• The KIPS Transactions:PartD
• /
• v.9D no.4
• /
• pp.579-586
• /
• 2002

This paper propose an efficient algorithm for mining association rules in the large main memory systems. To do this, the paper attempts firstly to extend the conventional algorithms such as DHP and Partition in order to be compatible to the large main memory systems and proposes secondly an algorithm to improve Partition algorithm by applying the techniques of the hash table and the bit map. The proposed algorithm is compared to the extended DHP within the experimental environments and the results show up to 65% performance improvement in comparison to the expanded DHP.

• The Journal of Society for e-Business Studies
• /
• v.15 no.4
• /
• pp.101-121
• /
• 2010

This paper presents an intelligent prefetching technique that significantly improves performance of hybrid fash-disk storage, a combination of flash memory and hard disk. Since flash memory embedded in a hybrid device is much faster than hard disk in terms of I/O operations, it can be utilized as a 'cache' space to improve system performance. The basic strategy for prefetching is to utilize sequential pattern mining, with which we can extract the access patterns of objects from historical access sequences. We use two techniques for enhancing the performance of hybrid storage with prefetching. One of them is to modify a FAST algorithm for mapping the flash memory. The other is to extend the unit of prefetching to a block level as well as a file level for effectively utilizing flash memory space. For evaluating the proposed technique, we perform the experiments using the synthetic data and real UCC data, and prove the usability of our technique.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.49 no.4
• /
• pp.26-33
• /
• 2012

As the flash memory has increased the market share of data storage in imbedded system and occupied the most of area in a system, It has a profound impact on system reliability. Flash memory is divided NOR/NAND-type according to the cell array structure, and is classified as SLC(Single Level Cell)/MLC(Multi Level Cell) according to reference voltage. Although NAND-type flash memory is slower than NOR-type, but it has large capacity and low cost. Also, By the effect of demanding mobile market, MLC NAND-type is widely adopted for the purpose of the multimedia data storage. Accordingly, Importance of fault detection algorithm is increasing to ensure MLC NAND-type flash memory reliability. There are many researches about the testing algorithm used from traditional RAM to SLC flash memory and it detected a lot of errors. But the case of MLC flash memory, testing for fault detection, there was not much attempt. So, In this paper, Extend SLC NAND-type flash memory fault detection algorithm for testing MLC NAND-type flash memory and try to reduce these differences.

• Journal of the Korean Institute of Telematics and Electronics D
• /
• v.36D no.5
• /
• pp.11-20
• /
• 1999

The data transfer width between the mocroprocessor and the memory comes to a critical part that limits system performance since the data transfer width has been as it was while the performance of a microprocessor is getting higher due to its continuous development in speed. And it is important that the memory should be in small size for the reduction of embedded microprocessor's price which is integrated on a single chip with the memory and IO circuit. In this paper, a mocroprocessor tentatively named as Extendable Instruction Set Computer(EISC) is proposed as the high code density 32 bit mocroprocessor architecture. The 32 bit EISC has 16 general purpose registers and 16 bit fixed length instruction which has the short length offset and small immediate operand. By using and extend register and extend flag, the offset and immediate operand could be extended. The proposed 32 bit EISC is implemented with an FPGA and all of its functions have been tested and verified at 1.8432MHz. And the cross assembler, the cross C/C++ compiler and the instruction simulator of the 32 bit EISC shows 140-220% and 120-140% higher code density than RISC and CISC respectively, which is much higher than any other traditional architectures. As a consequence, the EISC is suitable for the next generation computer architecture since it requires less data transfer width compared to any other ones. And its lower memory requirement will embedded microprocessor more useful.

