• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권11호
• /
• pp.4291-4310
• /
• 2015

This paper investigates the traffic offloading over unlicensed bands for two-tier multi-mode small cell networks. We formulate this problem as a Stackelberg game and apply a hierarchical learning framework to jointly maximize the utilities of both macro base station (MBS) and small base stations (SBSs). During the learning process, the MBS behaves as a leader and the SBSs are followers. A pricing mechanism is adopt by MBS and the price information is broadcasted to all SBSs by MBS firstly, then each SBS competes with other SBSs and takes its best response strategies to appropriately allocate the traffic load in licensed and unlicensed band in the sequel, taking the traffic flow payment charged by MBS into consideration. Then, we present a hierarchical Q-learning algorithm (HQL) to discover the Stackelberg equilibrium. Additionally, if some extra information can be obtained via feedback, we propose an improved hierarchical Q-learning algorithm (IHQL) to speed up the SBSs' learning process. Last but not the least, the convergence performance of the proposed two algorithms is analyzed. Numerical experiments are presented to validate the proposed schemes and show the effectiveness.

• 한국감성과학회:학술대회논문집
• /
• 한국감성과학회 2002년도 춘계학술대회 논문집
• /
• pp.195-200
• /
• 2002

ART (Adaptive Resonance Theory [1]) neural network and its variations perform non-hierarchical clustering by unsupervised learning. We propose a scheme "arboART" for hierarchical clustering by using several ART1.5-SSS networks. It classifies multidimensional vectors as a cluster tree, and finds features of clusters. The Basic idea of arboART is to use the prototype formed in an ART network as an input to other ART network that has looser distance criteria (Ishihara, et al., [2,3]). By sending prototype vectors made by ART to one after another, many small categories are combined into larger and more generalized categories. We can draw a dendrogram using classification records of sample and categories. We have confirmed its ability using standard test data commonly used in pattern recognition community. The clustering result is better than traditional computing methods, on separation of outliers, smaller error (diameter) of clusters and causes no chaining. This methodology is applied to Kansei evaluation experiment data analysis.

• 한국산학기술학회논문지
• /
• 제8권4호
• /
• pp.768-774
• /
• 2007

본 논문에서는 센서 네트워크의 수명을 길게 하기 위해 각 센서 및 클러스터 헤드에서의 데이터 전송량을 줄이기 위한 방법을 제안한다. 즉, 센서의 에너지 소모를 줄이기 위해 계층적 필터링을 제안한다. 계층적 필터링이란 센서 네트워크를 두 계층으로 나누어 필터링하는 것이다. 1계층 필터링은 클러스터 멤버에서 클러스터 헤드로 데이터를 전송시 필터링을 수행하고, 2계층 필터링은 클러스터 헤드에서 기지국으로 데이터를 전송시 필터링을 수행한다. 이는 일반적으로 필터의 폭을 넓혀 필터링을 많이 하는 것보다 필터링 효율은 증대시키면서 필터링에 따른 데이터 부정확성을 최소한 줄이는 효과를 가진다.

• International Journal of Computer Science & Network Security
• /
• 제23권8호
• /
• pp.121-136
• /
• 2023

Wireless Sensor Networks (WSNs) have many potential applications and unique challenges. Some problems of WSNs are: severe resources' constraints, low reliability and fault tolerant, low throughput, low scalability, low Quality of Service (QoS) and insecure operational environments. One significant solution against mentioned problems is hierarchical and clustering-based multipath routing. But, existent algorithms have many weaknesses such as: high overhead, security vulnerabilities, address-centric, low-scalability, permanent usage of optimal paths and severe resources' consumption. As a result, this paper is proposed an energy-aware, congestion-aware, location-based, data-centric, scalable, hierarchical and clustering-based multipath routing algorithm based on Numerical Taxonomy technique for homogenous WSNs. Finally, performance of the proposed algorithm has been compared with performance of LEACH routing algorithm; results of simulations and statistical-mathematical analysis are showing the proposed algorithm has been improved in terms of parameters like balanced resources' consumption such as energy and bandwidth, throughput, reliability and fault tolerant, accuracy, QoS such as average rate of packet delivery and WSNs' lifetime.

• 정보처리학회논문지C
• /
• 제14C권1호
• /
• pp.45-54
• /
• 2007

가장 위협적인 공격의 한 형태인 DDoS(Distributed Denial of Service) 공격은 다수의 공격 에이전트가 한꺼번에 많은 공격 트래픽을 특정 네트워크 또는 중요한 노드를 공격하는 특성을 갖고 있어 이로 인한 피해 지역 및 정도가 크다는 문제점이 있다. 이에 대한 기존의 많은 연구들은 탐지, 필터링, 추적 등에 집중되어 있고, 특히 피해 네트워크가 계층적인 구조를 갖고 있는 경우 특정 노드의 마비로 인해 하위 노드들의 정상 트래픽 전송이 어려워질 뿐 아니라, 탐지에 대한 다른 노드에 공지 및 추적을 위한 제어 트래픽 전송 또한 어려워 질 수 있다. 이에 본 논문에서는 계층적인 네트워크에서 이에 맞는 계층적인 오버레이를 구성하여, 공격 탐지 시 공지 및 추적을 위한 제어 트래픽을 오버레이를 이용해 전달하며, 공격 에이전트를 완전히 제거하기 전까지 정상적인 트래픽을 우회할 수 있는 DDoS 공격 감내 네트워크 구조를 제안한다. 또한 제안된 방법에서 오버레이 구성에 따른 오버헤드 분석과 공격 탐지 시 빠른 공격 차단 전달의 가능성과 신속성 및 정상 트래픽의 전송의 정도를 시뮬레이션을 통해 분석한다.

