• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.9
• /
• pp.3250-3265
• /
• 2014

Integrated WiMAX and WiFi networks is of great potential for the future due to the wider coverage of WiMAX and the high data transport capacity of WiFi. However, seamless and secure handover (HO) is one of the most challenging issues in this field. In this paper, we present a novel vertical HO authentication scheme with privacy preserving for WiMAX-WiFi heterogeneous networks. Our scheme uses ticket-based and pseudonym-based cryptographic methods to secure HO process and to achieve high efficiency. The formal verification by the AVISPA tool shows that the proposed scheme is secure against various malicious attacks and the simulation result indicates that it outperforms the existing schemes in terms of communication and computation cost.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.5
• /
• pp.989-998
• /
• 2011

Cloud computing provides computation, software, data access, and storage services that do not require end-user knowledge of the physical location and configuration of the system that delivers the services. Cloud computing providers have setup several data centers at different geographical locations over the Internet in order to optimally serve needs of their customers around the world. One of the fundamental challenges in geographically distributed clouds is to provide efficient algorithms for supporting inter-cloud data management and dissemination. In this paper, we propose a group quorum system (GQS)-based dissemination for improving the interoperability of inter-cloud in time-critical event dissemination service, such as computing policy updating, message sharing, event notification and so forth. The proposed GQS-based method organizes these distributed clouds into a group quorum ring overlay to support a constant event dissemination latency. Our numerical results show that the GQS-based method improves the efficiency as compared with Chord-based and Plume methods.

• ETRI Journal
• /
• v.45 no.6
• /
• pp.1035-1045
• /
• 2023

We propose a novel graphics processing unit (GPU) algorithm that can handle a large-scale 3D fast Fourier transform (i.e., 3D-FFT) problem whose data size is larger than the GPU's memory. A 1D FFT-based 3D-FFT computational approach is used to solve the limited device memory issue. Moreover, to reduce the communication overhead between the CPU and GPU, we propose a 3D data-transposition method that converts the target 1D vector into a contiguous memory layout and improves data transfer efficiency. The transposed data are communicated between the host and device memories efficiently through the pinned buffer and multiple streams. We apply our method to various large-scale benchmarks and compare its performance with the state-of-the-art multicore CPU FFT library (i.e., fastest Fourier transform in the West [FFTW]) and a prior GPU-based 3D-FFT algorithm. Our method achieves a higher performance (up to 2.89 times) than FFTW; it yields more performance gaps as the data size increases. The performance of the prior GPU algorithm decreases considerably in massive-scale problems, whereas our method's performance is stable.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.26 no.8
• /
• pp.1178-1186
• /
• 1989

In this paper, a systolic array for polyadic-nonserial DP problems is designed, the performance is analyzed and the efficient operating method is proposed. The algorithm is transformed to remove the broadcasting and global communication paths in the data dependence step by step. The transformed algorithm is mapping to the systolic array using the method proposed by D. I. Moldovan. The designed array is homogenous, had the processing elements of (n+1)/2 and 2n computation time ( n is the size of problem). In case of being many problems to process, the efficiency of array can be upward by inputing the problems successively. The interval between the initiations of two successive proboem instances is [n/2]+1 and the speed-up is about 4. The processor utilizations of each case are calculated.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.9
• /
• pp.1275-1278
• /
• 2021

Nowadays, fake news from newspapers and social media is a serious issue in news credibility. Some of machine learning methods (such as LSTM, logistic regression, and Transformer) has been applied for fact checking. In this paper, we present Transformer-based fact checking model which improves computational efficiency. Locality Sensitive Hashing (LSH) is employed to efficiently compute attention value so that it can reduce the computation time. With LSH, model can group semantically similar words, and compute attention value within the group. The performance of proposed model is 75% for accuracy, 42.9% and 75% for Fl micro score and F1 macro score, respectively.