• Journal of Korea Multimedia Society
• /
• v.3 no.2
• /
• pp.192-200
• /
• 2000

8 bit and 16 bit microprocessors are widely used in the small sited control machine. The embedded microprocessors which is integrated on a single chip with the memory and I/O circuit must have simple hardware circuit and high code density. This paper proposes a 16 bit high code density EISC(Extendable Instruction Set Computer) microprocessor. SE1608 has 8 general purpose registers and 16 bit fixed length instruction set which has the short length offset and small immediate operand. By using an extend register and extend flag, the offset and immediate operand in instruction could be extended. SE1608 is implemented with 12,000 gate FPGA and all of its functions have been tested and verified at 8MHz. And the cross assembler, the cross C/C++compiler and the instruction simulator of the SE1608 have been designed and verified. This paper also proves that the code density$.$ of SE1608 shows 140% and 115% higher code density than 16 bit microprocessor H-8300 and MN10200 respectively, which is much higher than traditional microprocessors. As a consequence, the SE1608 is suitable for the embedded microprocessor since it requires less program memory to any other ones, and simple hardware circuit.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2012.08a
• /
• pp.115-116
• /
• 2012

The demand for flexible electronic systems such as wearable computers, E-paper, and flexible displays has increased due to their advantages of excellent portability, conformal contact with curved surfaces, light weight, and human friendly interfaces over present rigid electronic systems. This seminar introduces three recent progresses that can extend the application of high performance flexible inorganic electronics. The first part of this seminar will introduce a RRAM with a one transistor-one memristor (1T-1M) arrays on flexible substrates. Flexible memory is an essential part of electronics for data processing, storage, and radio frequency (RF) communication and thus a key element to realize such flexible electronic systems. Although several emerging memory technologies, including resistive switching memory, have been proposed, the cell-to-cell interference issue has to be overcome for flexible and high performance nonvolatile memory applications. The cell-to-cell interference between neighbouring memory cells occurs due to leakage current paths through adjacent low resistance state cells and induces not only unnecessary power consumption but also a misreading problem, a fatal obstacle in memory operation. To fabricate a fully functional flexible memory and prevent these unwanted effects, we integrated high performance flexible single crystal silicon transistors with an amorphous titanium oxide (a-TiO2) based memristor to control the logic state of memory. The $8{\times}8$ NOR type 1T-1M RRAM demonstrated the first random access memory operation on flexible substrates by controlling each memory unit cell independently. The second part of the seminar will discuss the flexible GaN LED on LCP substrates for implantable biosensor. Inorganic III-V light emitting diodes (LEDs) have superior characteristics, such as long-term stability, high efficiency, and strong brightness compared to conventional incandescent lamps and OLED. However, due to the brittle property of bulk inorganic semiconductor materials, III-V LED limits its applications in the field of high performance flexible electronics. This seminar introduces the first flexible and implantable GaN LED on plastic substrates that is transferred from bulk GaN on Si substrates. The superb properties of the flexible GaN thin film in terms of its wide band gap and high efficiency enable the dramatic extension of not only consumer electronic applications but also the biosensing scale. The flexible white LEDs are demonstrated for the feasibility of using a white light source for future flexible BLU devices. Finally a water-resist and a biocompatible PTFE-coated flexible LED biosensor can detect PSA at a detection limit of 1 ng/mL. These results show that the nitride-based flexible LED can be used as the future flexible display technology and a type of implantable LED biosensor for a therapy tool. The final part of this seminar will introduce a highly efficient and printable BaTiO3 thin film nanogenerator on plastic substrates. Energy harvesting technologies converting external biomechanical energy sources (such as heart beat, blood flow, muscle stretching and animal movements) into electrical energy is recently a highly demanding issue in the materials science community. Herein, we describe procedure suitable for generating and printing a lead-free microstructured BaTiO3 thin film nanogenerator on plastic substrates to overcome limitations appeared in conventional flexible ferroelectric devices. Flexible BaTiO3 thin film nanogenerator was fabricated and the piezoelectric properties and mechanically stability of ferroelectric devices were characterized. From the results, we demonstrate the highly efficient and stable performance of BaTiO3 thin film nanogenerator.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.10a
• /
• pp.107-110
• /
• 2017

