• Journal of the Semiconductor & Display Technology
• /
• v.21 no.3
• /
• pp.45-49
• /
• 2022

The fourth industrial revolution, internet of things, and the expansion of online web services have increased an exponential growth and deployment in the number of cloud data centers (CDC). The cloud is emerging as new paradigm for delivering the Internet-based computing services. Due to the dynamic and non-linear workload and availability of the resources is a critical problem for efficient workload and resource management. In this paper, we propose the particle swarm optimization (PSO) based gated recurrent unit (GRU) neural network for efficient prediction the future value of the CPU and memory usage in the cloud data centers. We investigate the hyper-parameters of the GRU for better model to effectively predict the cloud resources. We use the Google Cluster traces to evaluate the aforementioned PSO-GRU prediction. The experimental shows the effectiveness of the proposed algorithm.

• Journal of Broadcast Engineering
• /
• v.20 no.2
• /
• pp.340-347
• /
• 2015

HEVC (High Efficiency Video Coding) achieves high coding efficiency by employing a quadtree-based coding unit (CU) block partitioning structure and various prediction units (PUs), and the determination of the best CU partition structure and the best PU mode based on rate-distortion (R-D) cost. However, the computation complexity of encoding also dramatically increases. In this paper, to reduce such encoding computational complexity, we propose three fast PU mode decision methods based on encoding information of upper depth as follows. In the first method, the search of PU mode of the current CU is early terminated based on the sub-CBF (Coded Block Flag) of upper depth. In the second method, the search of intra prediction modes of PU in the current CU is skipped based on the sub-Intra R-D cost of upper depth. In the last method, the search of intra prediction modes of PU in the lower depth's CUs is skipped based on the sub-CBF of the current depth's CU. Experimental results show that the three proposed methods reduce the computational complexity of HM 14.0 to 31.4%, 2.5%, and 23.4% with BD-rate increase of 1.2%, 0.11%, and 0.9%, respectively. The three methods can be applied in a combined way to be applied to both of inter prediction and intra prediction, which results in the complexity reduction of 34.2% with 1.9% BD-rate increase.

• Journal of Drive and Control
• /
• v.18 no.2
• /
• pp.1-8
• /
• 2021

In this work, a circuit with a hydraulic power unit is formulated as a means of predicting the behavior of the prefill valve in the future. The behavior of the prefill valve can be examined by the measurements of the configured power unit, and the performance is determined by using hydraulic pumps, relief valves, and hydraulic hoses that make up the power unit. In particular, pressure/flow pulsation generated by hydraulic pumps can cause instability in the prefill valve and cause noise-induced degradation of the overall performance and reliability of the hydraulic system containing the prefill valve. Therefore, to study the behavior and performance of the prefill valve in a relatively accurate manner, the prediction of the characteristics of the hydraulic power unit driving the prefill valve is very important. In this study, the pulsation characteristics of the hydraulic pump were analyzed to theoretically demonstrate its relationship with different settings of the power unit, such as relief valve pressure settings and the presence/absence of the hose.

• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2003.10a
• /
• pp.7-16
• /
• 2003

In this paper, we propose a probabilistic framework to predict the interaction probability of proteins. The notion of domain combination and domain combination pair is newly introduced and the prediction model in the framework takes domain combination pair as a basic unit of protein interactions to overcome the limitations of the conventional domain pair based prediction systems. The framework largely consists of prediction preparation and service stages. In the prediction preparation stage, two appearance pro-bability matrices, which hold information on appearance frequencies of domain combination pairs in the interacting and non-interacting sets of protein pairs, are constructed. Based on the appearance probability matrix, a probability equation is devised. The equation maps a protein pair to a real number in the range of 0 to 1. Two distributions of interacting and non-interacting set of protein pairs are obtained using the equation. In the prediction service stage, the interaction probability of a protein pair is predicted using the distributions and the equation. The validity of the prediction model is evaluated fur the interacting set of protein pairs in Yeast organism and artificially generated non-interacting set of protein pairs. When 80% of the set of interacting protein pairs in DIP database are used as foaming set of interacting protein pairs, very high sensitivity(86%) and specificity(56%) are achieved within our framework.

