ABSTRACT
Contents 9.1 Introduction ................................................................................................................... 222 9.2 Cognitive Radio Meets Multimedia: Introduction ......................................................... 223
9.2.1 Cognitive Radio: Emerging Solution for Spectrum Crisis ................................... 223 9.2.2 CR and Multimedia Communications ............................................................... 224
9.3 Multimedia over CR-Based Networks: Challenges and Problem Statement ................... 225 9.3.1 Multicasting Multimedia Content over CR Networks ........................................ 225 9.3.2 Loss Protection Schemes for Cognitive Multimedia Transmissions..................... 226 9.3.3 Prior Work on Multimedia Transmissions over CR Networks ............................ 226
9.4 Proposed Network Model .............................................................................................. 228 9.4.1 General Analysis ................................................................................................. 228 9.4.2 ULP Framework for CR Networks ..................................................................... 230
9.4.2.1 An Analytical Expression for NPU ..........................................................233 9.4.2.2 An Analytical Expression for Nsc .......................................................... 234 9.4.2.3 An Analytical Expression for [ ]SCE N s ................................................... 234
9.4.3 Equal Loss Protection Scheme for Secondary Transmissions .............................. 236
9.1 Introduction In this chapter, a recurrent use case of multimedia applications over CR networks, namely, video transmission, has been considered. More importantly, we focus on applications requiring a timely delivery of information with a tolerable amount of distortion. We focus on multicasting scenarios where one participant is providing access to a video application directly available to a given population of clients with heterogeneous reception bandwidths and QoS requirements. The stochastic profile of the licensed traffic to coexist with is the Poisson distribution. A special emphasis will be given to ULP and ELP assignment. In particular, a detailed description of the ULP scheme is given. The proposed ULP framework takes advantage of MDC properties, and makes use of the priority encoding transmission (PET) packetization framework of Albanese et al. This technique assigns different amounts FEC to different descriptions according to their importance, and Reed Solomon (RS) codes are employed to generate FEC symbols. The used MDC has the property that each packet acts as an independent block of the original message. For the ELP scheme, we recommend the use of LT codes for multicasting operations as they can generate on the fly as many encoding symbols as are needed to decode the transmitted data. In the case of erroneous transmission, the receiver needs just to wait for more packets to be received instead of acknowledging the transmission and requesting retransmission of the missed packets. Afterward, we develop an analytic expression for the success probability in both cases, which will be used to derive the PSNR formula of the finale stream as a function of the loss pattern. In addition, we recall two PSNR-based maximization algorithms. The outcome of the first algorithm is the length of the FEC coefficients associated with the ULP-based MDC scheme, and the amount of redundancy needed in the ELP case is investigated through the second algorithm, followed by a comparative study between the ULP and ELP modes to identify the scheme that best tackles the packet loss pattern. Later, a number of simulations have been performed to assess the effectiveness of the designed multicast architecture.
