• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1305-1308
• /
• 2003

A new technique is presented to construct a predictive model of plasma etch process. This was accomplished by combining a backpropagation neural network (BPNN) and a genetic algorithm (GA). The predictive model constructed in this way is referred to as a GA-BPNN. The GA played a role of controlling training factors simultaneously. The training factors to be optimized are the hidden neuron, training tolerance, initial weight magnitude, and two gradients of bipolar sigmoid and linear functions. Each etch response was optimized separately. The proposed scheme was evaluated with a set of experimental plasma etch data. The etch process was characterized by a $2^3$ full factorial experiment. The etch responses modeled are aluminum (A1) etch rate, silica profile angle, A1 selectivity, and dc bias. Additional test data were prepared to evaluate model appropriateness. The GA-BPNN was compared to a conventional BPNN. Compared to the BPNN, the GA-BPNN demonstrated an improvement of more than 20% for all etch responses. The improvement was significant in the case of A1 etch rate.

• Journal of Korean Institute of Industrial Engineers
• /
• v.10 no.2
• /
• pp.17-27
• /
• 1984

This paper is concerned with a branched pipe network system which transports some fluids or gas from multiple sources to multiple demand nodes. A nonlinear programming model is proposed for determining junction locations simultaneously with selection of pipe sizes and pump capacities such that the capital and operating costs of the system are minimized over a given planning horizon. To solve the model, a hierarchical decomposition method is developed with the junction location being the primary variable. With some values fixed for the primary, the other decision variables are found by linear programming. Then, using the postoptimality analysis of LP, junction locations are adjusted. We repeat this process until an optimum is approached. A simple example of designing a water distribution network is solved to illustrate the optimization procedure developed.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.5
• /
• pp.60-72
• /
• 1996

In this paper, a enw model-base three-dimensional (3-D) object recognition mehtod using hopfield network is proposed. To minimize deformation of feature values on 3-D rotation, we select 3-D shape features and 3-D relational features which have rotational invariant characteristics. Then these feature values are normalized to have scale invariant characteristics, also. The input features are matched with model features by optimization process of hopjfield network in the form of two dimensional arrayed neurons. Experimental results on object classification and object matching with the 3-D rotated, scale changed, an dpartial oculued objects show good performance of proposed method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.6 no.8
• /
• pp.1913-1925
• /
• 2012

In order to efficiently utilize limited radio resources, resource allocation schemes in OFDMA-based wireless networks have gained intensive attention recently. Instead of improving the throughput performance, the utility is adopted as the metric for resource allocation, which provides reasonable methods to build up the relationship between user experience and various quality-of-service (QoS) metrics. After formulating the optimization problem by using a weighted bipartite graph, a modified bipartite matching method is proposed to find a suboptimal solution for the resource allocation problem in OFDMA-based wireless systems with feasible computational complexity. Finally, simulation results are presented to validate the effectiveness of the proposed method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.2
• /
• pp.932-945
• /
• 2018

Everything in the world is becoming connected and interactive due to the Internet. The future of interactive smart environments such as smart cities, smart industries, or smart farms demand high network bandwidth, high network flexibility, and self-organization systems without costly hardware upgrades, and they provide a sustainable, scalable, and replicable smart environment backbone infrastructure. This paper presents a new Hybrid Software-Defined architecture for integrating Internet-of-Things technologies that are essential technologies for smart environments. It combines a software-defined networking infrastructure and a real-time distributed network framework with an advanced optimization to enable self-configuration, self-management, and self-adaption for providing seamless communication and efficiently managing a vast number of smart heterogeneous devices.

• Journal of Korea Society of Industrial Information Systems
• /
• v.7 no.5
• /
• pp.147-153
• /
• 2002

In this paper, a substrate network model to be used with BSIM3 MOSFET model for submicron MOSFETs in giga hertz frequencies and its direct parameter extraction with physically meaningful values are proposed. The proposed substrate network model includes a conventional resistance and single inductance originated from ring-type substrate contacts around active devices. Model parameters are extracted from S-parameter data measured from common-bulk configured MOS transistors with floating gate and use where needed without any optimization process. The proposed modeling technique has been applied to various-sized MOS transistors. The substrate model has been validated for frequency up to 300Hz.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.35 no.2
• /
• pp.53-69
• /
• 2010

This paper proposes an ant-based routing algorithm, Ant System-Routing in wireless Senor Networks(AS-RSN), for wireless sensor networks. Using a transition rule in Ant System, sensors can spread data traffic over the whole network to achieve energy balance, and consequently, maximize the lifetime of sensor networks. The transition rule advances one of the original Ant System by re-defining link cost which is a metric devised to consider energy-sufficiency as well as energy-efficiency. This metric gives rise to the design of the AS-RSN algorithm devised to balance the data traffic of sensor networks in a decentralized manner and consequently prolong the lifetime of the networks. Therefore, AS-RSN is scalable in the number of sensors and also robust to the variations in the dynamics of event generation. We demonstrate the effectiveness of the proposed algorithm by comparing three existing routing algorithms: Direct Communication Approach, Minimum Transmission Energy, and Self-Organized Routing and find that energy balance should be considered to extend lifetime of sensor network and increase robustness of sensor network for diverse event generation patterns.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.34 no.1
• /
• pp.33-41
• /
• 2011

This paper considers a topological optimization of a network design with mean packet delay and node connectivity constraints. The objective is to find the topological layout of links, at minimal cost. This Problem is known to be NP-hard. To efficiently solve the problem, a scatter search algorithm is proposed. An illustrative example is used to explain and test the proposal approach. Experimental results show evidence that the proposal approach performs more efficiently for finding a good solution or near optimal solution in comparison with a genetic approach.

• Journal of Korean Institute of Industrial Engineers
• /
• v.31 no.1
• /
• pp.10-19
• /
• 2005

The minimum spanning tree (MST) problem is one of the traditional optimization problems. Unlike the MST, the degree constrained minimum spanning tree (DCMST) of a graph cannot, in general, be found using a polynomial time algorithm. So, finding the DCMST of a graph is a well-known NP-hard problem of importance in communications network design, road network design and other network-related problems. So, it seems to be natural to use evolutionary algorithms for solving DCMST. Especially, when applying an evolutionary algorithm to spanning tree problems, a representation and search operators should be considered simultaneously. This paper introduces a new tree representation scheme and a genetic operator for solving combinatorial tree problem using evolutionary algorithms. We performed empirical comparisons with other tree representations on several test instances and could confirm that the proposed method is superior to other tree representations. Even it is superior to edge set representation which is known as the best algorithm.

• Journal of Korean Institute of Industrial Engineers
• /
• v.4 no.2
• /
• pp.97-105
• /
• 1978

A mathematical model is developed in order to describe the network configuration and heating distribution to a Central District Heating System for an Urban Residential Community. The purpose of using this model is to optimize operating costs and to distribute heat to the Residential Community efficiently. In particular, because of the inherent nonlinearity and dual optimization of the problem a dyamic programming approach is taken. It is turned out that the optimal cost of the system is a strong non-linear function of the network. In particular, it is found that increasing N, the number of houses, may not necessarily imply increased costs. It is felt that past failure of producing economical systems may be due to the improper attention given to the network.