• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.17 no.4
• /
• pp.213-220
• /
• 2024

This paper proposes an optimization strategy for performing Federated Learning between devices and edge clouds using WebAssembly (WASM). The proposed strategy aims to maximize efficiency by conducting partial training on devices and the remaining training on edge clouds. Specifically, it mathematically describes and evaluates methods to optimize data transfer between GPU memory segments and the overlapping of computational tasks to reduce overall training time and improve GPU utilization. Through various experimental scenarios, we confirmed that asynchronous data transfer and task overlap significantly reduce training time, enhance GPU utilization, and improve model accuracy. In scenarios where all optimization techniques were applied, training time was reduced by 47%, GPU utilization improved to 91.2%, and model accuracy increased to 89.5%. These results demonstrate that asynchronous data transfer and task overlap effectively reduce GPU idle time and alleviate bottlenecks. This study is expected to contribute to the performance optimization of Federated Learning systems in the future.

• Journal of Advanced Marine Engineering and Technology
• /
• v.37 no.4
• /
• pp.409-417
• /
• 2013

Mu-communication board supported by multi-communication is designed with Atxmega 128A1 which is a low power energy consuming of 8-bit microcontroller. ATxmega128A1 microcontroller consists of 8 UART(Universal asynchronous receiver/transmitter) ports which can be setting appropriate user interface having command line interpreter(CLI) program with each port, 2 kbytes EEPROM, 128 kbytes flash memory, 8 kbytes SRAM. 8 URAT ports are used for the multi communication modem, GPS module, etc. and EEPROM is used for saving a configuration for program running, and flash memory of 128 kbytes is used for storing a Firm Ware, and 8 kbytes SRAM is used for stack, storing memory of global variables while program running. If we uses the hybrid communication of path optimization of VHF, TRS and CDMA to remote control AtoN(aid to navigation), it is able to remove the communication shadow area. Even though there is a shadow area for individual communication method, we can select an optimum communication method. The compatibility of data has been enhanced as using of same data frame per communication devices. For the test, 8640 of data has been collected from the each buoy during 30 days in every 5 minutes and the receiving rate of the data has shown more than 99.4 %.

• The Journal of the Acoustical Society of Korea
• /
• v.19 no.7
• /
• pp.42-47
• /
• 2000

This paper describes a real-time implementation of a GSM-EFR (Global System for Mobil communications Enhanced Full Rate) speech coder using OakDSP core; a 16bit fixed-point Digital Signal Processor (DSP) by DSP Group, Inc. The real-time implemented speech coder required about 24MIPS for computation and 7.06K words and 12.19K words for code and data memory, respectively. The implemented GSM-EFR speech coder passes all of test vectors provided by ETSI (European Telecommunication Standard Institute), and perceptual speech quality measurement using MNB algorithm shows that the quality of the GSM-EFR speech coder is similar to the one of 32kbps ADPCM. The real-time implemented GSM-EFR speech coder which is the highest bit-rate mode of the GSM-AMR speech coder will be used as the basic structure of the GSM-AMR speech coder which is embedded in MODEM ASIC of IMT2000 asynchronous mode mobile station.

• Proceedings of the KIEE Conference
• /
• 2002.07d
• /
• pp.2344-2346
• /
• 2002

A vacuum control system has been developed for using Ethernet Multi Serial Device Severs (EMSDS) for the Pohang Accelerator Laboratory (PAL) storage ring. There are 124 vacuum ion pumps at the storage ring. It was a very important problem to solve the problem how to control such a big number of vacuum pumps distributed around the ring. After discussions, we decided to develop a serial to ethernet interrace device sever that will be mounted in the control system rack. It has a 32-bits microprocessor embedded Linux, 12 ports RS485 (or RS232) slave interface. one channel 10/100BaseTx ethernet host port, one channel UART host port, and 16 Mbytes large memory buffer. These vacuum pumps are connected to Ion-Pump serial controllers, which chop the AC current so as to control the current in the pumps. The EMSDS connect either 100BaseTx or 10BaseT ethernet networks to asynchronous serial ports for communication with serial device. It can simultaneously control up to 12 ion-pump serial controllers. 12 EMSDS are connected to a personal computer (PC) through the network. The PC can automatically control the EMSDS by sending a set of commands through the TCP/IP network. Upon receiving a command from a PC running under Windows2000 through the network, the EMSDS communicate through the stave serial interrace ports to ion-pump controller. We added some software components on the top of EPICS (Experimental Physics and Industrial Control System) toolkit.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.3
• /
• pp.509-516
• /
• 2008

