• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.11
• /
• pp.5243-5263
• /
• 2017

Data center networks provide critical bandwidth for the continuous growth of cloud computing, multimedia storage, data analysis and other businesses. The problem of low link bandwidth utilization in data center network is gradually addressed in more hot fields. However, the current scheduling strategies applied in data center network do not adapt to the real-time dynamic change of the traffic in the network. Thus, they fail to distribute resources due to the lack of intelligent management. In this paper, we present an improved adaptive traffic scheduling strategy utilizing the simulated annealing genetic algorithm (SAGA). Inspired by the idea of software defined network, when a flow arrives, our strategy changes the bandwidth demand dynamically to filter out the flow. Then, SAGA distributes the path for the flow by considering the scheduling of the different pods as well as the same pod. It is implemented through software defined network technology. Simulation results show that the bisection bandwidth of our strategy is higher than state-of-the-art mechanisms.

• The Journal of the Korea Contents Association
• /
• v.21 no.7
• /
• pp.164-170
• /
• 2021

In VLSI design, partitioning is a task of clustering objects into groups so that a given objective circuit is optimized. It is used at the layout level to find strongly connected components that can be placed together in order to minimize the layout area and propagation delay. The most popular algorithms for partitioning include the Kernighan-Lin algorithm, Fiduccia-Mattheyses heuristic and simulated annealing. In this paper, we propose a adapted genetic algorithm searching solution space for the circuit partitioning problem, and then compare it with simulated annealing and genetic algorithm by analyzing the results of implementation. As a result, it was found that an adaptive genetic algorithm approaches the optimal solution more effectively than the simulated annealing and genetic algorithm.

• Proceedings of the KIEE Conference
• /
• 2004.11c
• /
• pp.162-164
• /
• 2004

This paper proposes a new classification system combining the adaptive feature weighting algorithm using the genetic algorithm and the modified KNN rule. GA is employed to choose the middle value of weights and weights of features for high performance of the system. The modified KNN rule is proposed to estimate the class of test pattern using adaptive feature space. Experiments with the unconstrained handwritten digit database of Concordia University in Canada are conducted to show the performance of the proposed method.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.2 no.1
• /
• pp.59-64
• /
• 2002

In the sender-initiated load balancing algorithms, the sender continues to send an unnecessary request message fur load transfer until a receiver is found while the system load is heavy. Meanwhile, in the receiver-initiated load balancing algorithms, the receiver continues to send an unnecessary request message for load acquisition until a sender is found while the system load is light. These unnecessary request messages result in inefficient communications, low CPU utilization, and low system throughput in distributed systems. To solve these problems, in this paper, we propose a genetic algorithm based approach fur improved sender-initiated and receiver-initiated load balancing. The proposed algorithm is used for new adaptive load balancing approach. Compared with the conventional sender-initiated and receiver-initiated load balancing algorithms, the proposed algorithm decreases the response time and increases the acceptance rate.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.37 no.3
• /
• pp.248-254
• /
• 2009

The use of evolutionary algorithm is limited in the field of aerodynamics, mainly because the population-based search algorithm requires excessive CPU time. In this paper a coupling method with adaptive range genetic algorithm for floating point and back-propagation neural network is proposed to efficiently obtain a converged solution. As a result, it is shown that a reduction of 14% and 33% respectively in wave drag and its consumed time can be achieved by the new method.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.11a
• /
• pp.157-160
• /
• 2000

This paper deals with the fuzzy modeling for the complex and uncertain system in which conventional and mathematical models may fail to give satisfactory results. mGA(messy Genetic Algorithm) has more effective and adaptive structure than sGA with respect to using changeable-length string and VEGA(Virus Evolution Genetic) Algorithm) can search the global and local optimal solution simultaneously with reverse transcription operator and transduction operator. Therefore in this paper, the optimal fuzzy model is obtained using Virus-messy Genetic Algorithm(Virus-mGA). In this method local information is exchanged in population so that population may sustain genetic divergence. To prove the surperioty of the proposed approach, we provide the numerical example.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.5
• /
• pp.663-668
• /
• 2018

A combined CPG (Central Pattern Generator) based foot trajectory and GA (Genetic Algorithm) based joint compensation method is presented for adaptive humanoid walking. In order to increase an adaptability of humanoid walking for rough terrains, the experiment for rolling terrains are introduced. The CPG based foot trajectory method has been successfully applied to basic slops and variable slops, but has a limitation for the rolling terrains. The experiments are conducted in an ODE based Webots simulation environment using humanoid robot Nao to verify a stability of walking for various rolling terrains. The proposed method is compared to the previous CPG foot trajectory technique and shows better performance especially for the cascade rolling terrains.

• Journal of Power System Engineering
• /
• v.18 no.5
• /
• pp.129-136
• /
• 2014

This paper proposes improvement on sensorless vector control performance of a permanent magnet synchronous motor (PMSM) with sliding mode observer. An adaptive observer gain and second order cascade low-pass filter (LPF) were used to improve the estimation accuracy of the rotor position and speed. The adaptive observer gain was applied to suppress the chattering intensity and obtained by using the Lyapunov's stability criterion. The second order cascade LPF was designed for the system to escalate the filtering performance of the back-emf estimation. Furthermore, genetic algorithm was used to optimize the system PI controller's performance. Simulation results showed the effectiveness of the suggested improvement strategy. Moreover, the strategy was useful for the sensorless vector control of PMSM to operate on the low-speed area.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.234-234
• /
• 2000

A direct torque control(DTC) based sensorless speed control system which employs a new closed loop flux observer is proposed. The flux observer takes an adaptive scheduling gains where motet speed is used as the scheduling variable. Adaptive nature comes from the fact that the estimated values of stator resistance and speed are included as observer parameters. The parameters of the PI controllers adopted in the adaptive law for the estimation of stator resistance and motor speed are determined by simple genetic algorithm. Simulation results in low speed region are given for comparison between proposed and conventional flux estimate scheme.

• Proceedings of the KIEE Conference
• /
• 2001.07d
• /
• pp.2693-2695
• /
• 2001

In this paper, we present an adaptive image filter using genetic algorithm. The filter is robust to the characteristic variance of image and noise, by evolving the parameter and combination of image preprocessors properly. And we have adopted adaptive mutation strategy, which use different mutation rate for specific region of chromosome. The filter is implemented on FPGA board and controlled by host PC.