• Phonetics and Speech Sciences
• /
• v.8 no.3
• /
• pp.31-38
• /
• 2016

Speech inversion (acoustic-to-articulatory mapping) is not a trivial problem, despite the importance, due to the highly non-linear and non-unique nature. This study aimed to investigate the performance of Deep Neural Network (DNN) compared to that of traditional Artificial Neural Network (ANN) to address the problem. The Wisconsin X-ray Microbeam Database was employed and the acoustic signal and articulatory pellet information were the input and output in the models. Results showed that the performance of ANN deteriorated as the number of hidden layers increased. In contrast, DNN showed lower and more stable RMS even up to 10 deep hidden layers, suggesting that DNN is capable of learning acoustic-articulatory inversion mapping more efficiently than ANN.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.8
• /
• pp.3701-3727
• /
• 2016

Recently Service Function Chaining (SFC) is promising to innovate the network service mode in modern networks. However, a feasible implementation of SFC is still difficult due to the need to achieve functional equivalence with traditional modes without sacrificing performance or increasing network complexity. In this paper, we present a configurable network service chaining (CNSC) mechanism to provide services for network traffics in a flexible and optimal way. Firstly, we formulate the problem of network service chaining and design an effective service chain construction framework based on integrating software-defined networking (SDN) with network functions virtualization (NFV). Then, we model the service path computation problem as an integer liner optimization problem and propose an algorithm named SPCM to cooperatively combine service function instances with a network utility maximum policy. In the procedure of SPCM, we achieve the service node mapping by defining a service capacity matrix for substrate nodes, and work out the optimal link mapping policies with segment routing. Finally, the simulation results indicate that the average request acceptance ratio and resources utilization ratio can reach above 85% and 75% by our SPCM algorithm, respectively. Upon the prototype system, it is demonstrated that CNSC outperforms other approaches and can provide flexible and scalable network services.

• Proceedings of the Korean Institute Of Construction Engineering and Management
• /
• 2004.11a
• /
• pp.561-564
• /
• 2004

Crack sealing automation machines' have been continually developed since the early 1990's because of the effectiveness of crack sealing that would be able to improve safety, quality and productivity. It has been considered challenging problem to detect crack network in pavement which includes noise (oil marks, skid marks, previously sealed cracks and inherent noise). It is required to develop crack network mapping and modeling algorithm in order to accurately inject sealant along to the middle of cut crack network. The primary objective of this study is to propose a crack network mapping and modeling algorithm using neural network for improving the accuracy of the algorithm used in the APCS. It is anticipated that the effective use of the proposed algorithms would be able to reduce error rate in image processing for detecting, mapping and modeling crack network as well as improving quality and productivity compared to existing vision algorithms.

• Journal of information and communication convergence engineering
• /
• v.5 no.2
• /
• pp.131-135
• /
• 2007

In this paper, The proposed algorithm consists of three steps. In the first step, TD(Transition Density) calculation has to be performed. a CLB-based CPLD low-power technology mapping algorithm considered a Trade-off is proposed. To perform low-power technology mapping for CPLDs, a given Boolean network has to be represented in a DAG. Total power consumption is obtained by calculating the switching activity of each node in a DAG. In the second step, the feasible clusters are generated by considering the following conditions: the number of inputs and outputs, the number of OR terms for CLB within a CPLD. The common node cluster merging method, the node separation method, and the node duplication method are used to produce the feasible clusters. In the final step, low-power technology mapping based on the CLBs packs the feasible clusters. The proposed algorithm is examined using SIS benchmarks. When the number of OR terms is five, the experiment results show that power consumption is reduced by 30.73% compared with TEMPLA, and by 17.11 % compared with PLA mapping.

• Journal of Biomedical Engineering Research
• /
• v.17 no.2
• /
• pp.179-188
• /
• 1996

A clinically oriented 32-channel electroencephalogram (EEG) and evoked potential (EP) mapping system has been developed EEG and EP signals acquired from 32-channel electrodes attached on the heroid surface are amplified by a pre-amplifier which is separated from main amplifier and is located near the patient to reduce signal attenuation and noise contamination between electrodes and the amplifier. The amplified signals are further amplified by a main amplifier where various filtering and gain contr61 are achieved An automatic artifact rejection scheme is employed using neural network-based EEG and artifact classifier, by which examination time is substantially reduce4 The continuously measured EEG sigrlals are used for spectral mapping, and auditory and visual evoked potentials measured in synchronous to the auditory and visual stimuli are used for temporal evoked potential mapping. A user-friendly graphical interface based on the Microsoft Window 3.1 is developed for the operation of the system. Statistical databases for comparisons of group and individual are included to support a statistically-based diagnosis.