Recently, the operator modules to control external devices are concerned about automatic management system to process continuously changed signals. These signals are the stream data of which characteristics are several numbers. a short report interval and asynchronous report time. It is necessary that the system brings about high accuracy and real time process for stream data. The typical queries of these systems consist of the current query to search the latest signal value, the snapshot query at a past time, the sliding window query from a past time to current. In this paper, we propose the efficient method to manage the above signals by using a file structured database in small-size operating systems. We also propose a query model to accommodate various queries including the sliding window query. The file database in the QNX adopts a delta version and a shared memory buffering method for the resource limit of a small storage and a low computing power.

• Journal of KIISE:Computer Systems and Theory
• /
• v.28 no.5
• /
• pp.213-217
• /
• 2001

Excessive memory buffer requirement in continuous media playback is a serious impediment of wide spread usage of on-line multimedia service. Skewed access frequency of available video files provides an opportunity of re-using the date blocks which has been loaded by one session for later usage. We present novel algorithm which minimizes the buffer requirement in multiple sessions of multimedia playbacks. In continuous media playback originated from the disk, a certain amount of memory buffer is required to synchronize asynchronous disk. Read operation and synchronous playback operation. As aggregate playback bandwodth increases, larger amount of buffer needs to be allocated for this synchronization purpose. The focus of this work is to study the asymptotic behavior of the synchronization buffer requirement and to develop an algorithm coping with this excessive buffer requirement under bandwidth congestioon. We argue that in a large scale continuous media server, it may not be necessary to read the blocks for each session directly from the disk. The beauty of our work lies in the fact that it dynamically adapts to disk utilization of the server and finds the optimal way of servicinh the individual sessions while minimizing the overall buffer space requirement. Optimality of the proposed algorithm is shown by proof. The effectiveness and performance of the proposed scheme is examined via simulation.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.6
• /
• pp.1107-1114
• /
• 2004

System Packet Interface Level 4 Phase 2(SPI-4.2) is an interface for packet and cell transfer between a physical layer(PHY) device and a link layer device, for aggregate bandwidths of OC-192 ATM and Packet Over Sonet/SDH(POS), as well as 10Gbps Ethernet applications. SPI-4.2 core consists of Tx and Rx modules and supports full duplex communication. Tx module of SPI-4.2 core writes 64-bit data word and 14-bit header information from the user interface into asynchronous FIFO and transmits DDR(Double Data Rate) data over PL4 interface. Rx module of SPI-4.2 core operates in vice versa. Tx and Rx modules of SPI-4.2 core are designed to support maximum 256-channel and control the bandwidth allocation by configuring the calendar memory. Automatic DIP4 and DIP-2 parity generation and checking are implemented within the designed core. The designed core uses Xilinx ISE 5.li tool and is described in VHDL Language and is simulated by Model_SIM 5.6a. The designed core operates at 720Mbps data rate per line, which provides an aggregate bandwidth of 11.52Gbps. SPI-4.2 interface core is suited for line cards in gigabit/terabit routers, and optical cross-connect switches, and SONET/SDH-based transmission systems.

• The Journal of the Convergence on Culture Technology
• /
• v.6 no.4
• /
• pp.813-818
• /
• 2020

Recently, with the rapid increase in technology and users of smart devices, the smart watch market has grown, and its utility and usability are continuously expanding. The strengths of smartwatches are wearable portability, application immediacy, data diversity and real-time capability. Despite these strengths, smartwatches have limitations such as battery limitations, display and user interface size limitations, and memory limitations. In addition, there is a need to supplement developers and standard devices, operating system standard models, and killer application modules. In particular, monitoring and application of user's biometric information is becoming a major service for smart watches. The biometric information of such a smart watch generates a large amount of data in real time. In order to advance the biometric information service, stable peer-to-peer transmission of sensing data to a remote smartphone or local server storage must be performed. We propose a synchronization method to ensure wireless remote peer-to-peer transmission stability in a smart watch system. We design a wireless peer-to-peer transmission process based on this synchronization method, analyze asynchronous transmission process and proposed synchronous transmission process, and propose a transmission efficiency method according to an increase in transmission amount.