• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2001.07a
• /
• pp.37-40
• /
• 2001

This paper presents the modeling methodology of Zinc diffusion process applied for high-speed avalanche photodiode fabrication using neural networks. Three process factors (sealing pressure, amount of Zn$_3$P$_2$ source per volume, and doping concentration of diffused layer) are examined by means of D-optimal design experiment. Then, diffusion rate and doping concentration of Zinc in diffused layer are characterized by a static response model generated by training fred-forward error back-propagation neural networks. It is observed that the process models developed here exhibit good agreement with experimental results.

• Electronics and Telecommunications Trends
• /
• v.35 no.4
• /
• pp.34-41
• /
• 2020

The existing Von Neumann architecture places limits to data processing in AI, a booming technology. To address this issue, research is being conducted on computing architectures and artificial neural networks that simulate neurons and synapses, which are the hardware of the human brain. With high-speed, high-throughput data communication infrastructures, photonic solutions today are a mature industrial reality. In particular, due to the recent outstanding achievements of artificial neural networks, there is considerable interest in improving their speed and energy efficiency by exploiting photonic-based neuromorphic hardware instead of electronic-based hardware. This paper covers recent photonic neuromorphic studies and a classification of existing solutions (categorized into multilayer perceptrons, convolutional neural networks, spiking neural networks, and reservoir computing).

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.12B
• /
• pp.1052-1062
• /
• 2004

Recent studies have pointed out that existing high-speed TCP protocols have a severe unfairness and TCP friendliness problem. As the congestion window achieved by a high-speed TCP connection can be quite large, there is a strong possibility that the sender will transmit a large burst of packets. As such, the current congestion control mechanisms of high-speed TCP can lead to bursty traffic flows in hi인 speed networks, with a negative impact on both TCP friendliness and RTT unfairness. The proposed solution to these problems is to evenly space the data sent into the network over an entire round-trip time. Accordingly, the current paper evaluates this approach with a high bandwidth-delay product network and shows that pacing offers better TCP friendliness and fairness without degrading the bandwidth scalability.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.12C
• /
• pp.1207-1214
• /
• 2002

A scheduling scheme for high speed networks requires a low time complexity to schedule packets in a packet transmission time. High speed networks support a number of connections, different rates for each connection and variable packet length. Conventional round robin algorithms have a time complexity of O(1), but their short time fairness, latency and burstiness depend on the quantum of a connection due to serving several packets for a backlogged connection once a round. To improve these properties, we propose in this paper an efficient packet scheduling scheme which is based on the credits of a connection and has a time complexity of O(1). We also analyzed its performance in terms of short time fairness, latency and burstiness. The analysis results show that the proposed scheme can improve the performance compared with traditional round robin schemes. The proposed scheme can be easily utilized in high speed packet networks.

• Proceedings of the KIEE Conference
• /
• 1996.07b
• /
• pp.808-810
• /
• 1996

In this paper, neural network, which has learning capability, is used for fault type classification and fault section estimation for high speed relaying. The potential of the neural network approach is demonstrated by simulation using ATP. The instantaneous values of voltages and currents are used the inputs of neural networks. This approach determines the fault section directly. In this paper, back-propagation network(BPN) is used for fault type classification and fault section estimation and can use for high speed relaying because it determines fault section within a few msec.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.10
• /
• pp.4015-4033
• /
• 2015

The future mobile networks consist of hyper-dense heterogeneous and small cell networks of same or different radio access technologies (RAT). Integrating mobile networks of different RATs to provide seamless and smooth mobility service will be the target of future mobile converged network. Generally, handover from high-speed networks to low-speed networks faces many challenges from application perspective, such as abrupt bandwidth variation, packet loss, round trip time variation, connection disruption, and transmission blackout. Existing inter-RAT handover solutions cannot solve all the problems at the same time. Based on the high-layer convergence sublayer design, a new receiver-aided soft inter-RAT handover is proposed. This soft handover scheme takes advantage of multihoming ability of multi-mode mobile station (MS) to smooth handover procedure. In addition, handover procedure is seamless and applicable to frequent handover scenarios. The simulation results conducted in UMTS-WiMAX converged network scenario show that: in case of TCP traffics for handover from WiMAX to UMTS, not only handover latency and packet loss are eliminated completely, but also abrupt bandwidth/wireless RTT variation is smoothed. These delightful features make this soft handover scheme be a reasonable candidate of mobility management for future mobile converged networks.

