Opportunistic real-time video transmission over fading cognitive radio channels using hybrid digital analog coding
MetadataShow full item record
Cognitive radio is a wireless communication paradigm that has achieved greater research attention due to the increasing demand for high-quality multimedia services and scarcity of the available spectrum to support these services. This paradigm allows unlicensed (secondary) users to access the unused spectrum portion of the licensed (primary) users dynamically. However, an unlicensed user faces a high level of uncertainty because of random arrivals of the licensed users, a varying channel quality, and variability in terms of number and capacity of the available channels. On the other hand, fixed-rate source-channel single-layer coding in the traditional wireless communication systems suffers greatly from the cliff effect and the lack of scalability. Therefore, most conventional wireless transmission schemes may not perform well in a cognitive radio environment which requires the source encoder to be adaptive to channel variations and the arrival-rate of the licensed users. This thesis proposes a new approach to reliable transmission of real-time video over cognitive radio channels based on a hybrid digital analog (HDA) joint source-channel (JSCC) superposition coding scheme. The proposed scheme is capable of adapting itself to channel variations, primary-user arrival-rates, and exploits both the high coding gain of a digital video codec and the inherent capability of analog coding to adapt to the time-varying channel signal-to-noise ratio (CSNR) of a wireless channel. Experimental results demonstrate that the proposed HDA coding approach is far superior to the digital layered coding (DLC), single-layer digital coding, and pure analog coding in terms of decoded video quality, quality scalability, robustness to imperfect channel estimation, and transmission power utilization. It has been observed that HDA coding produces a video quality which is on the average 2.5-4.0 dB higher in peak signal-to-noise ratio (PSNR) compared to single-layer digital coding and as much as 15-19 dB higher compared to pure analog coding respectively. Also, upon close observation of the experimental comparisons between the HDA and DLC systems reveals that the HDA approach achieves up to 2-5 dB improvements in PSNR than the DLC system when intensive motion is present between video frames.