• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.8
• /
• pp.3900-3916
• /
• 2019

The modern mobile devices are typically equipped with multiple network interfaces, e.g., 4G LTE, Wi-Fi, Bluetooth, but the current implementation of TCP can support only a single path at the same time. The Multipath TCP (MPTCP) leverages the multipath feature and provides (i) robust connection by utilizing another interface if the current connection is lost and (ii) higher throughput than single path TCP by simultaneously leveraging multiple network paths. However, if the performance between the multiple paths are significantly diverse, the receiver may have to wait for packets from the slower path, causing reordering and buffering problems. To solve this problem, previous MPTCP schedulers mainly focused on predicting the latency of the path beforehand. Recent studies, however, have shown that the path latency varies by a large margin over time, thus the MPTCP scheduler may wrongly predict the path latency, causing performance degradation. In this paper, we propose a new MPTCP scheduler called, choose fastest subflow (CFS) scheduler to solve this problem. Rather than predicting the path latency, CFS utilizes the characteristics of these paths to reduce the overall flow completion time by redundantly sending the last part of the flow to both paths. We compare the performance through real testbed experiments that implements CFS. The experimental results on both synthetic packet generation and actual Web page requests, show that CFS consistently outperforms the previous proposals in all cases.

• Journal of Multimedia Information System
• /
• v.5 no.1
• /
• pp.59-62
• /
• 2018

The delay between input action and visual interface feedback ("Latency") in a touchscreen inking task reduces the user's performance. When the latency is less than 2.38ms, the user cannot perceive the latency in dragging task. This value is difficult to achieve on recent touchscreens and general purpose computers. So, methods of predicting touch points to reduce perceptible latency has been proposed. In general, touch points prediction is not perfect. When using point prediction, feedback of the predicted points is displayed on the screen, after a while, erased when the actual points are displayed. When this task is implemented by software, it causes additional latency to work to erase predicted points feedback. It therefore propose a platform for rendering point prediction feedback without additional latency by the FPGA. This platform transmits input points and HDMI signals rendering feedback of input points to the FPGA. The FPGA draws the feedback of points predicted based on the input points on the HDMI and displays the screen. Since hardware rendering changes the HDMI signal every frame, it does not require erasing work and rendering can be done within an early time regardless of the amount of rendering, so we will reduce the latency.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.34S no.10
• /
• pp.89-97
• /
• 1997

To extract time-varying evoked potential (EP), segement latency corrected average method is presented. This method is composed of three steps. First, adaptive filtering for reducing the effect of artifacts and removing background noise is performed. Next, validated intervals of individual segments are aligned, and latency components are detected by cross-correlation between the previously obtained and measured EPs within the intervals. Finally, after the detected latency component, responses of segments are groupe and averaged, the shole corrected EP signal is obtained. In the experiments, the resutls of the conventional methods including simple averaging, Woody's method, and peak component latency corrected averaging are obtained, the results compared with the present method for evaluating performance. Therefore, the presented method confirms that it reflects the latency variations of fundamental peaks and gets the improved EP.

• The Journal of the Korea Contents Association
• /
• v.10 no.10
• /
• pp.45-53
• /
• 2010

In this paper, we propose RFID network agent based user mobility support mechanism for secure information transmissions. When a user with RFID tag who enjoys service moves to different attachment points, the user mobility is supported to provide continuously service and the information can be securely transmitted. Although there is an existing study for providing user mobility, it has a problem which the handover latency time is long. To settle this problem, we present a scheme for minimizing handover latency time and compare handover latency time of existing and proposed scheme. We confirm that proposed mechanism has low latency time than existing mechanism by mathematical analysis using the queuing theory and show that proposed mechanism is more superior to the existing mechanism.

