• The Transactions of the Korea Information Processing Society
• /
• v.1 no.3
• /
• pp.367-379
• /
• 1994

This paper describes scheduling algorithms of the UNIX operating system and shows an analytical approach to approximate the average conditional response time for a process in the UNIX operating system. The average conditional response time is the average time between the submittal of a process requiring a certain amount of the CPU time and the completion of the process. The process scheduling algorithms in thr UNIX system are based on the priority service disciplines. That is, the behavior of a process is governed by the UNIX process schuduling algorithms that (ⅰ) the time-shared computer usage is obtained by allotting each request a quantum until it completes its required CPU time, (ⅱ) the nonpreemptive switching in system mode and the preemptive switching in user mode are applied to determine the quantum, (ⅲ) the first-come-first-serve discipline is applied within the same priority level, and (ⅳ) after completing an allotted quantum the process is placed at the end of either the runnable queue corresponding to its priority or the disk queue where it sleeps. These process scheduling algorithms create the round-robin effect in user mode. Using the round-robin effect and the preemptive switching, we approximate a process delay in user mode. Using the nonpreemptive switching, we approximate a process delay in system mode. We also consider a process delay due to the disk input and output operations. The average conditional response time is then obtained by approximating the total process delay. The results show an excellent response time for the processes requiring system time at the expense of the processes requiring user time.

• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.89-106
• /
• 2022

Recently, the amount of video data collected from smartphones, CCTVs, black boxes, and high-definition cameras has increased rapidly. According to the increasing video data, the requirements for analysis and utilization are increasing. Due to the lack of skilled manpower to analyze videos in many industries, machine learning and artificial intelligence are actively used to assist manpower. In this situation, the demand for various computer vision technologies such as object detection and tracking, action detection, emotion detection, and Re-ID also increased rapidly. However, the object detection and tracking technology has many difficulties that degrade performance, such as re-appearance after the object's departure from the video recording location, and occlusion. Accordingly, action and emotion detection models based on object detection and tracking models also have difficulties in extracting data for each object. In addition, deep learning architectures consist of various models suffer from performance degradation due to bottlenects and lack of optimization. In this study, we propose an video analysis system consists of YOLOv5 based DeepSORT object tracking model, SlowFast based action recognition model, Torchreid based Re-ID model, and AWS Rekognition which is emotion recognition service. Proposed model uses single-linkage hierarchical clustering based Re-ID and some processing method which maximize hardware throughput. It has higher accuracy than the performance of the re-identification model using simple metrics, near real-time processing performance, and prevents tracking failure due to object departure and re-emergence, occlusion, etc. By continuously linking the action and facial emotion detection results of each object to the same object, it is possible to efficiently analyze videos. The re-identification model extracts a feature vector from the bounding box of object image detected by the object tracking model for each frame, and applies the single-linkage hierarchical clustering from the past frame using the extracted feature vectors to identify the same object that failed to track. Through the above process, it is possible to re-track the same object that has failed to tracking in the case of re-appearance or occlusion after leaving the video location. As a result, action and facial emotion detection results of the newly recognized object due to the tracking fails can be linked to those of the object that appeared in the past. On the other hand, as a way to improve processing performance, we introduce Bounding Box Queue by Object and Feature Queue method that can reduce RAM memory requirements while maximizing GPU memory throughput. Also we introduce the IoF(Intersection over Face) algorithm that allows facial emotion recognized through AWS Rekognition to be linked with object tracking information. The academic significance of this study is that the two-stage re-identification model can have real-time performance even in a high-cost environment that performs action and facial emotion detection according to processing techniques without reducing the accuracy by using simple metrics to achieve real-time performance. The practical implication of this study is that in various industrial fields that require action and facial emotion detection but have many difficulties due to the fails in object tracking can analyze videos effectively through proposed model. Proposed model which has high accuracy of retrace and processing performance can be used in various fields such as intelligent monitoring, observation services and behavioral or psychological analysis services where the integration of tracking information and extracted metadata creates greate industrial and business value. In the future, in order to measure the object tracking performance more precisely, there is a need to conduct an experiment using the MOT Challenge dataset, which is data used by many international conferences. We will investigate the problem that the IoF algorithm cannot solve to develop an additional complementary algorithm. In addition, we plan to conduct additional research to apply this model to various fields' dataset related to intelligent video analysis.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.39 no.4
• /
• pp.188-200
• /
• 2002