According to researches, while the speed of processor and network in embedded devices is fast enough to meet user requirement, the IO speed is recognized as the main performance bottleneck. Meanwhile it is known that more than 70 percent of IOs are issued from SQLite database. Many researches related SQLite performance optimization is based on WAL mode because WAL mode optimized for write IO performance. In this paper, I propose to optimize SQLite with Atomic Write in the Rollback Journal Mode, which is mainly used in Android and Tizen. I have observed that Atomic Write have a significant write performance improvement(300%) by reducing write, file sync operation and memory usage improvement(80%). Additionally it can block JOJ(Journaling of Journal) and extend the life of the flash memory.

• Earthquakes and Structures
• /
• v.22 no.3
• /
• pp.231-243
• /
• 2022

Beam-column joints (BCJs) are recognized among the most crucial zones in reinforced concrete structures, as they are the critical elements subjected to a complex state of forces during a severe earthquake. Under such conditions, BCJs exhibit behaviors with impacts that extend to the whole structure and significantly influence its ductility and capability of dissipating energy. The focus of this paper is to investigate the effect of undamaged transverse beam (secondary beams) on the ductility of concrete BCJs reinforced with conventional steel and shape memory alloys bars using pushover analysis at tip of beam under different axial load levels at the column using a nonlinear finite element model in ABAQUS environment. A numerical model of a BCJ was constructed and the analysis outcomes were verified by comparing them to those obtained from previous experiments found in the literature. The comparison evidenced the capability of the calibrated model to predict the load capacity response of the joint. Results proved the ability of undamaged secondary beams to provide a noticeable improvement to the ductility of reinforced concrete joints, with a very negligible loss in load capacity. However, the effect of secondary beams can become less significant if the beams are damaged due to seismic effects. In addition, the axial load was found to significantly enhance the performance of BCJs, where the increase in axial load magnified the capacity of the joint. However, higher values of axial load resulted in greater initial stiffness of the BCJ.

• Journal of Computing Science and Engineering
• /
• v.5 no.2
• /
• pp.111-120
• /
• 2011

Data mining algorithms should exploit new hardware technologies to accelerate computations. Such goal is difficult to achieve in database management system (DBMS) due to its complex internal subsystems and because data mining numeric computations of large data sets are difficult to optimize. This paper explores taking advantage of existing multithreaded capabilities of multicore CPUs as well as caching in RAM memory to efficiently compute summaries of a large data set, a fundamental data mining problem. We introduce parallel algorithms working on multiple threads, which overcome the row aggregation processing bottleneck of accessing secondary storage, while maintaining linear time complexity with respect to data set size. Our proposal is based on a combination of table scans and parallel multithreaded processing among multiple cores in the CPU. We introduce several database-style and hardware-level optimizations: caching row blocks of the input table, managing available RAM memory, interleaving I/O and CPU processing, as well as tuning the number of working threads. We experimentally benchmark our algorithms with large data sets on a DBMS running on a computer with a multicore CPU. We show that our algorithms outperform existing DBMS mechanisms in computing aggregations of multidimensional data summaries, especially as dimensionality grows. Furthermore, we show that local memory allocation (RAM block size) does not have a significant impact when the thread management algorithm distributes the workload among a fixed number of threads. Our proposal is unique in the sense that we do not modify or require access to the DBMS source code, but instead, we extend the DBMS with analytic functionality by developing User-Defined Functions.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.30B no.5
• /
• pp.27-33
• /
• 1993

This paper describes the hardware implementation of 2-D FIR digital filter for a real-time image processing. Generally, the most time-consuming operation in signal processing is the multiplication operation. To avoid it in digital filter. Pelid and Liu proposed the distributed arithmetic method for the one-dimensional case. The implementation method proposed in this paper is to extend Pelid's method to two-dimensional FIR filter using simple ROM lookup table and to use the technique of pipe lining two main operations of memory access and arithmetic. As a result, the speed of our proposed hardware implementation is two times faster than that of conventional methods and can be close to the real time speed.