• Journal of Biomedical Engineering Research
• /
• v.22 no.3
• /
• pp.259-267
• /
• 2001

The visual evoked potentials(VEPs) is used to assist in the diagnosis of specific disorders associated with involvement of the sensory visual pathways. The P100 latency is an important parameter which is diagnosis of optic nerve disorders. There are characteristics of latency delay, wave distortion, amplitude deduction in abnormal subjects. It is difficult to diagnose in the case of producing peak at the P100 latency. In this paper, difference of pattern between normal VEPs and abnormal VEPs using the Choi-Williams distribution method is studied. We observed the relationship about time and spectrum. The result shown that normal VEPs had maximum spectral value at 20Hz~26.7Hz and abnormal VEPs had maximum spectral value at 16.7Hz~20Hz. Also normal VEPs spectrum is higher than abnormal VEPs spectrum.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.11
• /
• pp.2658-2663
• /
• 2014

To minimize low latency and improve the processing speed of the stock trading system, various technologies have been introduced. However, expensive network equipment has limitation for improving speed of trading system. Also, it is true that there is not much advantage by introducing those kind of systems. In this paper, we propose a low-Latency SPT(Safe Proper Time) scheme for overcoming the stock trading system in a cloud. The proposed method minimizes the CPI in order to reduce the CPU overhead that is based on the understanding of the kernel. and this approach satisfies the data timeliness.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.10
• /
• pp.2608-2617
• /
• 1999

One of the most important functionality that commercial video storage servers should provide is to maximize the number of concurrent streams and to minimize the initial latency of new requests. In this paper, we propose a data placement scheme whose disk read unit size can be twice large than that of conventional striping methods. The proposed scheme can significantly increase the number of concurrent streams, since the ratio of rotational latency time is decreased and the disks are effectively utilized. The disk scheduling scheme we propose guarantees constant initial latency time. We also propose a procedural design method for a storage server by introducing the concept of allowed initial latency. The comparison with previous research shows that the proposed scheme provides better performance.

• Journal of KIISE
• /
• v.42 no.11
• /
• pp.1322-1331
• /
• 2015

According to the hypervisor scheduler, the vCPU (virtual CPU) operates under two states: the running state and the stop state. When the vCPU is in the stop state, incoming events are delayed until that vCPU's state changes to the running state. The latency in handling such events that are sent to the vCPU is regarded as the I/O latency. Since a SMP (symmetric multiprocessing) VM (virtual machine) incorporates multiple vCPUs, the event latency on a SMP VM can vary according to specific vCPU that receives the event. In this paper, we propose a new scheme named event routing that sends events according to the operation state of each vCPU to reduce the event latency on an SMP VM. We implemented the proposed event routing scheme in Xen ARM hypervisor and confirmed the reduction of I/O latency from measuring the network RTT (round trip time) and the TCP bandwidth under a variety of testing conditions. The network RTT decreases by up to 94% and the TCP bandwidth increases up to 35% when compare to native Xen ARM.

• Journal of Internet of Things and Convergence
• /
• v.1 no.1
• /
• pp.1-4
• /
• 2015

This paper aims to present the mathematical model to find the mean web object transfer latency in the slow-start phase of TCP congestion control mechanism, which is one of the main control techniques of Internet. Mean latency is an important service quality measure of end-user in the network. The application area of the proposed latency model is the narrowband environment including multi-hop wireless network and Internet of Things(IoT), where packet loss occurs in the slow-start phase only due to small window. The model finds the latency considering initial window size and the packet loss rate. Our model shows that for a given packet loss rate, round trip time and initial window size mainly affect the mean web object transfer latency. The proposed model can be applied to estimate the mean response time that end user requires in the IoT service applications.

• Journal of Biomedical Engineering Research
• /
• v.12 no.3
• /
• pp.185-190
• /
• 1991

An automatized latency calculation method of an e-OAE(evoked otacoustic emission) is proposed. The e-OAE signal measured from a normal adult is averaged 1000 times to remove noises. This averaged signal is converted to digital signal and that is processed by IBM-AT computer for latency calculation. we separate the stimulated and the emitted signal on the time domain by a modified LPC (linear prediction coding) spectrum, and the latency is calculated by cross-correlation method. By proposed latency calculation method the latency is 7.9[ms] for normal adult. The performance of the proposed method is also compared with that of the auto-correlation and cross performance of the proposed method is also compared with that of the auto-correlation and cross-correlation method. The result show that the proposed method has same precision with the conventional methods and can automatically calculate latency without subjective observation.