An Intelligent Video Streaming Technique in Wireless Communications

This paper is concerned with an intelligent application of Moving Picture Expert Group (MPEG) video transmission over a Bluetooth Asynchronous Connection- Less (ACL) link. MPEG Variable Bit Rate (VBR) video is data hungry and presents excessive time delay and data loss over a wireless communication. In a Bluetooth channel transmission rate could also be irregular, due to i nterferences caused by Bluetooth channel noises or other nearby wireless devices. Th erefore, it is difficult to transmit MPEG VBR video over Bluetooth without time delay or data loss. A buffer entitled 'traffic-shaper' is introduced before the Host Cont roller Interface (HCI) of the Bluetooth protocol stack to prevent excessive overflow of MPE G video data over the network. This paper presents a novel Self-tuning Neuro-Fuzzy (SNF) scheme to monitor the traffic- shaping buffer output rate, in order to make sure t he video stream conforms to the traffic conditions of the Bluetooth ACL link. The Matlab-Si mulink computer simulation results demonstrate that the use of the SNF control scheme reduces time delay and data loss, as compared with a conventional MPEG VBR video transmission over Bluetooth.