Home Agent and Access Node for Mobile Sink in Mobile Wireless Sensor Networks

Achieving efficient data dissemination to a mobile sink moving is one of the main challenges in wireless sensor networks. In this paper, we describe DHA, a data Dissemination protocol based on Home agent and Access node. DHA introduces the home agent and access node concepts to represent the mobile sink in sensor networks. The movement of the mobile sink affects only the access nodes and home agent, and the sensor nodes can transfer data to the fixed home agent without knowing the position of mobile sink. All the sensor nodes except the home agent and access node don’t need to be updated in the direction of forwarding future data. DHA provides reliable and efficient data delivery to a mobile sink with minimum overheads. Our result shows that DHA increases the reliability of data delivery to the mobile sink and reduces the number of broadcast packets, thereby increasing the lifetime of sensor networks.

Joa Hyoung Lee, Jae Myung Kim, Byung Tae Jang, Eun-Ser Lee

Energy-Efficient Sensor Node Control Based on Sensed Data and Energy Monitoring

This paper proposes an energy-efficient sensor node control mechanism to prolong sensor networks’ lifespan by minimizing and equalizing energy consumption of sensor nodes. The proposed scheme newly defines three operational modes of a sensor node, which are normal, power-saving, and inactive. When the remaining energy of a sensor node is lower than the average remaining energy of all sensor nodes in the same network or cluster, it operates in the power-saving or inactive mode. This makes it possible to minimize and equalize energy consumption of each sensor node. The proposed mechanism additionally includes another scheme to prevent a sensor node transmitting duplicate sensed data. When a sensor node takes sensed data, it compares them with pre-sensed data to decide their duplicity. This makes it possible to avoid unnecessary energy consumption caused by transmitting duplicate sensed data. We implemented and simulated the proposed schemes with TinyOS and NS-2, respectively. The simulation results show that the proposed mechanism can efficiently reduce and equalize energy consumption, and therefore prolong sensor networks’ lifespan.

Ho-Guen Song, Dae-Cheol Jeon, Hee-Dong Park, Do-Hyeon Kim

A Data Recovery Method for High Accuracy in Data Centric Storage Schemes

In wireless sensor networks (WSNs), various data centric storage (DCS) schemes have been proposed to store the collected data and to efficiently process a query. The DCS scheme assigns divided data regions to sensor nodes and stores the collected data to the sensor which is responsible for the data region to process the query efficiently. However, since the whole data stored in a node will be lost when the fault of a node occurs, the accuracy of the query processing becomes low. In spite of the fact that such a serious problem exists, no works have been carried out on processing the faults of the nodes in the existing DCS schemes. To solve such a problem, we propose a novel data recovery method for high accuracy in DCS schemes with faults. The proposed method assures the high accuracy of the query result in the case of data losses due to the faults of the nodes in the DCS scheme. When a data loss occurs, the proposed method generates a compensation model against the area of the data loss data distribution patterns and data variation rate. It returns the query result including virtual data by using the compensation model. Therefore, it guarantees the query result with high accuracy in spite of the faults of the nodes. To show the superiority of our proposed method, we compare KDDCS with the proposed method with the existing DCS schemes without the data-loss recovery method. In the result, the DCS scheme (KDDCS) with the proposed method shows an about 98% accuracy rate on average in the various data sets. The accuracy of a query result is increased by about 37% over the existing DCS schemes without the data recovery method.

Junho Park, Dongook Seong, Jaesoo Yoo

A Delay Reduction Scheme Based on Network Coding for Voice Traffic in Large-Scale Wireless Sensor Networks

Recently, wireless sensor networks have been researched as core technologies for ubiquitous computing. In wireless sensor networks, a sensor node has typically small amount of memory, a processor with low computing power, and wireless interface of low bandwidth. Many sensor nodes should be placed to gather information in a large area because of small sensing range of the sensor nodes and limitation of wireless communications. In this paper, the cluster structure is used to organize large-scale sensor networks. Here, we propose a delay reduction scheme based on network coding for voice traffic by coping with the congestion problem in large-scale sensor networks. Finally, the simulation results show that the proposed scheme can reduce the transmission delay significantly.

Inwhee Joe, Kyunghwan Kim

Information Retrieval and Scheduling

A Vertical Search Engine for School Information Based on Heritrix and Lucene

The contents on the web are increasing exponentially as the rapid development of the Internet applications and services continues to expand. A problem in obtaining useful information from vast contents quickly and accurately is facing us while people are enjoying the convenience of the Internet. The immediate response to this problem is a Web Search Engine. We developed a vertical search engine for a certain domain like university. The search engine consists of Crawler, Indexer, and Searcher. The crawler component is implemented with Heritrix crawler based on the mechanism of recursion and archiving. A reusable, extensible index establishment and management subsystem are designed and implemented by open-source package named Lucene in the indexer component. An experiment has been done for Chungbuk National University web sites, and the number of documents the system retrieves is more than 4 hundred times on the average for typical keywords set than those from Google or university’s search engines.

Hyo-Bong Lee, Franco Nazareno, Seung-Hyun Jung, Wan-Sup Cho

An Improvement in Feature Feedback Using R-LDA with Application to Yale Database

This paper improves the performance of Feature Feedback and presents its application to face recognition. Feature Feedback has been introduced as a method which focuses on preprocessing the input data before classification. After extracting the features from original, Feature Feedback identifies the important part of the original data through the reverse mapping from the extracted features to the original space. In the feature extraction step, original feature feedback used PCA before LDA to avoid the small sample size problem but it has been shown that this may cause loss of significant discriminatory information. To overcome that problem, in the proposed method, we introduce feature feedback using regularized Fisher’s separability criterion to extract the features and apply it to face recognition using the Yale data. The experimental results show that the proposed method works well.

Lang Bach Truong, Sang-Il Choi, Gu-Min Jeong, Jeong-Min Seo

Effective Ranking Techniques for Book Review Retrieval Based on the Structural Feature

Buying products online is becoming a way of life for people. Buying books online is also one of a way of life. With an increasing number of people buying books online, reviews of books written by other people are becoming more important. However, review systems which are serviced by websites of online bookstores or collected reviews have limited browsing and searching functions. Moreover, search engine for reviews posted such websites doesn’t exist. Therefore, retrieval system which is proper for searching reviews has been required. To the best of our knowledge, this is the first work that studies ranking techniques for book review retrieval. In this paper, we propose ranking techniques for book review retrieval based on the structural features. We show that our ranking techniques outperform previous a ranking technique for theinternet information retrieval on searching for reviews.

Heungmo Ryang, Unil Yun

Efficient Food Retrieval Techniques Considering Relative Frequencies of Food Related Words

The scale of the Internet has become vast in the aspect of information. information. And, the performance of internet information retrieval systems are advanced. Now, researches of IIR (Internet Information Retrieval)systems are analysis of means of webpage based on keyword search. Recently, IIR system’s issue is system of searching necessary information for user. In this paper, we propose which search webpage of food related information and servicing the IIR system. Our system shows good performance than Commercial IIR services. We expect our system will be use-full IIR system.

Gwangbum Pyun, Unil Yun

Serial Scheduling Algorithm of LDPC Decoding

In this paper, we consider a layered decoding algorithm for a performance improvement of LDPC. The proposed algorithm is based on a serial scheduling algorithm for the layered decoding of LDPC in which we consider check node degree, variable node degree, and the edge connection between the each node. Simulation results show that the proposed method improves about 0.2dB SNR gain compared with random schedule in a layered decoding. Especially when the number of iteration is low, the performance gain gets larger, which case is applicable to the WPAN system which requires low power consumption and low latency.

Dong-Ho Kim, Ye-Hoon Lee

Design and Implementation of a Locking-Aware Scheduler for Multiprocessor Environments

Virtualization technology has known to be an efficient solution to reduce space and costs by using server consolidation and to provide flexible management of system resources. The widely used virtual machine monitor (VMM) such as Xen is not perfectly optimized for running over multi-processor systems. In order to run VMMs over multi-processor systems without performance degradation, the scheduling algorithms running in the VMM should be modified. This paper proposes a Locking-aware Scheduler (LAS) algorithm that has eliminated the lock-holder preemption (LHP) problem, which causes the inefficient use of CPU resources in multi-processor environments. By using LAS algorithm, Xen allows guest kernel to acquire spin-lock and prevents the preemption of related processors during operation. The performance results show that our proposed algorithm solves the LHP problem and reduces the total execution time.

Yongseob Lee, Wooseok Son, Sungyong Park

Graph Metrics for Predicting Speedup in Static Multiprocessor Scheduling

This paper presents a set of metrics for estimating the speedup achievable in static multiprocessor scheduling using a previously introduced Genetic Algorithm (GA) approach. This is of major importance because, although conventional wisdom suggests that metaheuristics such as GAs have the potential to improve over standard heuristics, little research has been conducted on characterizing the sorts of graphs that they should excel at. We describe several metrics and illustrate that four of them can predict the speed up with an accuracy of almost 90%.

Alan Sheahan, Conor Ryan

Hardware and Software Enginering

Accuracy Analysis of Power Characterization and Modeling

This paper presents a study on accuracy analysis of power models used for high-level power analysis in a chip design. The purpose is to find factors that are not modeled properly or not yet put into consideration. Different methods of characterizing input capacitance, effects of previous input states and input curves are discussed based on the comparisons with SPICE simulation results using 90nm bulk CMOS and 32nm Metal Gate/High-K Predictive Technology Model (PTM) models.

Xiaolan Bai, Hu Xu, May Huang

Design and Implementation of Turbo Decoder for TETRA Release 2 - TEDS

Terrestrial Trunked Radio (TETRA) is a digital trunked mobile radio standard developed by the European Telecommunications Standards Institute (ETSI). Especially, TEDS (TETRA Enhanced Data Services), part of TETRA Release 2, supports different channel bandwidths and modulation techniques, and turbo code has been adopted as a channel code with powerful error correcting capability. Due to its inherent high complexity and long latency of turbo code, the implementation of high-performance turbo decoder requires optimizations in various design levels. In this paper, we present the algorithmic modifications for low-complexity hardware implementation, architectural solutions and several optimizations for high-performance TEDS-compatible turbo decoder.

Ji-Hoon Kim

Performance Evaluation of 2x2 Space-Time Block Coded OFDM Systems under Time-Varying Fading Channels

In this paper, we evaluate the performance of space-time block coded orthogonal frequency division multiplexing (OFDM) systems with rate-2, full diversity, Golden code under time-varying fading channels. Simulation results show that the performance is very sensitive to the channel variation in the time domain.

