Kyu Ho Park

An efficient NAND flash file system for flash memory storage

In this paper, we present an efficient flash file system for flash memory storage. Flash memory, especially NAND flash memory, has become a major method for data storage. Currently, a block level translation interface is required between an existing file system and flash memory chips due to its physical characteristics. However, the approach of existing file systems on top of the emulating block interface has many restrictions and is, thus, inefficient because existing file systems are designed for disk-based storage systems. The flash file system proposed in this paper is designed for NAND flash memory storage while considering the existing file system characteristics. Our target performance metrics are the system booting time and garbage collection overheads, which are important issues in flash memory. In our experiments, the proposed flash file system outperformed other flash file systems both in booting time and garbage collection overheads.

Automatic human face location in a complex background using motion and color information

Abstract Automatic human face location in a complex background is one of the difficult problems in developing a fully automatic face recognition system and is still a relatively unexplored problem. In this paper, we propose an effective automatic face location system that can locate the face region in a complex background when the system is used as a pre-processor of a practical face recognition system for security. We use two natural visual cues, the motion and the color, to locate face regions in a real scene. In the first stage of the proposed system, we extract a velocity vector field of the face, and by thresholding this velocity field to detect the region showing a clear motion, the face region is separated from the background. In the next stage, the method based on the thresholding in hue space and knowledge-based location process is used to find the eye, eyebrow, and mouth regions of the face. According to our experiment, the proposed system shows fast and robust location results.

A Real-Time Discrete Event System Specification Formalismfor Seamless Real-Time Software Development

We present a time domain extension of the hierarchical and modular discrete event specification (DEVS) formalism. This extension is important for establishing a seamless real-time software development framework. Formalisms help describe a system unambiguously. If formal models are implemented without any consistent frameworks, however, it is hard to guarantee that there is no semantic gap between models and codes. Real-Time DEVS, named RTDEVS, is an extension of DEVS that can be characterized in three perspectives: the real time execution of models, the addition of time interval functions, and the activity specification for each state. After analyzing a system, the framework based on RTDEVS helps to expand each model of the system for executing in a real-time environment. In order to support the RTDEVS formalism, we propose abstract executive concepts based on the abstract simulator concepts of the DEVS formalism. Also, we implement an RTDEVS execution engine, named DEVS Executive, which runs on real-time Mach.

PFFS: a scalable flash memory file system for the hybrid architecture of phase-change RAM and NAND flash

In this paper, we present the scalable and efficient flash file system using the combination of NAND and Phase-change RAM (PRAM). Until now, several flash file systems have been developed considering the physical characteristics of NAND flash. However, previous flash file systems still have a high performance overhead and a scalability problem of the mounting time and the memory usage because, in most case, the metadata is written with several words at a single update even though the writes in NAND flash must be performed in terms of page, which is typically 2 KiB. The proposed flash file system called PFFS uses PRAM to mitigate the limitation of NAND flash. The PRAM is a next generation non-volatile memory and good for dealing with word level read/write of a small size of data. PFFS hence separates the metadata from the regular data in a file system and saves them into PRAM. Consequently, the PFFS manages all the files and directories in the PRAM and outperforms other flash file systems. The experimental results show that the performance of PFFS is 25% better than YAFFS2 for small-file writes while matching YAFFS2 performance for large writes and the mouting time and the memory usage of PFFS are O(1).

A Cooperative Clustering Protocol for Energy Saving of Mobile Devices with WLAN and Bluetooth Interfaces

One of the most widely used wireless communication standards is a Wireless Local Area Network (WLAN) (IEEE 802.11). However, WLAN has a serious power consumption problem. In this paper, we propose a novel energy saving approach that exploits the multiradio feature of recent mobile devices equipped with WLAN and Bluetooth interfaces. Unlike previous approaches, our work is based on clustering. In our work, a cluster is a Bluetooth Personal Area Network (PAN), which consists of one cluster head and several regular nodes. The cluster head acts as a gateway between the PAN and the WLAN, enabling the regular nodes to access the WLAN infrastructure via low-power Bluetooth. We present a distributed clustering protocol, Cooperative Networking protocol (CONET), which dynamically reforms clusters according to each nodes bandwidth requirement, energy use, and application type. CONET does not require modifications of existing wireless infrastructures because clustering is performed independently of WLAN access points. We implemented the CONET prototype with four wearable computing devices to evaluate the performance on real hardware. We also simulated CONET for large networks of more than 100 mobile nodes. Both results demonstrate that our approach is effective in reducing the power consumption of WLAN.

