• Proceedings of the IEEK Conference
• /
• 2000.06b
• /
• pp.153-156
• /
• 2000

This paper presents a SOC design methodology using the newly-emerging SystemC. The suggested methodology firstly uses SystemC to define blocks from the previously-developed system level algorithm with internal behavior and interface being separated and validate such a described blocks' functionality when integrated. Next, the partitioning between software and hardware is considered. With software, the interface to hardware is described cycle-accurate and the other internal behavior in conventional ways. With hardware, I/O transactions are refined gradually in several abstraction levels and internal behavior described on a function basis. Once hardware and software have been completed functionally, system performance analysis is performed on the built model with assumed performance factors and influences such decisions regressively as on optimum algorithm selection, partitioning and etc. The analysis then gives constraint information when hardware description undergoes scheduling and fixed-point trans- formation with the help of automatic translation tools or manually. The methodology enables C/C++ program developers and VHDL/Verilog users to migrate quickly to a co-design & co-verification environment and is suitable for SoC development at a low cost.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.38 no.6
• /
• pp.1-10
• /
• 2001

In this paper, we propose a synchronous reservation protocol that is efficient for supporting variable sized messages in a wavelength division multiplexing (WDM) based local network using a passive star topology. A. control channel is used to coordinate transmission on data channels. Time is slotted with fixed-sized slots. The network can accommodate a variable number of nodes, and can operate independently of the change of the number of nodes. So, any 'new' node can join the network anytime without network re-initialization. Moreover, with the protocol, we can avoid data channel and destination conflicts. We analyze the performance by using EPA method according to the variation of the end to end propagation delay with respect to one slot time, and validate the results by simulation.

• Journal of Broadcast Engineering
• /
• v.11 no.3 s.32
• /
• pp.320-325
• /
• 2006

An image is partitioned into non-overlapping $8{\times}8$ blocks fer JPEG compression. A scan order converter is placed before the JPEG encoder to provide $8{\times}8$ blocks from the pixels in raster scan order. In general, its architecture requires two line memories for storing eight lines separately to allow the concurrent memory access by both the camera and JPEG processors. Although such architecture is simple to be implemented, it can be inefficient due to too excessive memory requirement as the image resolution increases. However, no deterministic addressing equation has been developed for scan conversion. In this paper, an effective memory addressing algorithm is proposed that can be devised only by adders and subtracters to implement a scan converter based on the single line memory.

• Journal of Sensor Science and Technology
• /
• v.10 no.3
• /
• pp.187-195
• /
• 2001

A wireless telemetry measurement system, using ISFETs, was developed for environmental monitoring applications, industrial processes, medical treatments and so on. This system is composed of the measurement part located at remote site and a personal computer(PC) which control whole instrument process. The measurement part transmits measured data to the PC by RF transceiver, and transmitted data are analyzed and handled in the PC. Proposed system utilizing time division multiplexing for the transmission of 3channel ISFET signal. The measurement part has an identification number(ID) so that the PC controls many measurement parts separately. Experiments were performed using pH-ISFETs, and the implemented system operates well within designed specification and accuracy of 0.1 pH.

• Journal of KIISE:Computer Systems and Theory
• /
• v.35 no.4
• /
• pp.150-157
• /
• 2008

This paper presents a design and implementation of realtime video player that operates on a tiled-display system consisting of multiple PCs to provide a very large and high resolution display. In the proposed system, the master process transmits a compressed video stream to multiple PCs using UDP multicast. All slaves(PC) receive the same video stream, decompress, clip their designated areas from the decompressed video frame, and display it to their displays while being synchronized with each other. A simple synchronization mechanism based on the H/W clock of each slave is proposed to avoid the skew between the tiles of the display, and a flow-control mechanism based on the bit-rate of the video stream and a pre-buffering scheme are proposed to prevent the jitter The proposed system is implemented with Microsoft DirectX filter technology in order to decouple the video/audio codec from the player.