The WRR scheduling algorithm is widely used in ATM networks due to its simplicity and the low cost of hardware implementation. It guarantees minimum cell rate according to the weight of each queue. The fairness is a important factor for ABR service. That is, scheduling algorithm allocates network resources fairly to each VC. However, WRR algorithm shows worse performance on bursty traffic. Because it schedules input traffics according to predetermined weight, it can not satisfies fairness criteria, MCR plus equal share and Maximum of MCR or Max-Min share, defined by ATM Forum TM 4.1 specification. The Nabeshima et al algorithm is not adapt to network status rapidly because it is not compensate the weights of unused bandwidth to VCs and assign the unused bandwidth to VCs by RR method. In this paper, we propose a scheduling algorithm for satisfying the two fairness criteria, MCR plus equal share and Maximum of MCR or Max-Min share, among the six criteria defined by ATM Forum TM 4.1 specification. The WRR, Nabeshima et al, and the proposed scheduling algorithms are compared with respect to fairness and convergence time throughout experimental simulation. According to the simulation results, the proposed algorithm shows higher fairness and more rapid convergence than other algorithms.

• Journal of Korean Society of Transportation
• /
• v.20 no.3
• /
• pp.159-168
• /
• 2002

The Highway Capacity Manual(HCM) suggests estimating the average green time for the performance evaluation of the traffic actuated operation and Provides the average green time estimation model. However, the model provides with much room for improvements. This document proposes a new analytical model that overcomes the shortage of the HCM model. The average green times estimated by the HCM model and the proposed model were compared. A computer program using the proposed model was coded for the study, while the ACT348 program was used for the implementation of the HCM model Through the comparison study based on the 1,196 hypothetical simulation data surrogating field data, it was found that the average green times estimated by the proposed model yields much nicer one-to-one linear relationship to the simulation results than the ones from the HCM model in both exclusive-only and shared-permitted cases. The R2 values of the proposed and the HCM models with those cases are 0.90 and 0.56, and 0.86 and 0.57, respectively.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.2
• /
• pp.283-294
• /
• 1996

To operate the automatic devices of manufacturing process more effectively and to solve the needs of the resource sharing, network technology is applied to the unit control devices located in common manufacturing zone and operated by connecting them. This paper introduces deign and test operation of cell system for the management of multivendor PLC by applying to wall paper process based on the manufacturing standardization of CIM. The analysis of designed multivendor PLC group is performed by selecting the parameters such as variation of data packet size and node number of PLC ladder logic program and variation by analyzing the corresponding variation values of token rotation time and waiting time for the system operation. For the method of the analysis,the equation is considering the overhead such as indicated packet service time and transmission safety margin for PLC network connection in M/G/1 queue model,and it is applied to the cell system and PLC group manage-ment for operation .Through the experiment,it is found that the realtime processing is possible with expanded and better result than the resuit obtained by Jayasumana.Browaka with PLC scan time considered,response lower limit of 10-20 msec. data packet size not more than 50 bytes.and the number of nodes less than 40.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.38 no.10
• /
• pp.1-9
• /
• 2001

The MAC (Medium Access Control) protocol is needed for the OLT(Optical Line Termination) to allocate bandwidth to ONUs(Optical Network Units) and ONTs(Optical Network Terminations) dynamically in the ATM PON(Passive Optical Network). With the protocol, the OLT gathers ONUs' informations and provides grants efficiently to each ONU. Two important functions of the MAC protocol is the grant request procedure and the grant distribution algrithm. The latter has the greatest arithmetic portion in the TC(Transmission Convergence) module, occupies a relatively large portion of the overall chip area, has often been the limiting factor in terms of speed, and should be designed to guarantee the quality of service for various traffics. In this paper, we implement the MAC with the FPGA which can allocate grants dynamically according to the queue length information and the number of active ONUs and distribute grants uniformly to minimize the cell delay variation for each ONU. The structure of the MAC scheduler for the dynamic bandwidth assignment has a programmable look-up table. Also, it has a simple structure, the less chip area, and the lower delay time.