A fast path planning by path graph optimization

A fast path planning method by optimization of a path graph for both efficiency and accuracy is proposed. A conventional quadtree-based path planning approach is simple, robust, and efficient. However, it has two limitations. We propose a path graph optimization technique employing a compact mesh representation. A world space is triangulated into a base mesh and the base mesh is simplified to a compact mesh. The compact mesh representation is object-dependent; the positions of vertexes of the mesh are optimized according to the curvatures of the obstacles. The compact mesh represents the obstacles as accurately as the quadtree even though using much fewer vertexes than the quadtree. The compact mesh distributes vertexes in a free space in a balanced way by ensuring that the lengths of edges are below an edge length threshold. An optimized path graph is extracted from the compact mesh. An iterative vertex pushing method is proposed to include important obstacle boundary edges in the path graph. Dijkstras shortest path searching algorithm is used to search the shortest path in the path graph. Experimental results show that the path planning using the optimized path graph is an order of magnitude faster than the quadtree approach while the length of the path generated by the proposed method is almost the same as that of the path generated by the quadtree.

JFTL: A flash translation layer based on a journal remapping for flash memory

In flash memory-based storage, a Flash Translation Layer (FTL) manages the mapping between the logical addresses of a file system and the physical addresses of the flash memory. When a journaling file system is set up on the FTL, the consistency of the file system is guaranteed by duplications of the same file system changes in both the journal region of the file system and the home locations of the changes. However, these duplications inevitably degrade the performance of the file system. In this article we present an efficient FTL, called JFTL, based on a journal remapping technique. The FTL uses an address mapping method to write all the data to a new region in a process known as an out-of-place update. Because of this process, the existing data in flash memory is not overwritten by such an update. By using this characteristic of the FTL, the JFTL remaps addresses of the logged file system changes to addresses of the home locations of the changes, instead of writing the changes once more to flash memory. Thus, the JFTL efficiently eliminates redundant data in the flash memory as well as preserving the consistency of the journaling file system. Our experiments confirm that, when associated with a writeback or ordered mode of a conventional EXT3 file system, the JFTL enhances the performance of EXT3 by up to 20%. Furthermore, when the JFTL operates with a journaled mode of EXT3, there is almost a twofold performance gain in many cases. Moreover, the recovery performance of the JFTL is much better than that of the FTL.

Fast learning method for back-propagation neural network by evolutionary adaptation of learning rates

In training a back-propagation neural network, the learning speed of the network is greatly affected by its learning rate. None, however, has offered a deterministic method for selecting the optimal learning rate. Some researchers have tried to find the sub-optimal learning rates using various techniques at each training step. This paper proposes a new method for selecting the sub-optimal learning rates by an evolutionary adaptation of learning rates for each layer at every training step. Simulation results show that the learning speed achieved by our method is superior to that of other adaptive selection methods.

A heuristic triangulation algorithm for multiple planar contours using an extended double branching procedure

This paper presents a heuristic triangulation algorithm for reconstructing surfaces over a set of cross-sectional contours. The multiple branching problem, an important problem of conventional triangulation methods, is reated as a set of double branchings, and an algorithm based on countour merging is developed. Several imaginary contours are generated to handle the multiple branching problem with many branch contours. A double branching algorithm based on the partitioning of the root contour is also proposed. The results show that our method works well even for objects with many complicated branches.

THEMIS: A Mutually Verifiable Billing System for the Cloud Computing Environment

With the widespread adoption of cloud computing, the ability to record and account for the usage of cloud resources in a credible and verifiable way has become critical for cloud service providers and users alike. The success of such a billing system depends on several factors: The billing transactions must have integrity and nonrepudiation capabilities; the billing transactions must be nonobstructive and have a minimal computation cost; and the service level agreement (SLA) monitoring should be provided in a trusted manner. Existing billing systems are limited in terms of security capabilities or computational overhead. In this paper, we propose a secure and nonobstructive billing system called THEMIS as a remedy for these limitations. The system uses a novel concept of a cloud notary authority for the supervision of billing. The cloud notary authority generates mutually verifiable binding information that can be used to resolve future disputes between a user and a cloud service provider in a computationally efficient way. Furthermore, to provide a forgery-resistive SLA monitoring mechanism, we devised a SLA monitoring module enhanced with a trusted platform module (TPM), called S-Mon. The performance evaluation confirms that the overall latency of THEMIS billing transactions (avg. 4.89 ms) is much shorter than the latency of public key infrastructure (PKI)-based billing transactions (avg. 82.51 ms), though THEMIS guarantees identical security features as a PKI. This work has been undertaken on a real cloud computing service called iCubeCloud.