• International Journal of Aeronautical and Space Sciences
• /
• v.5 no.1
• /
• pp.64-74
• /
• 2004

The Grid is a communication service that collaborates dispersed highperformance computers so that those can be shared and worked together. It enablesthe analysis of large-scale problem with the reduction of computation time bycollaborating high performance computing resources in dispersed organizations. Thus,the present paper focuses on the efficient flow calculation using the Grid. To increaseparallel efficiency, a simple load balance algorithm for the Grid computing is proposedand applied to various aerodynamic problems.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.11A
• /
• pp.822-830
• /
• 2009

The spectrum efficiency of mobile communication networks can be improved dramatically adopting multiple antennas technologies. In order to guarantee the licensed rights of primary user (PU), the cognitive radio system should perform in a relatively low interference manner when it gets access to the spectrum of licensed networks. In this paper, we explore a uniformly distributed circular antenna array to implement beamforming algorithm that is accomplished by optimization method at the base station of cognitive radio networks, and therefore we can suppress the interference to PU by steering quite low transmission power toward PU and constructing a narrow beam toward cognitive user (CU). By reducing the constraint number of the optimization problem, we also propose a null steering algorithm that steers rather low radiation power toward PU, while the other areas in the same cell are covered by radiation power except the local area around PU. It is pursued to reduce the computation load and enlarge the capacity of cognitive radio networks extremely. The simulation results demonstrate that the proposed algorithms process superior performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.9
• /
• pp.4527-4547
• /
• 2018

Signcryption is a cryptographic primitive that provides authentication (signing) and confidentiality (encrypting) simultaneously at a lower computational cost and communication overhead. With the proposition of certificateless public key cryptography (CLPKC), certificateless signcryption (CLSC) scheme has gradually become a research hotspot and attracted extensive attentions. However, many of previous CLSC schemes are constructed based on time-consuming pairing operation, which is impractical for mobile devices with limited computation ability and battery capacity. Although researchers have proposed pairing-free CLSC schemes to solve the issue of efficiency, many of them are in fact still insecure. Therefore, the challenging problem is to keep the balance between efficiency and security in CLSC schemes. In this paper, several existing CLSC schemes are cryptanalyzed and a new CLSC scheme without pairing based on elliptic curve cryptosystem (ECC) is presented. The proposed CLSC scheme is provably secure against indistinguishability under adaptive chosen-ciphertext attack (IND-CCA2) and existential unforgeability under adaptive chosen-message attack (EUF-CMA) resting on Gap Diffie-Hellman (GDH) assumption and discrete logarithm problem in the random oracle model. Furthermore, the proposed scheme resists the ephemeral secret leakage (ESL) attack, public key replacement (PKR) attack, malicious but passive KGC (MPK) attack, and presents efficient computational overhead compared with the existing related CLSC schemes.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.5
• /
• pp.1847-1870
• /
• 2021

With the development of information and communication technologies, especially wireless networks and cell phones, the e-voting system becomes popular as its cost-effectiveness, swiftness, scalability, and ecological sustainability. However, the current e-voting schemes are faced with the problem of privacy leakage and further cause worse vote-buying and voter-coercion problems. Moreover, in large-scale voting, some previous e-voting system encryption scheme with pairing operation also brings huge overhead pressure to the voting system. Thus, it is a vital problem to design a protocol that can protect voter privacy and simultaneously has high efficiency to guarantee the effective implementation of e-voting. To address these problems, our paper proposes an efficient unidirectional proxy re-encryption scheme that provides the re-encryption of vote content and the verification of users' identity. This function can be exactly applied in the e-voting system to protect the content of vote and preserve the privacy of the voter. Our proposal is proven to be CCA secure and collusion resistant. The detailed analysis also shows that our scheme achieves higher efficiency in computation cost and ciphertext size than the schemes in related fields.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.59 no.5
• /
• pp.854-859
• /
• 2010

Distribution Automation System(DAS) is designed to improve operational efficiency by acquisition and control of remote data using its components such as central computation units, communication network and feeder remote terminal units. A conventional human machine interface of the DAS adopts a schematic diagram which is made by drawing power equipments on the geographic information system map. The single line diagram is more useful than the schematic diagram for the main tasks of distribution system operation such as protective relay coordination, service restoration and loss minimization. Since the configuration of the distribution line is changed according to the relocation of the open tie switches, the auto-drawing algorithm based on the connection between the sections and the switches is an essential technique. This paper proposes a new auto-drawing algorithm for a single line diagram of distribution systems based on tertiary tree and collision avoidance method. The feasibility of the proposed algorithm has been testified for various cases using practical distribution system with 12 feeders.