• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.5
• /
• pp.15-21
• /
• 2009

An Ethernet passive optical network (EPON) is an economical and efficient access network that received significant research attention in recent year. A MAC protocol of the PON, the next generation access network, is based primarily on TDMA (Time Division Multiple Access). In this paper, we addressed the problems of a dynamic bandwidth allocation (DBA) in Fairness and QoS Performance. We augmented the advanced bandwidth allocation algorithms to support efficient resource utilization by reducing the unused remaining bandwidth made by idle state ONUs. Our new proposed advanced bandwidth allocation algorithm can allocate effectively and fairly the bandwidths between end-users.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.8 no.4
• /
• pp.26-33
• /
• 2000

In-cylinde flow and mixture formation are key contributors to both idle stability and combustion stability at part load condition in SI engine. The real time measurements of air-fuel ration and in- cylinder residual gas fraction are particularly important to obtain a better understanding of the mechanisms for combustion and emissions especially during cold start and throttle transient condition. This paper reports the cycle resolved measurements of residual gas fraction and equivalence ration near speak plug with value timing change and their effects on combustion characteristics at part load. The results showed that the effect of intake value opening on the residual gas fraction was smaller than that of exhaust valve closing because of the decreases of exhaust gas reverse flow from exhaust port. The variation of equivalence ratio near spark plug increased with the increase of value overlap and it closely related with heat release rate and combustion stability

• Journal of Information Technology Services
• /
• v.10 no.3
• /
• pp.237-250
• /
• 2011

Recently, Wireless Sensor Networks (WSN) require energy performance and guaranteed delivery delay time, contrarily with previous MAC protocols that aim to high throughput mostly. In order to satisfy the new significant requirements, many MAC protocols of WSN employ and try to enhance the duty cycle mechanism which is energy efficient technique in MAC layer. This duty cycle mechanism is oriented by toggling the transceiver conditions composed of wakeup and sleep states. The synchronous MAC protocols perform the period synchronization process. Hence, these are energy efficient in periodic monitoring environment, but are inefficient in where an event is incurred rarely and infrequently. Otherwise, the performance of asynchronous MAC protocols are contrarily with synchronous protocols. In this paper, we design two models consisting Always-busy and Always-idle ti simplify the general network congestion conditions. Through these models, moreover, we analyze two types MAC protocols in terms of energy efficiency and delay performance by analytical results. Additionally, we also evaluate two MAC protocols with two gongestion models that we designed. By the analytical and simulated results, we provide the general and efficient decision method in which protocols are more appropriate in a certain WSN environment.

• Industrial Engineering and Management Systems
• /
• v.12 no.1
• /
• pp.51-62
• /
• 2013

This paper considers a single machine rescheduling problem whose original (efficiency related) objective is minimizing makespan. We assume that disruptions such as order cancelations and newly arrived orders occur after the initial scheduling, and we reschedule this disrupted schedule with the objective of minimizing a disruption related objective while preserving the original objective. The disruption related objective measures the impact of the disruptions as difference of completion times in the remaining (uncanceled) jobs before and after the disruptions. The artificial due dates for the remaining jobs are set to completion times in the original schedule while newly arrived jobs do not have due dates. Then, the objective of the rescheduling is minimizing the maximum earliness without tardiness. In order to preserve the optimality of the original objective, we assume that no-idle time and no tardiness are allowed while rescheduling. We first define this new problem and prove that the general version of the problem is unary NP-complete. Then, we develop three simple but intuitive heuristics. For each of the three heuristics, we find a tight bound on the measure called modified z-approximation ratio. The best theoretical bound is found to be 0.5 - ${\varepsilon}$ for some ${\varepsilon}$ > 0, and it implies that the solution value of the best heuristic is at most around a half of the worst possible solution value. Finally, we empirically evaluate the heuristics and demonstrate that the two best heuristics perform much better than the other one.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.2
• /
• pp.270-276
• /
• 2018

Apache Spark is one of the high performance in-memory computing frameworks for big-data processing. Recently, to improve the performance, general-purpose computing on graphics processing unit(GPGPU) is adapted to Apache Spark framework. Previous Spark-GPGPU frameworks focus on overcoming the difficulty of an implementation resulting from the difference between the computation environment of GPGPU and Spark framework. In this paper, we propose a Spark framework based on a heterogenous pipeline computing with OpenCL to further improve the performance. The proposed framework overlaps the Java-to-Native memory copies of CPU with CPU-GPU communications(DMA) and GPU kernel computations to hide the CPU idle time. Also, CPU-GPU communication buffers are implemented with switching dual buffers, which reduce the mapped memory region resulting in decreasing memory mapping overhead. Experimental results showed that the proposed Spark framework based on a heterogenous pipeline computing with OpenCL had up to 2.13 times faster than the previous Spark framework using OpenCL.

