• Journal of the Korea Society of Computer and Information
• /
• v.17 no.5
• /
• pp.59-70
• /
• 2012

RSA-CRT is one of the commonly used techniques to speedup RSA operation. Since RSA-CRT performs its operations based on the modulus of two private primes, it is about four times faster than RSA. In RSA, the two primes are normally thrown away after generating the public key pair. However, in RSA-CRT, the two primes are directly used in RSA operations. This led to hardware fault attacks which can be used to factor the public modulus. The most common way to counter these attacks is based on error propagation. In these schemes, all the outputs of RSA are affected by the infected error which makes it difficult for an adversary to use the output to factor the public modulus. However, the error propagation has sequentialized the RSA operation. Moreover, these schemes have been found to be still vulnerable to hardware fault attacks. In this paper, we propose two new RSA-CRT schemes which are both resistant to hardware fault attack and support parallel execution: one uses common modulus and the other one perform operations in each prime modulus. Both proposed schemes takes about a time equal to two exponentiations to complete the RSA operation if parallel execution is fully used and can protect the two private primes from hardware fault attacks.

• ETRI Journal
• /
• v.43 no.3
• /
• pp.483-510
• /
• 2021

In computational biology, desired patterns are searched in large text databases, and an exact match is preferable. Classical benchmark algorithms obtain competent solutions for pattern matching in O (N) time, whereas quantum algorithm design is based on Grover's method, which completes the search in $O(\sqrt{N})$ time. This paper briefly explains existing quantum algorithms and defines their processing limitations. Our initial work overcomes existing algorithmic constraints by proposing the quantum-based combined exact (QBCE) algorithm for the pattern-matching problem to process exact patterns. Next, quantum random access memory (QRAM) processing is discussed, and based on it, we propose the QRAM processing-based exact (QPBE) pattern-matching algorithm. We show that to find all t occurrences of a pattern, the best case time complexities of the QBCE and QPBE algorithms are $O(\sqrt{t})$ and $O(\sqrt{N})$, and the exceptional worst case is bounded by O (t) and O (N). Thus, the proposed quantum algorithms achieve computational speedup. Our work is proved mathematically and validated with simulation, and complexity analysis demonstrates that our quantum algorithms are better than existing pattern-matching methods.

• Journal of Information Processing Systems
• /
• v.5 no.2
• /
• pp.105-116
• /
• 2009

To solve the general problems surrounding the application of genetic algorithms in stereo matching, two measures are proposed. Firstly, the strategy of simplified population-based incremental learning (PBIL) is adopted to reduce the problems with memory consumption and search inefficiency, and a scheme for controlling the distance of neighbors for disparity smoothness is inserted to obtain a wide-area consistency of disparities. In addition, an alternative version of the proposed algorithm, without the use of a probability vector, is also presented for simpler set-ups. Secondly, programmable graphics-hardware (GPU) consists of multiple multi-processors and has a powerful parallelism which can perform operations in parallel at low cost. Therefore, in order to decrease the running time further, a model of the proposed algorithm, which can be run on programmable graphics-hardware (GPU), is presented for the first time. The algorithms are implemented on the CPU as well as on the GPU and are evaluated by experiments. The experimental results show that the proposed algorithm offers better performance than traditional BMA methods with a deliberate relaxation and its modified version in terms of both running speed and stability. The comparison of computation times for the algorithm both on the GPU and the CPU shows that the former has more speed-up than the latter, the bigger the image size is.

• Annual Conference of KIPS
• /
• 2014.04a
• /
• pp.206-209
• /
• 2014

Control message exchange is major job for cognitive radio to exist and use spectrum opportunistically. For this control message exchange they need a common control channel (CCC). Once this channel is affected by a primary user, communication stops until new CCC is setup. This takes substantial time and if they could not get free channel, this halt continues for long time. To prevent such cease of communication, we propose a combination of two networks, namely WLAN and UWB, to let the communication continue. In our proposed idea if the CCC of a certain CR in WLAN is affected, the CR changes its network from WLAN to UWB and keeps the communication because UWB cannot be affected by PU. In the proposed idea every cognitive radio has two transceivers; one for the overlay network (WLAN) and another UWB network. If a primary user is detected in the spectrum of a cognitive radio, it continues exchanging control messages under the UWB network and in parallel negotiates for a new CCC using the WLAN network. This idea solves the communication interruption until new CCC is setup.

