• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.6
• /
• pp.1357-1378
• /
• 2013

Energy constraint of wireless sensor networks makes energy saving and prolonging the network lifetime become the most important goals of routing protocols. In this paper, we propose an Energy Efficient Chain-based Routing Protocol (EECRP) for wireless sensor networks to minimize energy consumption and transmission delay. EECRP organizes sensor nodes into a set of horizontal chains and a vertical chain. Chain heads are elected based on the residual energy of nodes and distance from the header of upper level. In each horizontal chain, sensor nodes transmit their data to their own chain head based on chain routing mechanism. EECRP also adopts a chain-based data transmission mechanism for sending data packets from the chain heads to the base station. The simulation results show that EECRP outperforms LEACH, PEGASIS and ECCP in terms of network lifetime, energy consumption, number of data messages received at the base station, transmission delay and especially energy${\times}$delay metric.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.6 no.7
• /
• pp.1874-1893
• /
• 2012

An approach to detect abnormal activities based on reputations created individually by each node is vulnerable to a false accusation since intrusion detection in ad-hoc networks is done in a distributed and cooperative manner. Detection of false accusation is considered important because the efficiency or survivability of the network can be degraded severely if normal nodes were excluded from the network by being considered as abnormal ones in the intrusion detection process. In this paper, we propose an improved reputation-based intrusion detection technique to efficiently detect and manage false accusations in ad-hoc networks. Additionally, we execute simulations of the proposed technique to analyze its performance and feasibility to be implemented in a real environment.

• Journal of Information Processing Systems
• /
• v.6 no.3
• /
• pp.403-412
• /
• 2010

Multi-Relay Networks should accommodate mobile users of various speeds. The cellular system should meet the minimum residency time requirements for handover calls while considering an efficient use of available channels. In this paper, we design speed-sensitive handover under dynamic hierarchical cellular systems, in which mobile users are classified according to the mean speed of mobile users and each class has its cellular layer. In order to meet the minimum residency time, the cell size of each cellular layer is dynamically determined depending on the distributions of mean speeds of mobile users. Since the speed-dependent non-preferred cell can provide a secondary resource, overflow and take-back schemes are adopted in the system. We develop analytical models to study the performance of the proposed system, and show that the optimal cell size improves the blocking probability.

• ETRI Journal
• /
• v.24 no.5
• /
• pp.381-390
• /
• 2002

We analyze the performance of a telecommunications management network (TMN) system using models of networks of queues, Jackson's theorem, and simulation. TMN systems for managing public asynchronous transfer mode (ATM) networks generally have a four-level hierarchical structure consisting of a network management system, a few element management systems (EMSs), and several pairs of agents and ATM switches. We construct a Jackson's queuing network and present formulae to calculate its performance measures: distributions of queue lengths and waiting times, mean message response time, and maximum throughput. We perform a numerical analysis and a simulation analysis and compare the results.

• Proceedings of the KIEE Conference
• /
• 1995.07b
• /
• pp.959-961
• /
• 1995

The objective of most image proccessing applications is to extract meaningful information from one or more pictures. It is accomplished efficiently using neural networks, which is used in image classification and image recognition. In neural networks, background and meaningful information are processed with same weight in input layer. In this paper, we propose the image classification method using neural networks, especially EBP(Error Back Propagation). Preprocessing is needed. In preprocessing, background is compressed and meaningful information is emphasized. We use the quadtree approach, which is a hierarchical data structure based on a regular decomposition of space.