• Journal of Korea Multimedia Society
• /
• v.3 no.3
• /
• pp.320-332
• /
• 2000

Web consists of the grand virtual system which is made of all connected computers network. We can solve the huge problem which requires high quality in cost performance and powerful computing power to use a numerous idle state system on internet as process it parallel. However, we have to consider heterogeneous computing resources, accessibility, and reliability to carry out parallel system on global environment, not network but whole Internet. In this paper, We the WebImg system which has the power of web computing, and show the flexible task allocation strategy in heterogeneous hosts. Also, we evaluate its performance, moreover the proposed task allocation strategy supplies fault tolerance by controlling host situation at any time.

• Journal of Multimedia Information System
• /
• v.2 no.2
• /
• pp.233-240
• /
• 2015

Mobile devices are becoming increasingly sophisticated and the latest generation of smartphones now incorporates many diverse and powerful sensors. These sensors include acceleration sensor, magnetic field sensor, light sensor, proximity sensor, gyroscope sensor, pressure sensor, rotation vector sensor, gravity sensor and orientation sensor. The availability of these sensors in mass-marketed communication devices creates exciting new opportunities for data mining and data mining applications. In this paper, we describe and evaluate a system that uses phone-based accelerometers to perform activity recognition, a task which involves identifying the physical activity that a user is performing. To implement our system, we collected labeled accelerometer data from 10 users as they performed daily activities such as "phone detached", "idle", "walking", "running", and "jumping", and then aggregated this time series data into examples that summarize the user activity 5-minute intervals. We then used the resulting training data to induce a predictive model for activity recognition. This work is significant because the activity recognition model permits us to gain useful knowledge about the habits of millions of users-just by having them carry cell phones in their pockets.

• Journal of the military operations research society of Korea
• /
• v.17 no.1
• /
• pp.61-83
• /
• 1991

There has been a great deal of research dealing with the optimal replacement of stochastically deteriorating equipment and research in queueing systems with a finite calling population. However. there has been a lack of research which combines these two areas to deal with optimal replacement for a fixed number of machines and a limited number of repairmen. In this research, an optimal control policy for replacement involving two machines and one repairman is developed by investigating a class of age replacement policies in the context of controlling a G/M/1 queueing system with a finite calling population. The control policy to be imposed on this problem is an age-dependent control policy, described by the control limit $t^{\ast}$. The control limit is operational only when the repairman is idle; that is. if both machines are working, as soon as a machine reaches the age $t^{\ast}$ it is taken out of service for replacememt. We obtain the ${\epsilon}$-optimal control age which will minimize the long-run average system cost. An algorithm is developed that is applicable to general failure time distributions and cost functions. The algorithm does not require the condition of unimodality for implementation.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.9 no.12
• /
• pp.1353-1358
• /
• 2014

Tremendous amount of dummy packets are generally generated for faking over a wireless sensor network so as to keep the location privacy of nodes on the communication paths against the global eavesdropping. In this paper, a scoped-flooding protocol is designed for transferring data between each source and mobile sink(aka, basestation) where, the only nodes within the scope are allowed to issue dummy packets at every idle time so that the location privacy of the nodes on the paths is kept and the amount of dummy packets is reduced to the extend of the flooding scope. The size of the flooding diameter can be taken into consideration of the privacy level and the communication cost. We design a detailed specification of the protocol and verify several properties.

• Journal of the Korean Ceramic Society
• /
• v.44 no.10
• /
• pp.574-579
• /
• 2007

Porous $Al_2O_3$ ceramics were prepared by the gelcasting foams method (a slurry foaming process) with acrylamide monomer. The foaming and gelation behavior was investigated with the parameters such as the type and concentration of surfactant, solid loading of slurry, and the concentrations of initiator and catalyst. Density, porosity, microstructure, and strength of the green and sintered samples were characterized. Of the four kinds of surfactants tested, Triton X-114 showed the highest foaming ability for the solid loading of 55-30 vol%. The gelation condition giving the idle time off min was found to set the foamed structure without significant bubble enlargement and liquid lamella thinning. The green samples were fairly strong and machinable and showed maximum strength of 2.4 MPa in diametral compression. The sintered samples showed densities of 10-36% theoretical (i.e. porosity 90-64%) with a highly interconnected network of spherical pores with sizes ranging from 30 to $600{\mu}m$. The pore size and connectivity increased but the cell strut thickness decreased with decreasing the solid loading. Flexural strength of 37.8-1.7 MPa was obtained for the sintered samples.