• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.1-25
• /
• 2020

In this paper, we suggest an application system architecture which provides accurate, fast and efficient automatic gasometer reading function. The system captures gasometer image using mobile device camera, transmits the image to a cloud server on top of private LTE network, and analyzes the image to extract character information of device ID and gas usage amount by selective optical character recognition based on deep learning technology. In general, there are many types of character in an image and optical character recognition technology extracts all character information in an image. But some applications need to ignore non-of-interest types of character and only have to focus on some specific types of characters. For an example of the application, automatic gasometer reading system only need to extract device ID and gas usage amount character information from gasometer images to send bill to users. Non-of-interest character strings, such as device type, manufacturer, manufacturing date, specification and etc., are not valuable information to the application. Thus, the application have to analyze point of interest region and specific types of characters to extract valuable information only. We adopted CNN (Convolutional Neural Network) based object detection and CRNN (Convolutional Recurrent Neural Network) technology for selective optical character recognition which only analyze point of interest region for selective character information extraction. We build up 3 neural networks for the application system. The first is a convolutional neural network which detects point of interest region of gas usage amount and device ID information character strings, the second is another convolutional neural network which transforms spatial information of point of interest region to spatial sequential feature vectors, and the third is bi-directional long short term memory network which converts spatial sequential information to character strings using time-series analysis mapping from feature vectors to character strings. In this research, point of interest character strings are device ID and gas usage amount. Device ID consists of 12 arabic character strings and gas usage amount consists of 4 ~ 5 arabic character strings. All system components are implemented in Amazon Web Service Cloud with Intel Zeon E5-2686 v4 CPU and NVidia TESLA V100 GPU. The system architecture adopts master-lave processing structure for efficient and fast parallel processing coping with about 700,000 requests per day. Mobile device captures gasometer image and transmits to master process in AWS cloud. Master process runs on Intel Zeon CPU and pushes reading request from mobile device to an input queue with FIFO (First In First Out) structure. Slave process consists of 3 types of deep neural networks which conduct character recognition process and runs on NVidia GPU module. Slave process is always polling the input queue to get recognition request. If there are some requests from master process in the input queue, slave process converts the image in the input queue to device ID character string, gas usage amount character string and position information of the strings, returns the information to output queue, and switch to idle mode to poll the input queue. Master process gets final information form the output queue and delivers the information to the mobile device. We used total 27,120 gasometer images for training, validation and testing of 3 types of deep neural network. 22,985 images were used for training and validation, 4,135 images were used for testing. We randomly splitted 22,985 images with 8:2 ratio for training and validation respectively for each training epoch. 4,135 test image were categorized into 5 types (Normal, noise, reflex, scale and slant). Normal data is clean image data, noise means image with noise signal, relfex means image with light reflection in gasometer region, scale means images with small object size due to long-distance capturing and slant means images which is not horizontally flat. Final character string recognition accuracies for device ID and gas usage amount of normal data are 0.960 and 0.864 respectively.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.1
• /
• pp.21-27
• /
• 2020

In this paper, a hybrid scheduling algorithm is proposed for CMS(Combat Management System) to improve QoS(Quality of Service) based on DWDRR(Dynamic Weighted Deficit Round Robin) and priority-based scheduling method. The main proposed scheme, DWDRR is method of packet transmission through giving weight by traffic of queue and priority. To demonstrate an usefulness of proposed algorithm through simulation, efficiency in special section of the proposed algorithm is proved. Therefore, We propose hybrid algorithm between existing algorithm and proposed algorithm. Also, to prevent frequent scheme conversion, a hysteresis method is applied. The proposed algorithm shows lower packet loss rate and delay in the same traffic than existing algorithm.

• Journal of Korean Society for Quality Management
• /
• v.47 no.4
• /
• pp.823-835
• /
• 2019

Purpose: This study deals with a production planning and scheduling problem to minimize the total weighted tardiness on hybrid flow shop with sets of non-identical parallel machines on stages, where parallel machines in the set are dedicated to perform specific subsets of jobs and sequence-dependent setup times are also considered. Methods: A two-stage approach, that applies MILP model in the 1st stage and dispatching rules in the 2nd stage, is proposed in this paper. The MILP model is used to assign jobs to a specific machine in order to equalize the workload of the machines at each stage, while new dispatching rules are proposed and applied to sequence jobs in the queue at each stage. Results: The proposed two-stage approach was implemented by using a commercial MILP solver and a commercial simulation software and a case study was developed based on the spark plug manufacturing process, which is an automotive component, and verified using the company's actual production history. The computational experiment shows that it can reduce the tardiness when used in conjunction with the dispatching rule. Conclusion: This proposed two-stage approach can be used for HFS systems with dedicated machines, which can be evaluated in terms of tardiness and makespan. The method is expected to be used for the aggregated production planning or shop floor-level production scheduling.

