• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.40 no.8
• /
• pp.782-791
• /
• 1991

In the fields of pattern recognition such as speech recognition, several new techniques using Artifical Neural network Models have been proposed and implemented. In particular, the Multilayer Perception Model has been shown to be effective in static speech pattern recognition. But speech has dynamic or temporal characteristics and the most important point in implementing speech recognition systems using Artificial Neural Network Models for continuous speech is the learning of dynamic characteristics and the distributed cues and contextual effects that result from temporal characteristics. But Recurrent Multilayer Perceptron Model is known to be able to learn sequence of pattern. In this paper, the results of applying the Recurrent Model which has possibilities of learning tedmporal characteristics of speech to phoneme recognition is presented. The test data consist of 144 Vowel+ Consonant + Vowel speech chains made up of 4 Korean monothongs and 9 Korean plosive consonants. The input parameters of Artificial Neural Network model used are the FFT coefficients, residual error and zero crossing rates. The Baseline model showed a recognition rate of 91% for volwels and 71% for plosive consonants of one male speaker. We obtained better recognition rates from various other experiments compared to the existing multilayer perceptron model, thus showed the recurrent model to be better suited to speech recognition. And the possibility of using Recurrent Models for speech recognition was experimented by changing the configuration of this baseline model.

• Journal of the Korean Society for Aviation and Aeronautics
• /
• v.32 no.3
• /
• pp.44-52
• /
• 2024

The go-around procedure plays a crucial role in aviation safety, allowing pilots to abort unsafe landings and attempt a new approach. While existing studies have primarily focused on predicting the onset of go-arounds, relatively little attention has been paid to evaluating aircraft stability and performance after a go-around has been initiated. This study aims to address this gap by systematically assessing the dynamic response and stability of aircraft following a go-around using Quick Access Recorder (QAR) data. The methodology involves classifying go-around events into 'near-ground' and 'at-altitude' categories, and analyzing changes in pitch, descent rate, engine performance, and environmental factors after the initiation of the go-around to evaluate its stability and efficiency. The XGBoost machine learning algorithm is employed to model the aircraft's response post go-around and to predict stability across various go-around scenarios. The findings from this study provide insights that can enhance the safety and efficiency of go-around procedures through systematic analysis of QAR data, contributing to improvements in operational protocols and pilot training programs.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.11
• /
• pp.4244-4274
• /
• 2021

The utilization of UAVs in various fields has led to the development of flying ad hoc network (FANET) technology. In a network environment with highly dynamic topology and frequent link changes, the traditional routing technology of FANET cannot satisfy the new communication demands. Traditional routing algorithm, based on geographic location, can "fall" into a routing hole. In view of this problem, we propose a geolocation routing protocol based on multi-agent reinforcement learning, which decreases the packet loss rate and routing cost of the routing protocol. The protocol views each node as an intelligent agent and evaluates the value of its neighbor nodes through the local information. In the value function, nodes consider information such as link quality, residual energy and queue length, which reduces the possibility of a routing hole. The protocol uses global rewards to enable individual nodes to collaborate in transmitting data. The performance of the protocol is experimentally analyzed for UAVs under extreme conditions such as topology changes and energy constraints. Simulation results show that our proposed QLGR-S protocol has advantages in performance parameters such as throughput, end-to-end delay, and energy consumption compared with the traditional GPSR protocol. QLGR-S provides more reliable connectivity for UAV networking technology, safeguards the communication requirements between UAVs, and further promotes the development of UAV technology.

• Proceedings of the IEEK Conference
• /
• 1999.06a
• /
• pp.1033-1036
• /
• 1999

This paper proposes an efficient fault diagnosis for digital circuits using multilayer neural networks. The efficient learning algorithm is also proposed for the multilayer neural network, which is combined the steepest descent for high-speed optimization and the dynamic tunneling for global optimization. The fault-diagnosis system using the multilayer neural network of the proposed algorithm has been applied to the parity generator circuit. The simulation results shows that the proposed system is higher convergence speed and rate, in comparision with system using the backpropagation algorithm based on the gradient descent.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.11
• /
• pp.3780-3797
• /
• 2022

