EMUNE: Architecture for Mobile Data Transfer Scheduling with Network Availability Predictions

Adaptive schedulers for deadline-constrained content upload from mobile multihomed vehicles

“…Rahmati et al [3] present a technique for estimating and learning the Wi-Fi network conditions from a fixed node. Rathnayake et al [4] demonstrate how a prediction engine may be capable of forecasting future network and bandwidth availability, and propose a utility-based scheduling algorithm which uses the predicted throughput to schedule the data transfer over multiple interfaces from fixed nodes. These works heavily rely on channel performance predictions, and consider scheduling at the packet-level, i.e., which packet to send through which interface, to maximize the total throughput.…”

“…This would be the solution typically adopted on related works that aim at maximizing the total throughput [3], [4], [5]. We use the solution provided by the Greedy-in-Cost (GC) [1] heuristic as cost lower bound to evaluate the performance of the adaptive scheduler, while we evaluate the cost saving by comparison against the GT heuristic.…”

A framework for mobility prediction and high bandwidth utilization to support mobile multimedia streaming

“…Many frameworks for mobility prediction and network resources management have been proposed in the recent literature [4][5][6][7][8]. In [4], the authors proposed a framework that integrates user mobility prediction models with resource availability prediction models to keep a constant or less fluctuating streaming rate and to ultimately ensure steady QoE (Quality of Experience).…”

“…In their framework, UbiPaPaGo and UbiHandoff are located in the application service provider; so, all the operations are performed by the network system; this option causes an overload of work. Rathnayake et al [6] presented an architecture which enables client side decision making, taking into consideration network and bandwidth availability predictions, user preferences and the application requirements to optimally schedule data transfers while taking into account the uncertainty in predictions. Their proposed architecture consists of an API (Application Interface), a transport service and two main functional units (i.e., a prediction engine and a scheduling engine).…”

“…All in all, existing frameworks have one or more of the following limitations: (1) they do not take into account users' aggregation in users' mobility prediction model [4][5][6][7][8]; (2) all the operations are located in the network system [5]; which adds high burden to the network workload; (3) they are limited to only non-or near-real-time applications [6]; (4) they are limited to only next cell prediction [5][6][7][8]; (5) they do not identify the equipments (e.g., users' equipments or network system) which perform prediction schemes and maintain the databases [7,8]. In this paper, we propose a framework that proposes solutions to these limitations.…”