D2DLive: Iterative live video streaming algorithm for D2D networks

Abstract

Smart healthcare system incorporates new technologies making healthcare more efficient and more convenient. Virtual consultations provide interactions with healthcare professionals via video technology, which lessens the number of patients visiting health facilities and consequently reduces the risk of infections. Recently, device to device (D2D) communication showed extraordinary abilities to save the available resources and improve the network quality of service (QoS). Some of the existing algorithms for multimedia services over D2D networks consider only the signal to noise ratio (SNR) and ignore temporal requirements, which do not provide optimum performances. So, more efficient transmission mechanisms that improve the users’ QoS are needed to respond quickly to the ever-changing application demands. In response to these challenges, the research community began exploring novel solutions for video streaming, namely: Quantile-based Carrier-sense multiple access (CSMA), Flexible video transmission (Flexi) and Distributed Random approach (DR). However, these solutions introduce higher throughput, and jitter, especially in the case of live streaming. Other solutions are also available in the literature, but most of them cannot efficiently scale to large multi videos, multi-hops multi-user live video streaming. In particular, in most of these techniques, the chunk number, node transmission capacity or specific video delay, and time requirements are not taken into consideration in the path optimization.To overcome these shortcomings, we propose a novel D2DLive video streaming algorithm as a new solution for a higher QoS to improve medical data delivery. For each requested video, the proposed algorithm divides its content into small playable units called chunks that are transmitted using a new network selection algorithm. The proposed algorithm computes iteratively a suboptimal set of paths to be used to forward all the video chunks cooperatively to their destinations. Furthermore, we develop a new weight algorithm to evaluate the performance of the proposed algorithm in adhering to the upload and download capacities for each device while minimizing the transmission delay. Simulation results show that the proposed algorithm outperforms other established methods in terms of full delay, average upload, and download capacities.