• The Journal of the Korea Contents Association
• /
• v.8 no.1
• /
• pp.169-176
• /
• 2008

In this paper, we develop radio resource management queueing model in IEEE 802.IS networks considering both connection and packet level. In the upper level connection, we model connection admission control depending on availability of bandwidth and priority queue in each service class. In the lower level packet, we model dynamic bandwidth allocation considering threshold and availability of bandwidth in each service class simultaneously. Hierarchical model is built using an extended Petri Nets, SRN (Stochastic Reward Nets). Bandwidth utilization and normal throughput as performance index for all service classes of traffic are calculated and numerical results are obtained.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.39 no.7
• /
• pp.22-29
• /
• 2002

The Gauranteed Frame Rate service category was proposed by the ATM Forum as an enhancement to the UBR service. This service aims to support minimum cell rate gaurantee for each virtual connection and allow any excess bandwidth in the network to be shared among the contending VCs in a fair manner. We propose a new buffer management algorithm for GFR service through FIFO queuing discipline. Proposed scheme can provide minimum bandwidth guarantee for GFR VCs as well as improve the fairness among the competing GFR VCs on a single FIFO queue. From simulation results, we demonstrate the proposed scheme fulfills the requirement of GFR service as well as improves the TCP throughput.

• International Journal of Internet, Broadcasting and Communication
• /
• v.8 no.1
• /
• pp.33-41
• /
• 2016

Big data management is becoming an increasingly important issue in both industry and academia of information science community today. One of the important categories of big data generated from software systems is log data. Log data is generally used for better services in various service providers and can also be used to improve system reliability. In this paper, we propose a novel big data management architecture specialized for log data. The proposed architecture provides a scalable log management system that consists of client and server side modules for efficient handling of log data. To support large and simultaneous log data from multiple clients, we adopt the Hadoop infrastructure in the server-side file system for storing and managing log data efficiently. We implement the proposed architecture to support various client environments and validate the efficiency through measurement studies. The results show that the proposed architecture performs better than the existing logging architecture by 42.8% on average. All components of the proposed architecture are implemented based on open source software and the developed prototypes are now publicly available.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.05a
• /
• pp.312-313
• /
• 2017

Internet of Things(IoT) is an intelligent technology and service in which all objects communicate with each other through various networks. Recently Internet of Things(IoT) is one of the fields that is attracting attention as the development of ICT industry. MQTT is a protocol which is safe using TLS or adopting light packet structure for effciency of memory and power using. In this paper, when TLS is used the process of encryption / decryption in the broker occurs. We propose an efficient MQTT protocol through centralized key management by adding authentication server.

• Journal of KIISE:Computer Systems and Theory
• /
• v.35 no.11
• /
• pp.517-526
• /
• 2008

As the functionality of portable devices arc being enhanced and the performance is being greatly improved, power dissipations of battery driven portable devices are being increased. So, an efficient power management for reducing their power consumption is needed. In this paper, we propose a window-based DVS algorithm for MPEG Player. The proposed algorithm maintains the recently frame information and execution time received from MPEG player in window queue and dynamically adjusts (frequency, voltage) level based on window queue information. Our algorithm can be implemented in the common multimedia player as a module. We employed well-known MPlayer for the measurement of performance. The experimental result shows that the proposed algorithm reduces energy consumption by 56% on maximal performance.

• Journal of KIISE:Information Networking
• /
• v.29 no.5
• /
• pp.553-561
• /
• 2002

We present a novel service discipline, called linear service discipline, to serve multiple QoS queues sharing a resource and analyze its properties. The linear server makes the output traffic and the queueing dynamics of individual queues as a linear function of its input traffic. In particular, if input traffic is Gaussian, the distributions of queue length and output traffic are also Gaussian with their mean and variance being a function of input mean and input power spectrum (equivalently, autocorrelation function of input). Important QoS measures including buffer overflow probability and queueing delay distribution are also expressed as a function of input mean and input power spectrum. This study explores a new direction for network-wide traffic management based on linear system theories by letting us view the queueing process at each node as a linear filter.

• The Journal of the Korea Contents Association
• /
• v.18 no.3
• /
• pp.34-40
• /
• 2018

In this study, the optimal buffer size is calculated for seamless video playback on a mobile device. Buffer means the memory space for multimedia packet which arrives in mobile device for video play such as VOD service. If the buffer size is too large, latency time before video playback can be longer. However, if it is too short, playback service can be paused because of shortage of packets arrived. Hence, the optimal buffer size insures QoS of video playback on mobile devices. We model the process of buffering into a discret-time queueing model. Mean busy period length and mean waiting time of Geo/G/1 queue with N-policy is analyzed. After then, we uses the main performance measures to present numerical examples to decide the optimal buffer size on mobile devices. Our results enhance the user satisfaction by insuring the seamless playback and minimizing the initial delay time in VOD streaming process.