Download or read book Cross-layer Framework for Wireless Video Communication Over USRP written by Praveen Janarthanan and published by . This book was released on 2013 with total page 44 pages. Available in PDF, EPUB and Kindle. Book excerpt: It is well known that the modern day wireless networks involve a significant amount of video traffic. The present day physical layer standards support high bit-rates so that the HD video could be streamed. But though the bit-rates are high, sudden channel errors are unexpected and unavoidable. This reflects on the video quality very badly. This means there is still room for intelligent solution for efficient video transfer. Cross-layer approach is one of the solution. Due to presence of frame dependencies in video the frames are prioritized and assigned a relative priority index values which is used to provide unequal error protection. Also the transmission deadlines of each frame is different. Thus a time based adaptive MAC layer re-transmission is proposed in this thesis. In this thesis we mainly focus on developing a cross-layer framework for wireless unicast video transmission on software defined radios (SDR). SDR are being very flexible in implementing MAC and PHY layers, helps to easily implement a cross-layer design too. A MAC layer centric cross-layer approach is adopted here. This approach involves choosing an appropriate modulation scheme with the known knowledge of channel state available from PHY layer at the receiver, and relative priority index (RPI) of each video frame from the application layer and adapting MAC re-transmission based on transmission deadline (Td) of the video packet from APP layer. An in-band cross-layer signaling is adopted here, in which RPI and Td are written in the header of the rtp packet. The SNR values from receiver are sent back to the sender along with the MAC acknowledgment. A content aware time based adaptive re-transmission (TAR) and a heuristic modulation selection algorithm are implemented using this framework. MAC and PHY layers are developed on GNU radio with USRPs N210s. Evalvid tool at the application layer for video streaming and evaluation. Results show that the gain in the efficiency is mostly due to the TAR mechanism.