Jeongchang Kim

Energy Efficient File Transfer Mechanism Using Deduplication Scheme

In a mobile device, it is important to minimize battery consumption when we operate an application. In this paper, we propose an energy efficient file transfer scheme for mobile device by exploiting deduplication scheme. Generally, a mobile phone needs more power for WiFi work rather than CPU work, therefore we have to minimize WiFi work. We can minimize the number of packets by adapting deduplication which consumes lots of CPU power. Deduplication work consumes additional CPU power during data hashing, however overall energy consumption can be diminished. Experiment result shows that the proposed scheme significantly reduces battery consumption if a file has over 5% SRP(Space Reduction Percentage).

Ho Min Jung, Sung Woon Kang, Jin Kim, Wan Yeon Lee, Jeong Gun Lee, Young Woong Ko

A Context-Driven Development Methodology for Context-Aware Systems

Context-aware systems demand a customized development methodology because they have their own features such as ambiguous system scope, context modeling, and implementing context-dependent services. Furthermore, it has been known that these systems are more complex than traditional systems. In this paper, we meet this challenge by extending the unified process (UP), and add three workflows: context requirements, context modeling, and context testing. At each workflow, stakeholders analyze, model, and test for the perspective context. We also provide guidelines for task performance, artifact production and development specifications for each stage of the context. Our methodology supports the full development process and the best practice technologies including OOP, UML, and UP.

Jongmyung Choi, Rosa I. Arriaga, Hyun-Joo Moon, Eun-Ser Lee

Architecture of Software Engineering Guideline Execution

Application lifecycle management (ALM) integrates the lifecycle workflows such as, requirements management, architecture, coding, testing, tracking, and release management. Process automation and seamless traceability across the process are so important. But, different terminology and different model were used to implement the tool for specific workflow so that it is not possible to communicate between the tools. In this paper, we suggest new approach to achieve the seamless traceability and process automation. For these features, knowledge formalization is critical. Evidence-based medicine is good reference architecture to ALM2. Software Engineering ontology is powerful solution to semantic gap between tools and guideline-based process execution environment can be practical solution to process automation. In this paper, we suggest the architecture for automating the ALM using process management technology and guideline-based engineering.

Jeong Ah Kim, Seung-Yong Choi, Sun-Myung Hwang

Security Systems

Specification-Based Intrusion Detection System for WiBro

WiBro (Wireless Broadband), the service based on the IEEE 802.16e (mobile WiMAX) standard, is a wireless broadband Internet technology developed by the South Korean telecommunications industry. In this paper, we analyze security vulnerabilities of the WiBro standard, focusing on the initial authentication phase and propose a specification-based intrusion detection system that can detect those vulnerabilities. We first derive specifications from the normal operational process of the initial authentication based on PKMv2 EAP-AKA and formalize the derived specifications as a state transition diagram. The proposed system executes the intrusion detection based on those specifications and the state transition diagram. In this paper, to verify the detection capability of the proposed system, we construct an experimental network and execute scenario-based tests.

Yunho Lee, Sang-Guun Yoo, Juho Kim, Soojin Lee

An Efficient Password Authentication Method Using CAPTCHA

User authentication is one of the most important security requirements for online financial transactions services to support users on the web. These securityservices should posses features, such as secure tunneling connection, and sniping attackprevention, keylog attack prevention etc. Malicious attackers maycapture user information such as users entered authentication information (ID, Password, card number, bank account number, etc)through keyboard hooking.This paper proposesa keyboard hooking protected password input method using CAPTCHA. The proposed password input method is based on entering the password using mouse click or touch pad on the CAPTCHA image. The mapping of CAPTCHA image pixels is random.

Beum-Su Park, Amlan Jyoti Choudhury, Young-Sil Lee, Tae-Yong Kim, Hoon-Jae Lee

A Study on Wi-Fi Hacking Attack Using Web

Using Smart Phone, Tablet PC and laptop that SNS, or they want to surf the web to obtain information by connecting to the Internet. But if you use the hacked site Malicious code is downloaded incident occurs. These attacks by the attackers hacked into other servers within the same IP segment after that ARP Spoofing by inserting malicious code associated with a particular Web server intercepts Web Traffic. In this paper, Log in Smart Phone, Tablet PC and Laptop is based, SNS and web site. At this point, Network Packet analysis tools to analyze the contents of the Packet and Encrypted website hacking content and ID, password by hacking incident triggers. Smart Phone, Tablet PC and laptop are presented in the study of security measures for the incident. In this study, Smart Phone, Tablet PC and laptop will contribute that network security Enhance to the development of Hack defense technology.

Jin-Young Song, Dea-Woo Park

Improved Security Proofs of Delerablée’s Identity-Based Broadcast Encryption Scheme

Delerablée proposed the first identity-based broadcast encryption scheme with constant size ciphertexts and private keys in [7]. But, the scheme was proved to be secure in the selective identity security model. We prove the scheme is secure in the full security model under an interactive assumption.

Sun-Mi Park, Seong-Oun Hwang

Cryptanalysis of Nonce-Based Mutual Authentication Scheme Using Smart Cards

Recently, many user authentication schemes using smart cards have been proposed to improve the security weaknesses in user authentication process. In 2008, Liu et al. proposed a new mutual authentication scheme using smart cards which is based on nonce and can withstand the existing of forged attack. In this paper, we analyze the security weaknesses of Liu et al.’s scheme. Although Liu et al.’s scheme overcame the vulnerability of Shen et al.’s scheme, we have shown that Liu et al.’s scheme is still insecure against the Impersonation attack and password guessing attack, and does not provide the mutual authentication between the user and the remote server.

Young-Hwa An, Youngdo Joo

An Attack on an RFID Authentication Protocol Conforming to EPC Class 1 Generation 2 Standard

Radio Frequency Identification (RFID) is taking place of barcodes in our lives, thanks to its remote identification capability. However, being transmitted via radio waves, the information is exposed to many possible attacks, which may endanger the security and privacy of relevant individuals and organizations. Fortunately, there have been efforts to construct an efficient and secure protocol in conformance with EPC Class 1 Generation 2 Standard (Gen 2), the most popular standard for RFID passive tags. Yeh et al. proposed a mutual authentication protocol conforming Gen 2. However, the proposed protocol is found to be vulnerable to an attack. Thus, in this paper, we show the vulnerability of Yeh et al.’s protocol and propose countermeasures to secure the protocol.

Changbin Lee, Sunwoo Park, Kwangwoo Lee, Dongho Won

Abnormal Traffic Detection System of VoIP Based on SIP

VoIP services need, as other application services do, monitoring and abnormal traffic detection function. However, existing IP-based traffic analysis and detection technique is not appropriate for VoIP. This paper proposes traffic analysis technique that takes the characteristics of VoIP application layer into consideration, and a system which detects abnormal VoIP traffic.

Kyu-won Lee, Jae-Won Ji, Seung-Jae You, Geuk Lee

Robotics and RFID Systems

Emergency Response System Using a Surveillance Camera with Collaborative Multiple Robots

This paper proposes a new type of emergency response system that can detect an emergency situation using vision based human action recognition with collaborative mobile robots to deal with emergency situations. A vision-based emergency detection scheme that can detect suspicious situations by analyzing image sequences from a surveillance camera is developed. After detecting a human in image sequences, the detected human regions are tracked and human actions are then recognized using a motion history image and a neural network. When an emergency occurs, a mobile robot is sent to the emergency site to diagnose the situation at close distance. After reaching the target location, the robot transfers information on the situation such as sound and images.

Yong-Ho Seo, Hyun S. Yang

Development of a PC-Based Network Performance Evaluation System for Intelligent Robots

In this paper, a portable in-robot network evaluation system is developed based on a PC with a CAN card to analyze the network performance inside intelligent robots. A lot of sensors, actuators, and controllers have been installed in robots for their intelligent behavior. The complex wiring between sensors, actuators, and controllers due to the point-to-point connection has been being replaced by bus-type networks. Among several types of communication protocols for bus networks, the CAN(Controller Area Network) protocol has been widely used for in-vehicle networks and now it has been applied to in-robot networks. The real-time operation of the control system for a humanoid type robot is analyzed by using the developed evaluation system, and it is shown that the control sampling time should be determined with the consideration of the maximum standard deviation of each transmission period.

Kyung-Jung Lee, Jae-Oh Kim, Jae-Han Park, Moon-Hong Baeg, Hyun-Sik Ahn

Linear Velocity Based Predictive Control Design and Experiment for Pursuit-Evasion of a Multiple Small Scale Unmanned Helicopter

In this paper, linear velocity based predictive control of unmanned aerial vehicle (UAV) for pursuit-evasion scenario is presented. The issues of multi-UAV control system are resolved with distributed control structure, to control and coordination multiple pursuers. The group operation of UAV’s is realized by decentralized linear velocity predictor control, which interacts with other UAV through communication system. In this research Reynold’s biologically inspired steering behavior model has been utilized. The proposed algorithm is implemented on JR Voyager G-260 small scale helicopter controller and tested with HILS (Hardware-In-the-Loop Simulation). With the multiple UAV’s the variety of test from waypoint to pursuit-evasion is taken to demonstrate the effectiveness of our approach.

Vishnu Kumar Kaliappan, Hanmaro Yong, Agus Budiyono, Dugki Min

Real-Time Map Generation Using Constraint Delaunay Triangulation

Navigation and automatic map generation are critical in robotics. This paper presents a map generation and navigation algorithm using constraint Delaunay triangulation. A local map around a robot is generated and global map of the environment is updated in real-time using the sensor data collected by the robot. Applying Minkowski sum a collision-free path is created to guide the robot moving toward its objective. Experimental results are shown in the paper to present the accuracy of generated maps.

Peng Wang, Yi Liu, Taiheng Jin, May Huang

Improving Position Estimation Accuracy of Mobile Robot by Using UHF Passive RFID Tags

This paper proposes a practical and improved position estimation method for real-industry application in localizing a mobile robot that is designed to move along a pre-defined path. The proposed method replaces the use of proximity sensors with much more economical UHF passive RFID tags to perform localization tasks. Considering that there is already a wide dissemination of RFID tags in various industries, the proposed method allows efficient and effective estimation of the position of a target mobile robot by using the already-present matrix of RFID tags as a reference point for localization. Particularly, the proposed method makes use of two readers with identical emission configurations attached to a mobile robot to identify a reference RFID tag. By utilizing the RSSI obtained by the readers from the reference RFID tag, the precise position of the moving robot can be obtained. The experiments prove the effectiveness achieved by the proposed method in improving the accuracy of robot position estimation.