• The KIPS Transactions:PartC
• /
• v.9C no.4
• /
• pp.605-614
• /
• 2002

Amount of the web traffic web server handles are explosively increasing, which requires that the performance of the web server should be improved for the various web services. Although the analysis for the HTTP traffic with the proper tuning for the web server is essential, the research relevant to the subject are insignificant. In particular, although most of applications are implemented over HTTP 1.1 protocol, the researches mostly deal with the performance evaluation of the HTTP 1.0 protocol. Consequently, the modeling approach and the performance evaluation over HTTP 1.1 protocol have not been well formed. Therefore, basing on the HTTP 1.1 protocol supporting persistent connection, we present an analytical end-to-end tandem queueing model for web server to consider the specific hardware configuration inside web server beginning at accepting the user request until completing the service. we compare various performances between HTTP 1.0 and HTTP 1.1 under the overloading condition, and then analyze the characteristics of the HTTP traffic that include file size requested to web server, the OFF time between file transfers, the frequency of requests, and the temporal locality of requests. Presented model is verified through the comparing the server throughput according to varying requests rate with the real web server. Thereafter, we analyze the performance evaluation of the web server, according to the interrelation between TCP Listen queue size, the number of HTTP threads and the size of the network buffers.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.12
• /
• pp.167-175
• /
• 2019

Recently, health examination centers have been changing from auxiliary medical facilities to key and independent medical facilities. However, it is not easy to improve medical facilities, including health examination centers, due to the variable characteristics of the relationship between humans and space. Therefore, this study was done to develop a pedestrian-based discrete event simulation analysis program to examine the problems and develop methods for improvement. The program was developed to analyze five evaluation indices and the density of examinees. The problems were derived by analyzing the required time, capacity, and queue size for each examination through simulations. We reduced the examination time and moving distance, increased the capacity, and distributed the queues by adjusting the medical services and relocating the examination rooms. The results were then quantitatively verified by simulations.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.15 no.6
• /
• pp.187-193
• /
• 2015

The ICT healing platform sets a couple of goals including preventing chronic diseases and sending out early disease warnings based on personal information such as bio-signals and life habits. The 2-step open system(TOS) had a relay designed between the healing platform and the storage of personal health data. It also took into account a publish/subscribe(pub/sub) service based on large-scale connections to transmit(monitor) the data processing process in real time. In the early design of TOS pub/sub, however, the same buffers were allocated regardless of connection idling and type of message in order to encode connection messages into a deflate algorithm. Proposed in this study, the dynamic buffer allocation was performed as follows: the message transmission type of each connection was first put to queuing; each queue was extracted for its feature, computed, and converted into vector through tf-idf, then being entered into a k-means cluster and forming a cluster; connections categorized under a certain cluster would re-allocate the resources according to the resource table of the cluster; the centroid of each cluster would select a queuing pattern to represent the cluster in advance and present it as a resource reference table(encoding efficiency by the buffer sizes); and the proposed design would perform trade-off between the calculation resources and the network bandwidth for cluster and feature calculations to efficiently allocate the encoding buffer resources of TOS to the network connections, thus contributing to the increased tps(number of real-time data processing and monitoring connections per unit hour) of TOS.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.6
• /
• pp.17-23
• /
• 2020

In this paper, collecting real-time data frames that occur during RS232 communication between an HMI and PLC of automation equipment by sniffing real-time information data frames through MCU without modification of the HMI or PLC, a method is proposed that allows users to collect data without being dependent on the modification of PLC and HMI systems. The user collects necessary information from the sniffing data through the parsing operation, and the original communication interface is maintained by transmitting the corresponding sniffing frame to the destination. The MCU's UART communication interface circuit is physically designed according to the RS232 communication standard, and this additionally improves efficiency more so than an interrupt-based system by using the DMA device inside the MCU. In addition, the data frame IO operation is performed by logically separating the work of the DMA interrupt service routine from the work of the main thread using the circular queue. Through this method, the user receives the sniffing data frame between the HMI and PLC in RS232 format, and the frame transfer between PLC and HMI arrives normally at the original destination. By sniffing the data frame without further modification of the PLC and HMI, it can be confirmed that it arrives at the user system normally.