• Journal of KIISE:Information Networking
• /
• v.31 no.5
• /
• pp.519-531
• /
• 2004

Ad-hoc networks are temporary wireless systems composed of mobile nodes without any fixed infrastructure. The life time of each node in the ad-hoc network significantly affects the life time of whole ad-hoc network. A node which drained out its battery may incur the partition of whole network in some network topology The life time of each node depends on the battery capacity of each node. Therefore if all mobile nodes in the network live evenly long, the life time of the network will be longer. In this paper, we propose Power-Aware Dynamic Source Routing (PADSR) which selects the best path to make the life time of the network be longer. In PADSR, when a source node finds a path to the destination node, it selects the best path that makes nodes in the network live evenly long. To find the best path, PADSR considers the consumption of transmission energy and residual battery capacity of nodes upon the path. Consequently the network lives longer if we use PADSR.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.3C
• /
• pp.248-255
• /
• 2002

An RF module in compliance with the IEEE 802.11a standard has been designed and its performance has been measured. Conventional heterodyne architecture with 580MHz intermediate frequency has been realized. Measurement results show that the receiver has a low Noise Figure of 5dB, the maximum gain of 70dB, and dynamic range as wide as 61dB. Also, the SAW filter used for channel selection in the IF section allowed minimum inter-channel interference. in addition to satisfying the RF output power requirement, the transmitter features its output P1dB as high as 34dBm so that the high peak-to-average ratio of the Orthogonal Frequency Division Multiplexing (OFDM) modulation scheme can be handled with minimum nonlinear distortion. The output P1dB of 34dBm of the transmitter corresponds to back-off powers of 18dB and 11dB with respect to the output power for the low and the middle frequency bands, respectively.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.12B
• /
• pp.1076-1086
• /
• 2006

In providing video-on-demand (VoD) services to a number of clients through networks, the bandwidth requirements of video transmission restrict VoD services. For this reason, many significant broadcasting-based VoD schemes have been proposed to efficiently support services. However, the broadcasting-based VoD schemes approach needs frequency channel hopping, using many channels at the same time and managing many segments of a video. These make it difficult to implement. In this paper, we propose a Harmonic Staggered broadcasting scheme which has a simple structure and substantially improved VoD efficiency. The numerical results demonstrate that the viewer's waiting time of the Harmonic Staggered broadcasting scheme is close to the harmonic broadcasting scheme and the maximum buffer requirements of this can be adapted for demanding rate by adjusting the short front part of a video sizes.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.6
• /
• pp.1390-1403
• /
• 1998

Database sharing system (DSS) refers to a system for high performance transaction processing. In DSS, the processing nodes are locally coupled via a high speed network and share a common database at the disk level. Each node has a local memory, a separate copy of operating system, and a DB'\fS. To reduce the number of disk accesses, the node caches database pages in its local memory buffer. However, since multiple nodes may be simultaneously cached a page, cache consistency must be cnsured so that every node can always access the'latest version of pages. In this paper, we propose efficient cache consistency schemes in DSS, where the database is logically partitioned using primary copy authority to reduce locking overhead, The proposed schemes can improve performance by reducing the disk access overhead and the message overhead due to maintaining cache consistency, Furthermore, they can show good performance when database workloads are varied dynamically.

• Korean Journal of Optics and Photonics
• /
• v.7 no.4
• /
• pp.419-427
• /
• 1996

Thermally-stabilized fiber-optic current sensors are proposed and demonstrated. The sensor head is made of two coil formers combined perpendicularly. In this sensor head, bending-induced birefringences can be reduced to the level much smaller than those of the single former type because the eigen-axes of the two perpendicular coil formers are made orthogonal to each other. Moreover, thermal variation of the birefringence is also expected to be minimized by the orthogonality of the two polarization eigen-axes. We changed the temperature slowly in the range of 20~45$^{\circ}C$ during 100 minutes. The overall linearity of the sensor is better than 1.2% in the range of 0~1000A. The long-term fluctuation of the sensor is less than 1% when measured for 3 hours at 500A and room temperature. Two orthogonally-polarized laser diodes are combined together to make the incident beam unpolarized. In the signal processing, the signals are separated by two parts and normalized respectively, which minimize the efects of optical fluctuations coming from sources, connectors, etc.