Sangdo Park, Taeho Kim, Hongchul Lee

A Centralized Arbitration Scheme for UHF Band RFID Readers

When attempting to identify multiple tags, the reader may face up a simultaneous interfering transmission from transponders which may causes a collision as tags share the same communication channel. Also, reader collision may occur when neighboring readers use the same channel at the same time. This paper proposes the centralized arbitration scheme using a RFID-Arbiter. The RFID-Arbiter operates as a special tag in the air interface so that hardware of reader does not need to be modified. By the centralized arbitration scheme, readers that want to communicate with tags are allotted to operation at different times. For performance evaluation, the proposed centralized arbitration scheme is modeled and simulated by using discrete event system formalism.

Sung Woo Bae, Jun-Seok Park, Ha-Ryoung Oh, Yeong-Rak Seong

Analysis of the Future Internet of Things Capabilities for Continuous Temperature Monitoring of Blood Bags in Terrestrial Logistic Systems

This paper presents an analysis for the challenges and opportunities for the definition of a solution for traceability, tracking and monitoring based on Future Internet of Things to transport of goods, which require an item-level continuous monitoring solution, such as the case of perishable goods, food, medicines and dangerous goods due to the specific needs of these products in terms of environment status (humidity, lighting and temperature) during the transport, in order to preserve its integrity before, during and after the transport process. For that reason, this work describes a particular use case focused on the implementation of the solution for careful monitoring of blood bags, since this requires the communication with the items in order to control the temperature in the bags to detect that it has not been out of range, and control during the transport the temperature in the trailer in order to avoid that situation. This analysis is based on the Internet of Things to support global connectivity through the Future Internet for each one of the items, and on the Web of Things to support the monitoring of the item status and access to services at anytime during the transportation of goods through Embedded Web Services (RESTful).

Miguel Castro, Antonio J. Jara, Antonio F. G. Skarmeta

Pattern Recognition, Image Processing and Clustering

New Single Camera Vehicle Detection Based on Gabor Features for Real Time Operation

This paper presents a new novel approach for automatic vehicle detection from a live video based on texture Gabor features. Vehicle detection is a pivotal part in collision avoidance systems, blind-spot monitoring, and self-guided vehicles. This system uses a low cost camera mounted near the rear-view mirror to obtain the live video. The Gabor filter features have been used to identify the potential vehicles in the frames by using the support vector machine. The initial detection of potential vehicle candidates has been improved by using correlation techniques. A robust detection technique is developed in which the vehicles are detected with accuracy up to 9 % in day light image sequences.

Mirza Waqar Baig, Syed Jahanzeb Hussian Pirzada, Ehsan ul Haq, Hyunchul Shin

Effect of Different Window and Wavelet Types on the Performance of a Novel Crackle Detection Algorithm

Pulmonary crackles are used as indicators for the diagnosis of different pulmonary disorders. Crackles are very common adventitious sounds which have transient characteristic. From the characteristics of crackles such as timing and number of occurrences, the type and the severity of the pulmonary diseases can be obtained. In this study, a novel method is proposed for crackle detection, which uses time- frequency and time-scale analysis, and the performance comparison for different window types in time-frequency analysis and also for different wavelet types in time-scale analysis is presented. In the proposed method, various feature sets are extracted using time-frequency and time-scale analysis for different windows and wavelet types. The extracted feature sets are fed into support vector machines both individually and as an ensemble of networks. Besides, as a preprocessing stage in order to improve the success of the model, frequency bands containing no-information are removed using dual tree complex wavelet transform, which is a shift invariant transform with limited redundancy and an improved version of discrete wavelet transform. The comparative results of individual feature sets and ensemble of sets with pre-processed and non pre-processed data for different windows and wavelets are proposed.

Gorkem Serbes, C. Okan Sakar, Yasemin P. Kahya, Nizamettin Aydin

Wavelets and Decision Trees for Target Detection over Sea Surface Using Cosmo-Skymed SAR Data

We are investigating the combination of wavelets and decision trees to detect ships and other maritime surveillance targets from medium resolution SAR images. Wavelets have inherent advantages to extract image descriptors while decision trees are able to handle different data sources. In addition, our work aims to consider oceanic features such as ship wakes and ocean spills. In this incipient work, Haar and Cohen-Daubechies-Feauveau 9/7 wavelets obtain detailed descriptors from targets and ocean features and are inserted with other statistical parameters and wavelets into an oblique decision tree.

Rafael L. Paes, Aylton Pagamisse

An Efficient Online Signature Verification Scheme Using Dynamic Programming of String Matching

The necessity to authenticate individuals is rapidly increasing day by day with the explosive growth of E-commerce, E-finance, PDA, etc. Handwritten signature is the most widely used and easiest way to verify a person. Online signature verification is a very active and hot topic in the field of biometric research. It is a potential candidate to replace traditional password-based security system as the password can be forgotten, stolen or guessed. Online signature verification deals with both spatial and temporal features of signature. Therefore, it is difficult to forge. This paper proposes a novel online signature verification technique using dynamic programming of string matching. The performance of the proposed approach is evaluated for both genuine signatures and skilled forgeries on SVC2004 database. The proposed approach produces a False Acceptance Rate (FAR) of 4.13% and False Rejection Rate (FRR) of 5.5% with an Equal Error Rate (ERR) of 5%.