• 정보처리학회논문지C
• /
• 제13C권1호
• /
• pp.103-112
• /
• 2006

센서 네트워크에서 센서 노드는 제한된 배터리 용량을 가지기 때문에 에너지 효율적인 라우팅을 통한 네트워크 수명의 최대화가 매우 중요하다. 따라서 많은 라우팅 프로토콜들이 개발되었으며 이는 크게 평면 라우팅과 계층적 라우팅으로 분류된다. 최근 계층적 라우팅 방안에 초점을 맞추고 많은 연구가 이루어지고 있으며 대표적인 방안으로 LEACH가 있다. 본 논문에서는 LEACH의 문제점을 분석하고 이를 보완하기 위한 새로운 라우팅 방안으로 ENTER (ENergy efficient Two-tiEr Routing protocol)를 제안한다. ENTER는 LEACH에서 사용된 분산 알고리즘을 통해 클러스터를 형성하며 클러스터 헤드간에 경로를 형성하여 싱크 노드로 통합된 데이터를 전송함으로써 에너지 손실을 줄이고 네트워크 수명을 증가시킨다. 시뮬레이션을 통해 LEACH와 ENTER의 성능을 비교하였으며, 제안된 ENTER가 더 효율적으로 에너지를 이용함으로써 네트워크 수명이 증가함을 알 수 있었다.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2001년도 춘계학술대회 학술발표 논문집
• /
• pp.187-190
• /
• 2001

Due to the recent proliferation of the internet, it is broadly granted that the necessity of the automatic document categorization has been on the rise. Since it is a heavy time-consuming work and takes too much manpower to process and classify manually, we need a system that categorizes them automatically as their contents. In this paper, we propose the automatic E-mail response system that is based on 2 hierarchical document clustering methods. One is to get the final result from the classifier trained seperatly within each class, after clustering the whole documents into 3 groups so that the first classifier categorize the input documents as the corresponding group. The other method is that the system classifies the most distinct classes first as their similarity, successively. Neural networks have been adopted as classifiers, we have used dendrograms to show the hierarchical aspect of similarities between classes. The comparison among the performances of hierarchical and non-hierarchical classifiers tells us clustering methods have provided the classification efficiency.

• 한국멀티미디어학회논문지
• /
• 제24권4호
• /
• pp.550-557
• /
• 2021

Image steganalysis is a technique for detecting images with steganographic algorithms applied, called stego images. With state-of-the-art CNN-based steganalysis methods, we can detect stego images with high accuracy, but it is not possible to know which steganographic algorithm is used. Identifying stego images is essential for extracting embedded data. In this paper, as the first step for extracting data from stego images, we propose a hierarchical CNN structure for senary classification of steganographic algorithms. The hierarchical CNN structure consists of multiple CNN networks which are trained to classify each steganographic algorithm and performs binary or ternary classification. Thus, it classifies multiple steganogrphic algorithms hierarchically and stepwise, rather than classifying them at the same time. In experiments of comparing with several conventional methods, including those of classifying multiple steganographic algorithms at the same time, it is verified that using the hierarchical CNN structure can greatly improve the classification accuracy.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권7호
• /
• pp.3369-3385
• /
• 2019

There are various research challenges in vehicular ad hoc networks (VANETs) that need to be focused until an extensive deployment of it becomes conceivable. Design and development of a scalable routing algorithm for VANETs is one of the critical issue due to frequent path disruptions caused by the vehicle's mobility. This study aims to provide a novel road-aware routing protocol for vehicular networks named as Two-level hierarchical Hybrid Road-Aware (THERA) routing for vehicular ad hoc networks. The proposed protocol is designed explicitly for inter-vehicle communication. In THERA, roads are distributed into non-overlapping road segments to reduce the routing overhead. Unlike other protocols, discovery process does not flood the network with packet broadcasts. Instead, THERA uses the concept of Gateway Vehicles (GV) for the discovery process. In addition, a route between source and destination is flexible to changing topology, as THERA only requires road segment ID and destination ID for the communication. Furthermore, Road-Aware routing reduces the traffic congestion, bypasses the single point of failure, and facilitates the network management. Finally yet importantly, this paper also proposes a probabilistical model to estimate a path duration for each road segment using the highway mobility model. The flexibility of the proposed protocol is evaluated by performing extensive simulations in NS3. We have used SUMO simulator to generate real time vehicular traffic on the roads of Gangnam, South Korea. Comparative analysis of the results confirm that routing overhead for maintaining the network topology is smaller than few previously proposed routing algorithms.

• 정보처리학회논문지C
• /
• 제10C권3호
• /
• pp.355-366
• /
• 2003

오늘날 무선 이동 망의 빠른 기술 발전으로 인해 이동 인터넷 서비스의 시장이 급속히 팽창하고 있다. 본 논문에서는 무선 이동 망에서 이동 호스트의 QoS 보장을 위한 효율적 자원 예약 프로토콜인 Proxy MRSVP(PMRSVP)를 제안한다. PMRSVP는 수정된 지역 위치등록을 이용하여 무선 이동 망에서 기존의 Best Effort 서비스의 개선책으로 제시되고 있는 프로토콜들의 문제점인 메시지 과다 발생을 줄인다. 또한, Mobile Agent(MA)들이 Corresponding Host(CH)의 프록시 역할을 하도록 함으로써 신호 메시지 발생 양과 자원 예약 비용이 MRSVP와 Hierarchical MRSVP(HMRSVP) 방법에 비해 낮다는 것을 보인다. 본 논문의 우수성을 보이기 위해 MRSVP와 HMRSVP 그리고 PMRSVP를 도메인내(intradomain)와 도메인간(interdomain)의 위치 등록 비용을 포함한 자원 예약 비용을 비교한다.