Transactions of the Institute of Systems, Control and Information Engineers
Online ISSN : 2185-811X
Print ISSN : 1342-5668
ISSN-L : 1342-5668
Volume 25 , Issue 2
Showing 1-3 articles out of 3 articles from the selected issue
  • Shinsuke Satake, Hiroshi Inai, Tomoya Saito, Tsuyoshi Arai
    2012 Volume 25 Issue 2 Pages 19-27
    Published: 2012
    Released: May 16, 2012
    Most browsers establish multiple connections and download files in parallel to reduce the response time. On the other hand, a web server limits the total number of connections to prevent from being overloaded. That could decrease the response time, but would increase the loss probability, the probability of which a newly arriving client is rejected. This paper proposes a connection admission control method which accepts only one connection from a newly arriving client when the number of connections exceeds a threshold, but accepts new multiple connections when the number of connections is less than the threshold. Our method is aimed at reducing the response time by allowing as many clients as possible to establish multiple connections, and also reducing the loss probability. In order to reduce spending time to examine an adequate threshold for web server administrators, we introduce a procedure which approximately calculates the loss probability under a condition that the threshold is given. Via simulation, we validate the approximation and show effectiveness of the admission control.
    Download PDF (528K)
  • Kazuyoshi Hatada, Kentaro Hirata, Kohei Yamada, Shinji Otani
    2012 Volume 25 Issue 2 Pages 28-38
    Published: 2012
    Released: May 16, 2012
    This paper considers power assist control methods for periodic motions. In most power assist applications, additional force is applied in proportion to the instantaneous value of force generated by human. However, human persistent tasks are sometimes periodic and it can be shown that the proportional assist is not optimal for such motions in the sense of energy efficiency. Therefore, we propose power assisting methods to suppress the velocity fluctuation similar to the repetitive control. We compare the effectiveness and the energy efficiency of each method through numerical simulation and experiments.
    Download PDF (2538K)
  • Takayuki Wada, Takamitsu Itani, Yasumasa Fujisaki
    2012 Volume 25 Issue 2 Pages 39-44
    Published: 2012
    Released: May 16, 2012
    Stopping rules are developed for stochastic approximation which is an iterative method for solving an unknown equation based on its consecutive residuals corrupted by additive random noise. It is assumed that the equation is linear and the noise is independent and identically distributed random vectors with zero mean and a bounded covariance. Then, the number of iterations for achieving a given probabilistic accuracy of the resultant solution is derived, which gives a rigorous stopping rule for the stochastic approximation. This number is polynomial of the problem size.
    Download PDF (259K)