Ahmed Galib Reza, Hyotaek Lim, Md Jahangir Alam

Topological Graph Matching Based Dot Pattern Recognition Scheme for Smart Book

In this paper, we present a design and implementation of smart book system using pen based augmented reality approach. Student has been given a smart pen and a smart study book, which is a usual printed material with CMY inks for contents and K ink for perceptually invisible dot patterns. When a student touches a smart study books with our smart pen, we show his/her multimedia (visual/audio) information which is exactly related with the selected region. Smart pen includes (1) IR LED/IR pass filter for illumination independent extraction of the dot patterns, and (2) camera for image captures. From the image sequences, we perform topological graph analysis which determines the topological distance between dot pixels, and perform error correction decoding using four position symbols and five CRC symbols. Our scheme can embed 16 bit information, which is more than 200% larger than previous scheme, which supports 7 bits or 8 bits information at most.

Jae Youn Shim, Seong-Whan Kim

Speaker Identification Using Ensembles of Feature Enhancement Methods

In this paper, we propose a classifier ensemble of various channel compensation and feature enhancement methods for robust speaker identification on various environments. The proposed ensemble system is constructed with 15 classifiers including three channel compensation methods (including CMS and variance normalization, and without compensation) and five feature enhancement methods (including PCA, kernel PCA, greedy kernel PCA, kernel multimodal discriminant analysis, and without enhancement). Experimental results show that the proposed ensemble system gives the highest average speaker identification rate in various environments (channels, noises, and sessions).

IL-Ho Yang, Min-Seok Kim, Byung-Min So, Myung-Jae Kim, Ha-Jin Yu

Estimating Principal Properties on Triangular Meshes

In this paper, we estimate the principal curvatures of surfaces represented as triangular meshes using a generic model constructed from Euler Formula. Normal curvatures are sampled in the vertices on the direction of a 1-ring neighborhood of a target vertex. Our model is then fitted to a normal curvature graph that connects the pairs of sampled directions and normal curvatures. The principal properties of the surface at the target vertex are estimated from the parameters of the fitted model. We have tested our algorithm on triangular meshes generated from analytic surfaces and show that the errors in our technique are acceptable. We also apply the method to mesh models and discuss degenerate cases.

Kyungha Min

Image Processing for Automated Analysis of the Fluorescence In-Situ Hybridization (FISH) Microscopic Images

The paper describes automated segmentation and analysis of the microscopic images resulting from fluorescence in-situ hybridization (FISH) analysis. FISH is a popular molecular cytogenetic method. The output of a single FISH analysis is a set of several tens or hundreds microscopic images — a single evaluated sample is of roughly 20mm diameter. The goal of an automated evaluation is to replace the subjective evaluation of images by the laboratory technician to achieve higher uniformity of results. Following explanation of the principle of the method and the typical contents of images, the processing flow of image segmentation is outlined and the results are presented on several example images. With emphasis on a low-cost solution, the ITK library is used for implementation.

Jan Schier, Bohumil Kovář, Eduard Kočárek, Michal Kuneš

An Efficient Edge Detection Using Raster CNN Simulator

This paper presents a cellular neural network based edge detection using Raster CNN Simulator. The software is designed to handle with both gray level and color images. The experimental result of Raster CNN Simulator is compared with traditional edge detection operators Canny and Sobel. Simulation results show that the proposed simulator is accurately detecting the complete image edge and also save the computation time.

V. Murugesh, Kyung-Tae Kim

The Size and Position Detection of the Small Target in Infrared Image

In this paper, we present a technique used to detect the location and size of small targets in a multi-resolution image using a cubic facet model. The input image is divided into multi-resolution images. We apply the facet model and the local maxima conditions to each level of the multi-resolution images. We then detect the location of the small target. We derive the location for the maximum of the



, i.e. the local maxima value of the facet model, and use this for the location of the small target. We can detect small targets of various sizes in each level of the multi-resolution images. In this paper, we experimented using various infrared images possessing a small target. The conventional facet model method applies a mask. However, the proposed method applies a mask to the multi-resolution images. We verified our method by varying the mask size and altering the size of the small target. We found that our proposed algorithm can detect the location and size of the small target.

Gyoon-Jung Lee, Ji-Hwan Park, Jae-Heum Joo, Ki-Gon Nam

Compressed Sensing Meets the Human Visual System

Foveation principles suggested by the Human Visual System have already been used with significant success for compression purposes on both 1D and 2D data. The method provides spatially variable quality of the reconstructed information, enabling better approximation of specific regions of interest. Combining this approach with the principles behind Compressed Sensing theory enable further improvement of compression ratio performances, as indicated by experimental results on a set of natural images.

Iulian B. Ciocoiu

Integer KLT Design Space Exploration for Hyperspectral Satellite Image Compression

The Integer KLT algorithm is an approximation of the Karhunen-Loève Transform that can be used as a lossless spectral decorrelator. This paper addresses the application of the Integer KLT to lossless compression of hyperspectral satellite imagery. Design space exploration is carried out to investigate the impact of tiling and clustering techniques on the compression ratio and execution time of Integer KLT. AVIRIS hyperspectral images are used as test image data and the spatial compression is carried out with JPEG2000. The results show that clustering canspeed up the execution process and can increase the compression performance.

Nor Rizuan Mat Noor, Tanya Vladimirova

Model-Based Clustering Analysis of Student Data

Many students fail because of academic and financial difficulties and dissatisfaction with their instruction and academic environment. Educational data mining(EDM) community tries to find solutions for such problems by mining student’s data. There are a wide variety of current methods popular within educational data mining but the topics of EDM research are changing. Discovery with models has recently gained widespread use in EDM and it was the second most common category of EDM research by 2008-2009. In this article an important sub method of discovery with models (model based cluster analysis) is addressed. We explain the rule of MBCA in EDM, MBCA algorithms, interests of MBCA, and an experimental comparison of MBCA algorithms, and attributes selection methods on student’s data.