• Asian Pacific Journal of Cancer Prevention
• /
• v.13 no.1
• /
• pp.97-101
• /
• 2012

Objective: With the background of aging population in China and advances in clinical medicine, the amount of operations on old patients increases correspondingly, which imposes increasing challenges to critical care medicine and geriatrics. The study was designed to describe information on the length of ICU stay from a single institution experience of old critically ill gastric cancer patients after surgery and the framework of incorporating data-mining techniques into the prediction. Methods: A retrospective design was adopted to collect the consecutive data about patients aged 60 or over with a gastric cancer diagnosis after surgery in an adult intensive care unit in a medical university hospital in Shenyang, China, from January 2010 to March 2011. Characteristics of patients and the length their ICU stay were gathered for analysis by univariate and multivariate Cox regression to examine the relationship with potential candidate factors. A regression tree was constructed to predict the length of ICU stay and explore the important indicators. Results: Multivariate Cox analysis found that shock and nutrition support need were statistically significant risk factors for prolonged length of ICU stay. Altogether, eight variables entered the regression model, including age, APACHE II score, SOFA score, shock, respiratory system dysfunction, circulation system dysfunction, diabetes and nutrition support need. The regression tree indicated comorbidity of two or more kinds of shock as the most important factor for prolonged length of ICU stay in the studied sample. Conclusions: Comorbidity of two or more kinds of shock is the most important factor of length of ICU stay in the studied sample. Since there are differences of ICU patient characteristics between wards and hospitals, consideration of the data-mining technique should be given by the intensivists as a length of ICU stay prediction tool.

• ETRI Journal
• /
• v.38 no.6
• /
• pp.1064-1073
• /
• 2016

An adaptive speech streaming method to improve the perceived speech quality of a software-based multipoint control unit (SW-based MCU) over IP networks is proposed. First, the proposed method predicts whether the speech packet to be transmitted is lost. To this end, the proposed method learns the pattern of packet losses in the IP network, and then predicts the loss of the packet to be transmitted over that IP network. The proposed method classifies the speech signal into different classes of silence, unvoiced, speech onset, or voiced frame. Based on the results of packet loss prediction and speech classification, the proposed method determines the proper amount and bitrate of redundant speech data (RSD) that are sent with primary speech data (PSD) in order to assist the speech decoder to restore the speech signals of lost packets. Specifically, when a packet is predicted to be lost, the amount and bitrate of the RSD must be increased through a reduction in the bitrate of the PSD. The effectiveness of the proposed method for learning the packet loss pattern and assigning a different speech coding rate is then demonstrated using a support vector machine and adaptive multirate-narrowband, respectively. The results show that as compared with conventional methods that restore lost speech signals, the proposed method remarkably improves the perceived speech quality of an SW-based MCU under various packet loss conditions in an IP network.

• ETRI Journal
• /
• v.15 no.3
• /
• pp.27-34
• /
• 1994

There have been traditional approaches to model radio propagation path loss mechanism both theoretically ad empirically. Theoretical approach is simple to explain and effective in certain cases. Empirical approach accommodates the terrain configuration and distance between base station and mobile unit along the propagation path only. In other words, it does not accommodate natural terrain configuration over a specific area. In this paper, we propose a spatial prediction technique for the mobile radio propagation path loss accommodating complete natural terrain configuration over a specific area. Statistical uncertainty analysis is also considered.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2000.03b
• /
• pp.591-598
• /
• 2000

In the present study, the variation of settlement, pore water pressure and undrained shear strength through model tests were measured. Also, the variation of water content, unit weight and shear strength by the vane shear tests were observed. In this study, appropriate deposit time of construction equipments used in treatment of hydraulic fills is determined from the prediction curve of increased shear strength in dredged fills.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1993.06a
• /
• pp.1013-1016
• /
• 1993

This paper describes a new type of neuron model, the inputs of which are interfered with one another. It has a high mapping ability with only single unit. The learning speed is considerably improved compared with the conventional linear type neural networks. The proposed neuron model was successfully applied to the prediction problem of chaotic time series signal.

• The Journal of Korea Robotics Society
• /
• v.16 no.3
• /
• pp.283-290
• /
• 2021

We propose a task and motion planning algorithm for clearing obstacles and wiping surfaces, which is essential for surface disinfection during the pathogen disinfection process. The proposed task and motion planning algorithm determines task parameters such as grasping pose and placement location during the planning process without using pre-specified or discretized values. Furthermore, to quickly inspect many unit motions, we propose a motion feasibility prediction algorithm consisting of collision checking and an SVM model for inverse mechanics and self-collision prediction. Planning time analysis shows that the feasibility prediction algorithm can significantly increase the planning speed and success rates in situations with multiple obstacles. Finally, we implemented a hierarchical control scheme to enable wiping motion while following a planner-generated joint trajectory. We verified our planning and control framework by conducted an obstacle-clearing and surface wiping experiment in a simulated disinfection environment.