An adaptive PID neuron network controller for congestion control

Li Yu; Zibo Shi; Yantai Shu

doi:10.1117/12.733055

10 September 2007 An adaptive PID neuron network controller for congestion control

Li Yu, Zibo Shi, Yantai Shu

Proceedings Volume 6773, Next-Generation Communication and Sensor Networks 2007; 67730J (2007) https://doi.org/10.1117/12.733055
Event: Optics East, 2007, Boston, MA, United States

Abstract

We propose a congestion controller based on the Proportional-Integral-Differential Neuron Network (PIDNN). As existing controllers, our controller employs the queue size in bottleneck link router as a congestion indicator to trigger packet dropping. The target queue length and the feedback, actual queue length, act as the controller's two input signals. The packet dropping probability is computed by PIDNN controller with its simple embedded algorithm in term of the predefined state function and output function. Thus, the dropping probability decides to drop or to accept an incoming packet so that the queue length is kept at (or near) the target level. This controller's performance is examined under various network configurations, and compared to proposed congestion algorithms, including PI and RED. Our simulation results show that, with comparable simple implementation, this scheme has short response time, better robustness, and more adaptability, especially under highly dynamic network and heavy traffic load.

Citation Download Citation

Li Yu, Zibo Shi, and Yantai Shu "An adaptive PID neuron network controller for congestion control", Proc. SPIE 6773, Next-Generation Communication and Sensor Networks 2007, 67730J (10 September 2007); https://doi.org/10.1117/12.733055

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available