The estimation of radio map (RM) is a fundamental and critical task for the network planning and optimization performance of mobile communication. In this paper, a RM estimation method is proposed based on a deep dual learning structure. This method can simultaneously and accurately reconstruct the urban building map (UBM) and estimate the RM of the whole cell by only part of the measured reference signal receiving power (RSRP). Our proposed method implements UBM reconstruction task and RM estimation task by constructing a dual U-Net-based structure, which is named RadioCycle. RadioCycle jointly trains two symmetric generators of the dual structure. Further, to solve the problem of interference negative transfer in generators trained jointly for two different tasks, RadioCycle introduces a dynamic weighted averaging method to dynamically balance the learning rate of these two generators in the joint training. Eventually, the experiments demonstrate that on the UBM reconstruction task, RadioCycle achieves an F1 score of 0.950, and on the RM estimation task, RadioCycle achieves a root mean square error of 0.069. Therefore, RadioCycle can estimate both the RM and the UBM in a cell with measured RSRP for only 20% of the whole cell.

• Journal of Intelligence and Information Systems
• /
• v.3 no.1
• /
• pp.47-68
• /
• 1997

This paper discusses a method of modified Elman network(1990) for nonlinear predictions and its a, pp.ication to forecasting daily exchange rate returns. The method consists of two stages that take advantages of both time domain filter and modified feedback networks. The first stage straightforwardly employs the filtering technique to remove extreme noise. In the second stage neural networks are designed to take the feedback from both hidden-layer units and the deviation of outputs from target values during learning. This combined feedback can be exploited to transfer unconsidered information on errors into the network system and, consequently, would improve predictions. The method a, pp.ars to dominate linear ARMA models and standard dynamic neural networks in one-step-ahead forecasting exchange rate returns.

• Journal of Korean Institute of Industrial Engineers
• /
• v.14 no.1
• /
• pp.91-101
• /
• 1988

Break-even analysis is a simple and useful tool in decisions and planning activities though its use is somewhat limited to short-term analysis. The subject is discussed in the fields of engineering economics, production management, cost and managerial accounting, finance, marketing, and so on. Conventional break-even analysis suits the case of stable price and low interest rate. In this paper, we try to overcome the limit by considering following factors, namely, time value of money, depreciation, tax, and capital gains. Also, considering learning effect, we increase applicability to a new project which raises certain changes such as a replacement of production process, an employee turnover, etc. Thus, we suggest a model which has a dynamic break-even quantity per period for the project. Furthermore, we examine the effect of inflation in break-even analysis.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.10
• /
• pp.3394-3408
• /
• 2014

Conventional mobile state (MS) and base station (BS) association based on average signal strength often results in imbalance of cell load which may require more powerful processor at BSs and degrades the perceived transmission rate of MSs. To deal with this problem, a Markov decision process (MDP) for load balancing in a multi-cell system with multi-carriers is formulated. To solve the problem, exploiting Sarsa algorithm of on-line learning type [12], ${\alpha}$-controllable load balancing algorithm is proposed. It is designed to control tradeoff between the cell load deviation of BSs and the perceived transmission rates of MSs. We also propose an ${\varepsilon}$-differential soft greedy policy for on-line learning which is proven to be asymptotically convergent to the optimal greedy policy under some condition. Simulation results verify that the ${\alpha}$-controllable load balancing algorithm controls the behavior of the algorithm depending on the choice of ${\alpha}$. It is shown to be very efficient in balancing cell loads of BSs with low ${\alpha}$.

• ETRI Journal
• /
• v.35 no.6
• /
• pp.1048-1057
• /
• 2013

In recent years, augmented reality (AR) technologies have been the subject of great interest among many communities. In education applications, old-fashioned materials (or textbooks) are still used, despite remarkable AR developments in the industrial area. We present an AR system for education. Our system consists of an authoring tool that can be used to create educational content, a viewer that plays that content, and an engine to manage the tool and viewer. In our system, a marker unit recognizes a marker printed on a plane or a cubic plane by adaptively adjusting the threshold to have an excellent recognition rate in diverse environments and acquires corresponding data of the marker. Based on the system, we test 142 elementary school students for increased educational benefits using our system.

• Proceedings of the KOSOMBE Conference
• /
• v.1996 no.11
• /
• pp.199-202
• /
• 1996

In this paper, we suggested the combination of HMM(Hidden Markov Model) and MLP (Multi-Layer Perceptron) with GA(genetic algorithm) for a recognition of EMG signals. To describe EMG signal's dynamic properties, HMM algorithm was adapted and due to its outstanding abilities in static signal classification MLP was connected as a real processor. We also used GA( Genetic Algorithm) for improving MLP's learning rate. Experimental results showed that the suggested classifier gave higher EMG signal recognition rates with faster learning time than other one.