• The KIPS Transactions:PartC
• /
• v.11C no.4
• /
• pp.527-538
• /
• 2004

In this paper, we present the flow aggregation method and the FLTC(flow lasting time control) algorithm to reduce the number of flows and solve the scalability problem in high speed IP switching networks. The flow aggregation based on the destination address could reduce the total number of flows, improve the label efficiency, and increase the total amount of the switched packets. The FLTC algorithm also eliminates the waste of label by deleting the flow binding efficiently. With the traces of real Internet traffics, we evaluate the performance of these schemes by simulation. The label efficiency, the average number of label used, and the percentage of packets switched and the number of packets switched are used as performance measures for this simulation.

• International Journal of Advanced Culture Technology
• /
• v.7 no.4
• /
• pp.327-333
• /
• 2019

Wind speed data constitute important weather information for aircrafts flying at low altitudes, such as drones. Currently, the accuracy of low altitude wind predictions is much lower than that of high-altitude wind predictions. Deep neural networks are proposed in this study as a method to improve wind speed forecast information. Deep neural networks mimic the learning process of the interactions among neurons in the brain, and it is used in various fields, such as recognition of image, sound, and texts, image and natural language processing, and pattern recognition in time-series. In this study, the deep neural network model is constructed using the wind prediction values generated by the numerical model as an input to improve the wind speed forecasts. Using the ground wind speed forecast data collected at the Boseong Meteorological Observation Tower, wind speed forecast values obtained by the numerical model are compared with those obtained by the model proposed in this study for the verification of the validity and compatibility of the proposed model.

• Proceedings of the IEEK Conference
• /
• 2000.07a
• /
• pp.101-104
• /
• 2000

Nowadays, the issue of congestion control in high-speed communication networks becomes critical in view of the bandwidth-delay products for efficient data flow. In particular, the fact that the congestion is often accompanied by the data flow from the high-speed link to low-speed link is important with respect to the stability of closed-loop congestion control. The Virtual-Connection Network (VCN) in Gigabit Ethernet networks is a packet-switching based network capable of implementing cell- based connection, link-by-link flow-controlled connection, and single- or multi-destination virtual connections. VCN described herein differ from the virtual channel in ATM literature in that VCN have link-by-link flow control and can be of multi-destination. VCNs support both connection-oriented and connectionless data link layer traffic. Therefore, the worst collision scenario in Ethernet CSMA/CD with virtual collision brings about end-to-end delay. Gigabit Ethernet networks based on CSMA/CD results in non-deterministic behavior because its media access rules are based on random probability. Hence, it is difficult to obtain any sound mathematical formulation for congestion control without employing random processes or fluid-flow models. In this paper, an analytical method for the design of a congestion control scheme is proposed based on Smith＇s principle to overcome instability accompanied with the increase of end-to-end delays as well as to avoid cell losses. To this end, mathematical analysis is provided such that the proposed control scheme guarantees the performance improvement with respect to bandwidth and latency for selected network links with different propagation delays. In addition, guaranteed bandwidth is to be implemented by allowing individual stations to burst several frames at a time without intervening round-trip idle time.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.05a
• /
• pp.568-573
• /
• 2003

This paper describes a study of terrestrial digital TV transmission systems using wideband opto-electronic techniques such as electro-absorption modulators, wavelength division multiplex to support cost effective, high speed, flexible and interactive digital TV networks including ultra-high speed subscriber networks. In particular, the national DTV networks to be installed is required huge amount of cost. Therefore, the network should be carefully designed for DTV, communication and multimedia services in the future. In this paper, wideband optoelectronic DTV systems are suggested.