Multipath forwarding consists of using multiple paths simultaneously to transport data over the network. While most such techniques require endpoint modifications, we investigate how multipath forwarding can be done inside the network, transparently to endpoint hosts. With such a network-centric approach, packet reordering becomes a critical issue as it may cause critical performance degradation. We present a Software Defined Network architecture which automatically sets up multipath forwarding, including solutions for reordering and performance improvement, both at the sending side through multipath scheduling algorithms, and the receiver side, by resequencing out-of-order packets in a dedicated in-network buffer. We implemented a prototype with commonly available technology and evaluated it in both emulated and real networks. Our results show consistent throughput improvements, thanks to the use of aggregated path capacity. We give comparisons to Multipath TCP, where we show our approach can achieve a similar performance while offering the advantage of endpoint transparency.

The increasing uptake of smart home appliances, such as lights, smoke-alarms, power switches, baby monitors, and weighing scales, raises privacy and security concerns at unprecedented scale, allowing legitimate and illegitimate entities to snoop and intrude into the family's activities. In this paper we first illustrate these threats using real devices currently available in the market. We then argue that as more such devices emerge, the attack vectors increase, and ensuring privacy/security of the house becomes more challenging. We therefore advocate that device-level protections be augmented with network-level security solutions, that can monitor network activity to detect suspicious behavior. We further propose that software defined networking technology be used to dynamically block/quarantine devices, based on their network activity and on the context within the house such as time-of-day or occupancy-level. We believe our network-centric approach can augment device-centric security for the emerging smart-home.

The increasing heterogeneity and asymmetry in wireless network environments makes QoS guarantees in terms of delays and throughput a challenging task. In this paper, we study a novel scheduling algorithm for multipath transport called Delay Aware Packet Scheduling (DAPS) which aims to reduce the receiver's buffer blocking time considered as a main parameter to enhance the QoS in wireless environments. We develop an analytical model of maximum receiver's buffer blocking time and extend the DAPS algorithm considering implementation issues. Performance evaluations based on ns-2 simulations highlight the enhanced QoS that DAPS can provide. With reference to the classical multipath transport protocol CMT-SCTP, we observe a significant reductions of the receiver's buffer occupancy, down by 77%, and the application delay, down by 63%.

Physiological sensors are widely used in user studies, often by practitioners with limited expertise in networking. However, large data volumes, and processing times often prevent the use of a single computer to collect the readings in real time. With multiple collection machines appear the problems of data aggregation and, more importantly, synchronisation. This paper describes how the OML reporting library allows solving the aggregation problem at low cost by introducing a lightweight instrumentation reporting to a centralised database. However, with unknown delays in network paths during aggregation and unreliable clocks on acquisition machines, synchronisation is hard to attain. We present a preliminary study of the theoretical feasibility of post hoc synchronisation corrections, supported by an experiment applying correction techniques to artificially impaired clocks and network transmissions. Based on the results of this experiment this paper highlights potential improvements.

In this paper, we introduce the Moana network infrastructure. It draws on well-adopted practices from the database and software engineering communities to provide a robust and expressive information-sharing service using hypergraph-based network indirection. Our proposal is twofold. First, we argue for the need for additional layers of indirection used in modern information systems to bring the network layer abstraction closer to the developer's world, allowing for expressiveness and flexibility in the creation of future services. Second, we present a modular and extensible design of the network fabric to support incremental architectural evolution and innovation, as well as its initial evaluation.

International audience

Modern mobile devices comprise multiple interfaces for heterogeneous network technologies. However, currently implemented mechanisms to decide which one(s) to use and distribute application flows accordingly (i.e., solving the multihomed flow management problem, MFM) are rather coarse and do not leverage the opportunities. A user-centric quality-aware (QA) MFM has been proposed which optimises network use based on user-perceivable metrics such as application quality as well as energy and monetary costs. This paper refines this approach by providing a single method for both real-time and elastic (i.e., TCP-based) traffic, and uses realistic available capacity estimation. We evaluate this method in OPNET-simulated LTE and WLAN mobile networks. We study the impact of methods used to trigger the decision algorithm, and investigate the influence of an increasing number of users employing the QA-MFM technique on both the user-perceivable metrics and the global network performance. We find that on-demand triggering performs better than a static periodic method. We also demonstrate that the proposed approach out-performs classical network selection techniques in terms of application quality. We also show that the QA-MFM is not too greedy as to not scale with a number of users, and has a positive effect on the network loads, by preemptively adapting applications parameters to match network conditions.

We consider the problem of network selection and flow distribution for a multihomed mobile device. We argue the benefits of a holistic approach which considers user- and application-centric metrics such as quality, energy consumption and monetary cost, rather than the commonly used network-centric metrics. We thus introduce the multihomed flow management problem which combines network selection, flow distribution and application flow awareness. We formulate it as a constrained optimisation problem and compare it to commonly used techniques: single network selection and load balancing. For selected interactive applications, we use empirical network measurements to evaluate the optimal solutions obtained by the three approaches. We show that, by exploiting the flexibility of application parameters, it is possible to achieve the potentially conflicting goals of maintaining high application quality while reducing both the power consumption and cost of network use.

In the context of mobile and pervasive networking, it is not uncommon to experience frequent loss of connectivity. Congestion control algorithms usually mistake resulting packets losses for congestion events and unnecessarily reduce their network usage. We propose an enhancement to the TCP-Friendly Rate Control (TFRC) protocol to overcome this issue. Our contribution proposes to suspend the data transfer, in a way similar to Freeze-TCP, when an upcoming disconnection is predicted. Furthermore, a probing mechanism is introduced to enable fast adaptation to new network conditions. We present a description of the additional freezing and resuming mechanisms. This proposal has been implemented in ns-2. Simulation results are thus compared to the regular behavior and show encouraging improvements.