Bandwidth aggregation techniques in heterogeneous multi-homed devices: A survey

Abstract

The widespread deployment of various networking technologies, coupled with the exponential increase in end-user data demand, have led to the proliferation of multi-homed, or multi-interface enabled, devices. These trends drove researchers to investigate a wide spectrum of solutions, at different layers of the protocol stack, that utilize available interfaces in such devices by aggregating their bandwidth. In this survey paper, we provide an overview and examine the evolution of bandwidth aggregation solutions over time. We begin by describing the bandwidth aggregation problem. We then investigate the common features of proposed bandwidth aggregation systems and break them down into two major categories: layer-dependent and layer-independent features. Afterwards, we discuss the evolution trends in the literature and discuss some open challenges requiring further research. We end the survey with a brief presentation of related work in tangential research areas.

An efficient bandwidth aggregation algorithm using game theory for multimedia transmission

“…Similarly, at the Medium Access Layer various bandwidth aggregation solutions like Multi-radio Unification Protocol [7] have been proposed. Habak et al [8] surveyed various Multipath data transmission and Bandwidth Aggregation at different layers. They highlighted several challenges and some open research problems in bandwidth aggregation.…”

MPTCP Tunnel: An Architecture for Aggregating Bandwidth of Heterogeneous Access Networks

“…In this class, bandwidth aggregation is carried out at the mobile end with multiple network interfaces [14][15][16][17], especially in wireless access networks. Nevertheless, most of the hosts connected to the 4 Wireless Communications and Mobile Computing fixed access network do not have multiple-network-access capability.…”

Bandwidth Aggregation Across Multiple Smartphone Devices

“…There are several more complex scheduling algorithms described in [7], however specific discussion of these techniques is not within the scope of this thesis. The methods mentioned above are meant to highlight the importance of scheduling.…”

“…[7] describes granularity as the unit of data (referred to as a scheduling-unit for the remainder of this paper) that can be assigned to a network interface. In a broad sense, we have two options: packet-level or connection-level.…”

“…[7] identifies the features offered by all existing solutions regardless of the layer in which they are implemented: interface characteristics estimation, application characteristics estimation, scheduling and communication patterns.…”