Mirwais Tanai, Jongwan Kim, Joong Hyuk Chang

Numerical Evaluation of Clustering Methods with Kernel PCA

Kernel methods are ones that, by replacing the inner product with positive definite function, implicitly perform a nonlinear mapping of the input data into a high-dimensional feature space. The clustering methods using kernel function (kernel clustering methods) are superior in accuracy to the conventional ones such as K-Means (KM) and Neural-Gas (NG). But, it seems that kernel clustering methods do not always show sufficient ability of clustering. One method to improve them is to find expression of approximation for data in the feature space. In this paper, we introduce the kernel PCA and apply it to clustering methods as KM and NG. Further, we apply it to derived kernel method, which means twice application of kernel functions. The simulation results show that the proposed clustering methods are superior in terms of accuracy to the conventional methods.

Hiromi Miyajima, Noritaka Shigei, Tomiyuki Shiiba

Data Mining

Improved Sampling for Triangle Counting with MapReduce

The problem of counting the number of triangles in a graph has gained importance in the last few years due to its importance in many data mining applications. Recently, Tsourkakis, et al. proposed DOULION, which is based on a simple sampling idea but works very well on many of the important graphs. In this preliminary report, we show that DOULION may not be very correct on special cases of graphs and argue that it may not fulfill the main purpose of the triangle counting problem for real-world graphs. We then present improvements on DOULION and show that it works better, much better in some cases, than DOULION.

Jin-Hyun Yoon, Sung-Ryul Kim

A Resource Discovery Scheme for Mobile Peer-to-Peer Networks with Data Dissemination

In this paper, we propose a resource discovery scheme of mobile P2P with data dissemination that considers the interests and mobility of the peer. The proposed scheme manages a local information table, a resource index table, and a routing table in local database to enhance the accuracy and cost of the resource discovery. Each peer is recognized through a signal function and then the cost of data dissemination is reduced by timestamp messages. The transmitted resource is stored in the resource index table by ranking function. Our ranking function considers the distance, direction and peer’s interest item in order to improve the performance of a resource discovery.

Kyoung-Soo Bok, Dong-Won Kwak, Jae-Soo Yoo

A Novel Outlier Detection Method for Spatio-Tempral Trajectory Data

The development of mobile device technology and localization technology makes the collection of spatio-temporal information from moving objects much easier than before, and outlier detection for spatio-temporal trajectory is becoming increasingly attractive to data mining community. However, there is a lack of serious studies in this area. Several existing trajectory outlier methods such as the partition-and-detect framework can only deal with the trajectory data which only includes spatial attributes. It cannot be applied to the spatio-temporal trajectory data which includes both spatial and temporal attributes. In this paper, we propose an enhanced partition-and-detect framework to detect the outliers of spatio-temporal trajectory data. In this framework, we mainly introduce an outlier detection method which uses trajectory MBBs(Minimum Boundary Boxs). Based on this enhanced framework, we propose a congestion outlier detection method. Finally, the efficiency and accuracy are evaluated through experiments which use a real traffic dataset called US Highway 101 Dataset.

Yan Li, Weonil Chung, Hae-Young Bae

Enumeration Tree Based Emerging Patterns Mining by Using Two Different Supports

Recently, the analysis of power load in the electrical industry has becomes an important element for the concern of customer safety. In power system related studies, data mining techniques are used in power load analysis and they can help decision making in the electrical industry. In this paper, for using emerging patterns to define and analyze the significant difference of safe and non-safe power load lines, and identifying which line is potentially unsafe, we proposed an incremental TFP-tree algorithm for mining emerging patterns that can search efficiently within memory limitation. Especially, the use of two different minimum supports makes the algorithm possible to mine most number of emerging patterns and efficiently handle the incrementally increased, large size of data sets such as power consumption data.

Minghao Piao, Jong Bum Lee, Ho Sun Shon, Unil Yun, Keun Ho Ryu

Assessment Building a Method for Risk Model of Mountain Bike Accident Based on Classification Techniques

In recent years, many researchers have studied the risk models, which are able to prevent various accidents occurred on riding mountain bike. The bikers prevent themselves from the accidents using a model, which is able to predict the risk of mountain bike. The objective of this paper is to build a method of risk model for predicting mountain bike accident using information of mountain bike courses obtained by global positioning system and difficulty of courses for bikers. We evaluate the classification methods such as C5.0, Bayesian network, neural network, and support vector machine to find a method with higher accuracy comparing with other. Also, we build a risk model using factors with high risk on riding mountain bike. Proposed method ensures better selection of a course and reduction of the accident risk for mountain bikers.

Dong-Hyok Seo, Suk Ho Jung, Soo-Ho Park, Kwang-Sun Ryu, Hyeongsoo Kim, Dong Gyu Lee, Keun Ho Ryu

Extracting and Ranking Relevant Terms of Personalized Search Query

The pervasiveness of Internet use in daily life has made it possible for people to retrieve a large amount of information through the web. However, the exponential growth in the quantity of information on the web has imposed limits on the performance of online search engines, which retrieve enormous amounts of unwanted information. With so much unwanted information, web users now require more time and effort, than in the past, to search for the required information. In this study, a method that uses query expansion to quickly bring wanted information to web users has been proposed.