• Journal of information and communication convergence engineering
• /
• v.10 no.3
• /
• pp.276-283
• /
• 2012

Nowadays, wireless sensor networks (WSNs) comprise a tremendously growing infrastructure for monitoring the physical or environmental conditions of objects. WSNs pose challenges to mitigating energy dissipation by constructing a reliable and energy saving network. In this paper, we propose a novel network construction and routing method by defining three different duties for sensor nodes, that is, node gateways, cluster heads, and cluster members, and then by applying a hierarchical structure from the sink to the normal sensing nodes. This method provides an efficient rationale to support the maximum coverage, to recover missing data with node mobility, and to reduce overall energy dissipation. All this should lengthen the lifetime of the network significantly.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.11
• /
• pp.5427-5445
• /
• 2019

While deep neural networks have achieved remarkable performance in representation learning, a huge amount of labeled training data are usually required by supervised deep models such as convolutional neural networks. In this paper, we propose a new representation learning method, namely generative adversarial networks (GAN) based bagging deep convolutional autoencoders (GAN-BDCAE), which can map data to diverse hierarchical representations in an unsupervised fashion. To boost the size of training data, to train deep model and to aggregate diverse learning machines are the three principal avenues towards increasing the capabilities of representation learning of neural networks. We focus on combining those three techniques. To this aim, we adopt GAN for realistic unlabeled sample generation and bagging deep convolutional autoencoders (BDCAE) for robust feature learning. The proposed method improves the discriminative ability of learned feature embedding for solving subsequent pattern recognition problems. We evaluate our approach on three standard benchmarks and demonstrate the superiority of the proposed method compared to traditional unsupervised learning methods.

• 제어로봇시스템학회:학술대회논문집
• /
• 1994.10a
• /
• pp.498-503
• /
• 1994

The paper demonstrates that holon networks can be used effectively for identification of nonlinear dynamical systems. The emphasis of the paper is on modeling of complicated systems which have a great deal of uncertainty and unknown interactions between their elements and parameters. The concept of applying a quantitative model building, for example, to environmental or ecological systems is not new. In a previous paper we presented a holon network model as an another alternative to quantitative modeling. Holon networks have a hierarchical construction where each level of hierarchy consists of networks with reciprocal actions among their elements. The networks are able to evolve by self-organizing their structure and adapt their parameters to environments. This was achieved by an autonomous decentralized adaptation algorithm. In this paper we propose a new emergent evolution algorithm. In this algorithm the initial holon networks consists of only a few elements and it grows gradually with each new observation in order to fit their function to the environment. Some examples show that this algorithm can lead to a network structure which has sufficient flexibility and adapts well to the environment.

• Journal of KIISE:Software and Applications
• /
• v.33 no.10
• /
• pp.877-885
• /
• 2006

Recently it is proposed that the Bayesian networks used as conversational agent for topic inference is useful but the Bayesian networks require much time to model, and the Bayesian networks also have to be modified when the scripts, the database for conversation, are added or modified and this hinders the scalability of the agent. This paper presents a method to improve the scalability of the agent by constructing the Bayesian network from scripts automatically. The proposed method is to model the structure of Bayesian networks hierarchically and to utilize Noisy-OR gate to form the conditional probability distribution table (CPT). Experimental results with ten subjects confirm the usefulness of the proposed method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.1
• /
• pp.145-164
• /
• 2014

Network virtualization provides a promising tool to allow multiple heterogeneous virtual networks to run on a shared substrate network simultaneously. A long-standing challenge in network virtualization is the Virtual Network Embedding (VNE) problem: how to embed virtual networks onto specific physical nodes and links in the substrate network effectively. Recent research presents several heuristic algorithms that only consider single topological attribute of networks, which may lead to decreased utilization of resources. In this paper, we introduce six complementary characteristics that reflect different topological attributes, and propose three topology-aware VNE algorithms by leveraging the respective advantages of different characteristics. In addition, a new KS-core decomposition algorithm based on two characteristics is devised to better disentangle the hierarchical topological structure of virtual networks. Due to the overall consideration of topological attributes of substrate and virtual networks by using multiple characteristics, our study better coordinates node and link embedding. Extensive simulations demonstrate that our proposed algorithms improve the long-term average revenue, acceptance ratio, and revenue/cost ratio compared to previous algorithms.