Font Size: a A A

Compressed Sensing Based Wireless Video Multicast

Posted on:2015-01-30Degree:MasterType:Thesis
Country:ChinaCandidate:H ChenFull Text:PDF
GTID:2268330428977634Subject:Circuits and Systems
Abstract/Summary:PDF Full Text Request
Recently, with the wireless network and multimedia technologydevelopment, the wireless video multicast services have increased rapidly, suchas mobile TV, media sharing, broadcasting of sport events, and hot news inairport, it has become an inevitable demand to the real-time transmission.However, in the wireless multicast, since these users have different channelcharacteristics, including the bit rate they can support and the packet-loss ratewhich they are suffering from, they rarely get expected performance suitable fortheir individual channels. Compressed sensing/compressive sampling (CS) is anew method of signal processing. The CS measurements without structuralcharacteristics, which means that each data have equal significance in thereconstruction. The CS shows huge space in the wireless video multicast.This paper proposes a multicast system based on compressed sensing. Atthe encoder, the same video stream is transmitted to all multiple users. Manyheterogeneous receivers face the different channel condition and the differentreception. The receiver with good channel will receive more measurements soas to recover a better quality without interfering with others, which guaranteesthe system with a graceful degradation.The main works are as follows:1. This paper proposes an improved compressed-sensing-based wirelessvideo multicast (iCS-WVM) technique. At the encoder, we partition each frameapply BCS to get a number of measurements.Then, All measurement data arepacked and then transmitted over the channel with white Gaussian noise andpacket losses. At the decoder side, each receiver reconstructs the signal with thereceived packets and the Motion Estimation. The experiment shows that thereconstruction of signal with a linear relationship to the number of packets,besides, CS shows a good robustness in the wireless video multicastenvironment.2. A distributed compressed-sensing-based video multicast, DCS-cast, isproposed in this paper. To further improve the reconstructed quality of video recovery, DCS-Cast based on the iCS-WVM, added the Orthogonal FrequencyDivision Multiplex (OFDM) channels and the Least Squares linear algorithm(LSLE).The result shows that LLSE with a good effect in the low channelcondition. DCS-Cast has the graceful degradation and robustness in thereconstruction.
Keywords/Search Tags:Compressed Sensing, Video Multicast, Graceful degradation, heterogeneous receivers, robust
PDF Full Text Request
Related items