• Journal of information and communication convergence engineering
• /
• v.5 no.1
• /
• pp.59-63
• /
• 2007

In this paper, a CLB-based CPLD low-power technology mapping algorithm considered a Trade-off is proposed. To perform low-power technology mapping for CPLDs, a given Boolean network has to be represented in a DAG. The proposed algorithm consists of three steps. In the first step, TD(Transition Density) calculation has to be performed. Total power consumption is obtained by calculating the switching activity of each node in a DAG. In the second step, the feasible clusters are generated by considering the following conditions: the number of inputs and outputs, the number of OR terms for CLB within a CPLD. The common node cluster merging method, the node separation method, and the node duplication method are used to produce the feasible clusters. In the final step, low-power technology mapping based on the CLBs packs the feasible clusters. The proposed algorithm is examined using SIS benchmarks. When the number of OR terms is five, the experiment results show that power consumption is reduced by 30.73% compared with TEMPLA, and by 17.11 % compared with PLA mapping.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.6
• /
• pp.1081-1086
• /
• 2023

The rapid growth and increasing complexity of modern networks have highlighted the limitations of traditional network architectures. The emergence of SDN (Software-Defined Network) in response to these challenges has changed the existing network environment. The SDN separates the control unit and the data unit, and adjusts the network operation using a centralized controller. However, this structure has also recently caused a huge amount of traffic due to the rapid spread of numerous Internet of Things (IoT) devices, which has not only slowed the transmission speed of the network but also made it difficult to ensure quality of service (QoS). Therefore, this paper proposes a method of load distribution by switching the IP and any server (processor) from the existing data processing scheduling technique, RR (Round-Robin), to mapping when a large amount of data flows in from a specific IP, that is, server overload and data loss.

• MALSORI
• /
• no.48
• /
• pp.93-105
• /
• 2003

Since telephone channel has bandlimited frequency characteristics, speech signal over the telephone channel shows degraded speech quality. In this paper, we propose an algorithm using neural network to reconstruct wideband speech from its narrowband version. Although single neural network is a good tool for direct mapping, it has difficulty in training for vast and complicated data. To alleviate this problem, we modularize the neural networks based on appropriate clustering of the acoustic space. We also introduce fuzzy computing to compensate for probable misclassification at the cluster boundaries. According to our simulation, the proposed algorithm showed improved performance over the single neural network and conventional codebook mapping method in both objective and subjective evaluations.

• Journal of Positioning, Navigation, and Timing
• /
• v.8 no.4
• /
• pp.193-200
• /
• 2019

The auto-encoder network which is a good candidate to handle the modeling of the signal strength attenuation is designed for denoising and compensating the distortion of the received data. It provides a non-linear mapping function by iteratively learning the encoder and the decoder. The encoder is the non-linear mapping function, and the decoder demands accurate data reconstruction from the representation generated by the encoder. In addition, the adaptive network width which supports the automatic generation of new hidden nodes and pruning of inconsequential nodes is also implemented in the proposed algorithm for increasing the efficiency of the algorithm. Simulation results show that the proposed method can improve the neural network training surface to achieve the highest possible accuracy of the signal modeling compared with the conventional modeling method.

• 한국가시화정보학회:학술대회논문집
• /
• 2004.11a
• /
• pp.52-55
• /
• 2004

Thermo-chromic Liquid Crystal(TLC) particles were used as temperature sensor for thermal fluid flow. $1K\times1K$ CCD color camera and Xenon Lamp(500W) were used for the visualization of a Hele-Shaw cell. The characteristic between the reflected colors from the TLC and their corresponding temperature shows strong non-linearity. A neural network known as having strong mapping capability for non-linearity is adopted to quantify the temperature field using the image of the flow. Improvements of color-to-temperature mapping was attained by using the local color luminance (Y) and hue (H) information as the inputs for the constructed neural network.