Rateless code based opportunistic multicasting over cognitive radio networks

Abstract

Cognitive radio (CR) represents an exciting new paradigm on spectrum utilization and potentially more bandwidth for exploding multimedia traffic. We focus on the layer encoded video multicast problem over CR and contribute 1) an opportunistic multicasting framework, based on rateless forward error correction (FEC) codes, 2) a mechanism for adaptation of data fragment size to improve transmission efficiency, 3) a joint secondary channel (SC) and video data selection algorithm. By adapting fragment size, tracking video group receiving rate, and adapting transmission parameters, we are able to realize the opportunistic multicasting advantage in a challenging CR environment. The proposed overall SC and video data selection algorithm finds the best rateless FEC code length, physical layer modulation and coding schemes (MCS), fragment size and SC combination to increase the effective throughput and heuristically reach maximum system utility. Favorable results comparing to other algorithms showcase the improved performance.

Atomic fragmentation for efficient opportunistic multicasting over cognitive radio networks

“…In this section, we compare the presented zero waste scheduling (ZWS) scheme to the RCOM scheme [3], which selects fragment sizes from the MCS dependent tile capacities, in order to minimize waste and cross tile dependency and OSAS, as presented in [5].…”

“…Each video has four layers with source rate R v,l = 250kbps. The CR network parameters correspond to PrU setting one in [3]. The differences in schemes settings are listed in TABLE I.…”

“…The outline of the zero waste scheduling (ZWS) scheme, which is using the atomic fragmentation can be found in [3].…”

“…The CR BS is multicasting V video, with L layers each and the layers will be further denoted as (v, l). The BS processes the incoming video-data before transmission, by fragmentation and using rateless forward error control (FEC) [3].…”