Jung-Hun Lee, Suh-Hyun Cheon

A Sentiment Analysis of Audiences on Twitter: Who Is the Positive or Negative Audience of Popular Twitterers?

Microblogging is a new informal communication medium of blogging that differs from a traditional blog in which content is much shorter. Microbloggers post about topics that describe their current status. Twitter is a popular microblogging service and social media where users read and write millions of short messages on any topics within the 140-character limit. Twitter is used to find the context of social trend. Popular or influential users tweet about their status and are retweeted, mentioned or replied to by their audience. A sentiment analysis of the tweets of popular users and their audiences discovers whether the audience is favorable or unfavorable to the views expressed by such popular users. We conducted a content analysis of over 1,000,000 tweets mentioning or replying to their thirteen most influential users to discover the sentiment of the audience. Arguably, Twitter messages reflect the silent landscape of sentiment towards its most popular users. This study uses the technique of Sentiment Analysis to deliver a valid popularity indicator or measure.

Younggue Bae, Hongchul Lee

Human Computer Interaction

Reading Desk for Preschool Children and Older People with Emotional Speech Synthesis

In this paper, we introduce a reading desk designed to read books to the older people and children. For this purpose, we propose a reading desk together with an emotional speech synthesis system for Korean. The reading desk system provides a wireless audio output unit, and the reading desk is directly connected to a laptop computer in order to identify the current user and target reading material. The emotional speech synthesis system for Korean is a prosody re-synthesis system that has the option of providing four different emotions such as anger, fear, happiness, and sadness. Therefore, this system is also able to modify the speech rate and intensity information of speech as much as users want. We analyzed 240 pieces of emotional speech in order to extract distinct prosody structures for each emotion in Korean. The evaluation results show that we have achieved 48.5% of the recognition rate for happiness among four emotions, and with enough training experience, the average recognition rate has improved up to 95.5% for all emotions.

Ho-Joon Lee, Yong-Jae Lee, Jong C. Park

Background Reflectance Modeling for Robust Finger Gesture Detection in Highly Dynamic Illumination

We aim to develop an ’accessory-free’ or ’minimum accessory’ interface used for communication and computation without the requirement of any specified gadgets such as finger markers, colored gloves, wrist bands, or touch screens. We detect various types of gestures, by finding fingertip point locations in a dynamic changing foreground projection with varying illumination on an arbitrary background using visual segmentation by reflectance modeling as opposite to recent approaches which use IR(invisible) channel to do so. The overall performance of the system was found to be adequately fast, accurate, and reliable. The objective is to facilitate in the future, a direct graphical interaction with mobile computing devices equipped with mini projectors instead of conventional displays. We term this a dynamic illumination environment as the projected light is liable to change continuously both in time and space and also varies with the content displayed on colored or white surface.

Armin Mustafa, K. S. Venkatesh

A New Approach for Human-Robot Interaction Using Human Body Language

In order to make the interaction between human and robot easier and more diversity, in this paper, we construct a system in which human users can use the body language to control robot doing some works. At first, the human body data is collected via 3D camera. And then, we extract the skeleton feature. Based on the human posture and Semaphore system, the international communicative method, robot can get the character in the Alphabet. Finally, robot combines the separate character into the understandable message and executes what user wants to do. In simulation, we show the results in which the iRobot can move up, down, turn left, turn right...based on the received body message.

Nhan Nguyen-Duc-Thanh, Daniel Stonier, SungYoung Lee, Dong-Han Kim

An Improvement of Mobile Web Interface Design Optimized for Tablet PC Environment

Internet connection through mobile web has been abruptly increased with the recent universalities of smart phone and tablet PC. On the other hand, it might be more difficult to have free surfing internet, smoothly running mobile web and/or easily approaching various mobile sites because there are a lot of domestic web sites that do not keep the web standard. They just seem to obey various performing capabilities and outfit design. This will be the main reason of decreasing usability and accessibility when tablet PCs and mobile terminals are connecting to the Internet. In this paper, we propose a novel method of mobile web interface to improve its usability and accessibility optimized for mobile environment. Our interface tool can be easily used by industrial or individual users to access various web sites through tablet PC’s or other mobile communication media. Applying the proposed interface to a company web site, which is the final focus of this paper, we prove its advanced performance and increased usability. And then we offer this new direction of user interface design to the developers of industrial or individual mobile web sites.

Hyoung-Yong Oh, Byung-Won Min, Yong-Sun Oh

Text Input Tool for Immersive VR Based on 3 × 3 Screen Cells

This paper presents a noble text input tool for immersive VR by exploiting 3x3 screen cells. Our tool presents an input interface, which is similar to that of mobile phones. As the character layout on the interface, we exploit mobileQWERTY layout that arranges alphabet characters in 3x3 keys according to the QWERTY-similarity. In order to present more convenient pointing scheme for a key on the interface, we present a Screen Cell-based scheme, which divides the whole screen in front of a user into 3x3 structure. Instead of pointing a small key in the interface with a 3D pointing device, which shows less accuracy than familiar 2D pointing devices such as mouse, users can point the key indirectly by pointing its corresponding cell. We develop a VR game using our text input tool. Finally, our tool is compared to other text input tools: onscreen keyboard and onscreen keypad with direct input. The result of the comparison shows our tool records best performance and learnability.

Kyungha Min


Weitere Informationen

Premium Partner