›› 2014, Vol. 53 ›› Issue (5): 73-78.

Previous Articles     Next Articles

USB Bulk Transfer Research Based on SARSA Learning Algorithm

ZHANG Qiuyun  ,JIANG Hong   

  1. School of Information Engineering, Southwest University of Science and Technology, Mianyang 621010, China
  • Received:2013-11-25 Online:2014-09-25 Published:2014-09-25

Abstract: USB is widely used in PC peripherals, and the increase of transmission data results in increasingly high requirement on transmission efficiency. However, there is a serious waste of USB bandwidth resources, due to the USB system software and characteristics of devices. A USB bulk transfer transaction scheduling method with SARSA learning algorithm for this problem, intelligently allocating the transactions in each frame based on current environment is designed. Simulation results show that this method significantly improves the utilization effectiveness of USB bandwidth and the throughput, compared to the method of the USB system, in the case of multiple bulk transfers.

Key words: USB, SARSA, bulk transfer, utilization effectiveness of USB bandwidth

CLC Number: