• Journal of Internet of Things and Convergence
• /
• v.9 no.1
• /
• pp.65-70
• /
• 2023

In this paper, we propose an autonomous driving system using an end-to-end model to improve lane departure and misrecognition of traffic lights in a vision sensor-based system. End-to-end learning can be extended to a variety of environmental conditions. Driving data is collected using a model car based on a vision sensor. Using the collected data, it is composed of existing data and data with outlayers removed. A class was formed with camera image data as input data and speed and steering data as output data, and data learning was performed using an end-to-end model. The reliability of the trained model was verified. Apply the learned end-to-end model to the model car to predict the steering angle with image data. As a result of the learning of the model car, it can be seen that the model with the outlayer removed is improved than the existing model.

• Phonetics and Speech Sciences
• /
• v.11 no.1
• /
• pp.41-49
• /
• 2019

An end-to-end speech recognition model consisting of a single integrated neural network model was recently proposed. The end-to-end model does not need several training steps, and its structure is easy to understand. However, it is difficult to understand how the model recognizes speech internally. In this paper, we visualized and analyzed the attention-based end-to-end model to elucidate its internal mechanisms. We compared the acoustic model of the BLSTM-HMM hybrid model with the encoder of the end-to-end model, and visualized them using t-SNE to examine the difference between neural network layers. As a result, we were able to delineate the difference between the acoustic model and the end-to-end model encoder. Additionally, we analyzed the decoder of the end-to-end model from a language model perspective. Finally, we found that improving end-to-end model decoder is necessary to yield higher performance.

• 한국정보통신설비학회:학술대회논문집
• /
• 2009.08a
• /
• pp.333-336
• /
• 2009

DSON is a service overlay network architecture that can guarantee end-to-end quality of service to home network area. DSON makes service possible to be provided dynamically. This paper gives service model and scenario using DSON architecture.

• Journal of Institute of Control, Robotics and Systems
• /
• v.11 no.2
• /
• pp.175-185
• /
• 2005

The scheduling methods of the distributed real-time systems have been proposed. However, they have some weak points. They did not schedule both sporadic and periodic tasks and messages at the same time or did not consider the end-to-end constraints such as precedence relations between sporadic tasks. This means that system scheduling must guarantee the constraints of practical systems and be applicable to them. This paper proposes a new scheduling method that can be applied to more practical model of distributed real-time systems. System model consists of sporadic and periodic tasks with precedence relations and sporadic and periodic messages and has end-to-end constraints. The proposed method is based on a binary search-based period assignment algorithm, an end-to-end laxity-based priority assignment algorithm, and three kinds of schedulability analysis, node, network, and end-to-end schedulability analysis. In addition, this paper describes the application model of sporadic tasks with precedence constraints in a distributed real-time system, shows that existing scheduling methods such as Rate Monotonic scheduling are not proper to be applied to the system having sporadic tasks with precedence constraints, and proposes an end-to-end laxity-based priority assignment algorithm.

• Journal of Industrial Technology
• /
• v.29 no.B
• /
• pp.33-39
• /
• 2009

The need for end-to-end security has been increased with the widespread use of mobile devices in wireless internet access applications such as mobile commerce. The WAP1.x has an end-to-end security problem that the message transmitted between the mobile device and the Web server is decrypted inside the gateway within a short time. To overcome this problem, several protocols including WAP2.0 has been proposed. These protocols require that the heavy modules such as TLS or data compression modules should be installed on the mobile device with limited resources. This paper proposes a new WTLSW(WTLS-TLS at Web server) protocol and a new WAP2.0 proxy model in order to ensure end-to-end security between the mobile device and the Web server and to be appropriate for mobile devices with limited resources.

• Korean Journal of Computational Design and Engineering
• /
• v.17 no.1
• /
• pp.1-6
• /
• 2012

The End Coupling is main component of the aluminum hot roll process. The End Coupling is used for transmission of rotational power with heavy-duty load. Fracture of the End Coupling cause serious economic losses because an End Coupling is a very expensive component and it takes a long time to repair it. Therefore, preventing the destruction of the End Coupling is essential for ensuring a long mechanical life cycle. In this paper, the parametric study on the End Coupling was performed in order to minimize maximum stress under operation loads. To verify the interference of spindle assembly with modified End Coupling, kinematics simulation was performed by applying the various combination type and dynamic boundary condition of the spindle assembly. The interference of optimized model was not occurred during combination process and driving process. As a result of an optimum design for life extension on End Coupling, the maximum stress of modified End Coupling was lower than that of the initial model by 26%.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2001.11b
• /
• pp.826-831
• /
• 2001

Electromagnetic forces generate vibrations in the end-winding of large generators. A finite element analysis using a commercial S/W is performed to calculate electromagnetic force of end-winding in two pole generator for 500 MW fossil power plant. Also, this paper presents analytical and experimental modal analysis results of generator end- winding. Using validated FE model, 3D electromagnetic model which computes the forces on the end-winding is coupled with a 3D mechanical model which calculates the dynamic displacement and stress under electromagnetic forces. These results will be used to evaluate reliability of end-winding and applied to update model.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10c
• /
• pp.754-756
• /
• 2001

현재 인터넷 사용자의 증가와 더불어 사용자들이 원하는 서비스가 다양해짐에 따라 기존의 Best effort 서비스로는 사용자들이 요구하는 응용 서비스들을 제대로 지원할 수가 없게되었고, 이에 IETF(Internet Engineering Task Force)애서는 사용들의 QoS 제공을 위해 Integrated service model과 Differential service model이 제안되었다. 그러나 소규모 망에서 적용 가능한 Intserv와 대규모 망에서 사용적절한 Diffserv망의 service 방식이 다르기 때문에 End-to-End간에 QoS 제공을 위해서는 이에 대한 망의 연동이 필요하다. 본 논문에서는 두 망간의 연동을 위한 Framework을 제안하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.9
• /
• pp.3110-3125
• /
• 2014

Network coding improves TCP's performance in lossy wireless networks. However, the complex congestion window evolution of network coded TCP (TCP-NC) makes the analysis of end-to-end throughput challenging. This paper analyzes the evolutionary process of TCP-NC against lossy links. An analytic model is established by applying a two-dimensional Markov chain. With maximum window size, end-to-end erasure rate and redundancy parameter as input parameters, the analytic model can reflect window evolution and calculate end-to-end throughput of TCP-NC precisely. The key point of our model is that by the novel definition of the states of Markov chain, both the number of related states and the computation complexity are substantially reduced. Our work helps to understand the factors that affect TCP-NC's performance and lay the foundation of its optimization. Extensive simulations on NS2 show that the analytic model features fairly high accuracy.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.5
• /
• pp.530-536
• /
• 2021

Recently, due to the development of deep learning, end-to-end speech recognition, which directly maps graphemes to speech signals, shows good performance. Especially, among the end-to-end models, conformer shows the best performance. However end-to-end models only focuses on the probability of which grapheme will appear at the time. The decoding process uses a greedy search or beam search. This decoding method is easily affected by the final probability output by the model. In addition, the end-to-end models cannot use external pronunciation and language information due to structual problem. Therefore, in this paper conformer with lexicon transducer is proposed. We compare phoneme-based model with lexicon transducer and grapheme-based model with beam search. Test set is consist of words that do not appear in training data. The grapheme-based conformer with beam search shows 3.8 % of CER. The phoneme-based conformer with lexicon transducer shows 3.4 % of CER.