• Annual Conference of KIPS
• /
• 2020.11a
• /
• pp.57-60
• /
• 2020

순위다중패턴매칭문제는 길이가 n인 텍스트 T와 패턴들의 집합 P' = {P₁,P₂…,P_k}가 주어졌을 때, P'에 속하는 패턴들과 상대적인 순위가 일치하는 T의 모든 부분문자열들의 위치를 찾는 문제이다. P'에서 가장 짧은 패턴의 길이가 m, 가장 긴 패턴의 길이를 $\bar{m}$, 모든 패턴들의 길이의 합을 M, q개의 연속된 문자들을 q-그램이라 할 때, 기존에 텍스트의 핑거프린트를 이용하여 순위다중패턴매칭문제를 $O(q!+nqlogq+Mlog\bar{m}+nM)$ 시간에 해결하는 알고리즘이 제시되었다. 본 논문에서는 텍스트의 핑거프린트를 활용하여 O(max(q!,M,n))개의 스레드를 이용하여 순위다중패턴매칭문제를 평균적으로 $O(\bar{m}+qlogq+n/q!)$ 시간에 해결하는 병렬 구현 방법을 제시한다. 실험 결과, n = 1,000,000, k = 1,000, m = 5, q = 3일 때, 본 논문에서 제시하는 병렬 구현 방법은 기존의 순차 알고리즘보다 약 19.8배 빠르게 수행되었다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10c
• /
• pp.214-217
• /
• 2006

XML 문서의 검색을 위한 질의 언어인 XQuery는 다양한 데이터 소스로부터 가져온 고유한 구조를 가진 질의 결과로 구성할 수 있도록 설계되어 XML질의 언어의 표준이 되었다. XQuery를 이용해 특별히, 분산 환경에서 다중 XML문서 를 대상으로 하는 통합 질의의 경우, 질의 처리 계획을 결정하는 것은 처리 효율과 직결된다. 따라서 질의 처리 계획을 결정하는 요소 중 하나인 조인 처리 방법의 연구는 중요하다. 그러나 통합 질의에서 조인구조를 기준으로 단일 XML문서에 대한 질의 처리방법을 결정하는 것은 쉽지 않다. 본 논문에서는 분산환경에서 다중 XML문서를 대상으로 하는 조인을 포함한 다양한 통합 질의를 대상으로 실험을 통해 병렬처리 방법과 순차처리 방법 그리고 두 가지 처리방법을 조합한 하이브리드 방법을 적용하여 처리 시간을 비교 분석하고, 다중 문서에 대한 효율적인 조인방법과 순서를 모색한다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2011.07a
• /
• pp.396-398
• /
• 2011

본 논문은 두 대의 카메라와 한 대의 프로젝터로 구성된 Pro-cam시스템을 이용하여, 출력된 패턴 영상을 카메라로 촬영하고 이를 기반으로 Depth Map을 계산하는 모듈의 실시간 처리를 위한 GPU기반 병렬처리 기법을 제안한다. 입력받은 영상으로부터 구조광의 패턴을 해석하고, Depth Map을 계산하기 위해서, Dynamic pattern decoding하는 과정은 프로젝터의 패턴영상과 촬영된 카메라 패턴영상 간의 관계를 반복적으로 비교하므로, 이를 GPU 프로그래밍을 이용하여 병렬 처리를 통해 고속화하였다. 결과적으로 본 논문에서는 기존 CPU에서 수행했던 속도에 비해 약 18배정도 속도를 개선 할 수 있었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.1
• /
• pp.340-367
• /
• 2018

Storage services integration can be done for achieving high availability, improving data access performance and scalability while preventing vendor lock-in. However, multiple services environment management and interoperability have become a critical issue as a result of service architectures and communication interfaces heterogeneity. Storage federation model provides the integration of multiple heterogeneous and self-sufficient storage systems with a single control point and automated decision making about data distribution. In order to integrate diverse heterogeneous storage services into a single storage pool, we are proposing a storage service federation framework named BoxBroker. Moreover, an automated decision model based on a policy-driven data distribution algorithm and a service evaluation method is proposed enabling BoxBroker to make optimal decisions. Finally, a demonstration of our proposal capabilities is presented and discussed.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.6
• /
• pp.505-514
• /
• 2020

Since a large amount of training data is typically needed to train Deep Neural Networks (DNNs), a parallel training approach is required to train the DNNs. The Stochastic Gradient Descent (SGD) algorithm is one of the most widely used methods to train the DNNs. However, since the SGD is an inherently sequential process, it requires some sort of approximation schemes to parallelize the SGD algorithm. In this paper, we review various efforts on parallelizing the SGD algorithm, and analyze the computational overhead, communication overhead, and the effects of the approximations.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2002.11a
• /
• pp.405-408
• /
• 2002

Cluster is the cost-effective parallel processing environment, and consists of the off-the-shelf computers connected by the computer networks. The characteristics of cluster are the node heterogeneity, the variety of node load, and the variety of network load. Because these characteristics influence the performance of parallel program executions, the load sharing for cluster is important, and by using the proper load sharing strategy, we can reduce the execution time of parallel programs. In this paper, we propose modified GSS algorithm, aGSS. In the proposed load sharing algorithms aGSS, the size of tasks are decided using the BogoMTPS of node. From the result of out experiments, we conclude that the proposed aGSS algorithm is effective in the heterogeneous cluster.