Section: New Results


Routing in Wireless Sensor Networks

Participants : Emmanuel Nataf [contact] , Patrick-Olivier Kamgueu, Nesrine Khelifi.

We have formalized our previous work on the routing protocol for wireless sensor network by fuzzy logic specifications. The rules of routing metric composition are now valid for any network depth and we demonstrated its quality by real experimentation [36] . This work is done in the context of the associated team we build with the Cameroun and the Inria international lab LIRIMA.

For potentially very large wireless sensor network, our routing or any other routing, can not limit traffic bottleneck near the network root. Network depth should also be reduced as hop by hop communication is a factor which strongly increases data loss rate. Considering theses problems Nesrine Khelifi PhD student of the Manouba University in Tunisia spent 3 months within the Madynes team trying to limit the depth of the network by splitting it under the supervision of network quality observers we had to define.

Operator calculus based routing in Wireless Sensor Networks

Participants : Evangelia Tsiontsiou, René Schott, Stacey Staples [Southern Illinois University Edwardsville] , Jamilla Benslimane, Bilel Nefzi, Ye-Qiong Song [contact] .

Recently, Operator calculus (OC) has been developed by Schott and Staples with whom we collaborate. We make use of OC methods on graphs to solve path selection in the presence of multiple constraints. Based on OC, we developed a distributed algorithm for path selection in a graph. This approach has been applied to efficiently solve a joint routing, channel and time slot scheduling optimization problem in UWB wireless sensor networks [6] . We also designed a new routing protocol which makes use of this algorithm: the Operator Calculus based Routing Protocol (OCRP). In OCRP, a node selects the set of eligible next hops based on the given constraints and the distance to the destination. It then sends the packet to all eligible next hops. The protocol is implemented in Contiki OS (Rime profile) and emulated for TelosB motes using Cooja. We compared its performance against tree and directional flooding routing and showed the advantages of our technique [28] . Our ongoing work consists in its comparison with RPL to show its practical contribution to handle simultaneously several IETF ROLL routing metrics. This work is part of Lorraine AME Satelor project granted by Lorraine Region.

Probabilistic Energy-Aware Routing for Wireless Sensor Networks

Participants : Evangelia Tsiontsiou, Bernardetta Addis, Alberto Ceselli [Universita degli Studi di Milano] , Ye-Qiong Song [contact] .

Healthcare applications are considered as promising fields for Wireless Sensor Networks (WSNs). Thanks to WSNs, patients can be monitored in hospitals or smart home environments, providing health improvement, or emergency care. A key issue is the limited battery of sensors; indeed, current WSN research trends for healthcare applications include energy efficient routing and network lifetime guarantee mechanisms, among others. One of our ongoing work consists in designing a Smart Probabilistic Energy-Aware Routing Protocol (SPEAR) for WSNs which aims at maximizing the network lifetime by keeping low energy consumption and balancing network traffic between nodes. Our experimental campaign reveals that our SPEAR protocol outperforms the popular Energy Aware Routing Protocol (EAR) from the literature, proving to be more effective in extending the network lifetime. This work has resulted in a conference submission. It is part of Lorraine AME Satelor project granted by Lorraine Region.

Energy-aware IP networks management

Participants : Bernardetta Addis [contact] , Giuliana Carello [DEIB, Politecnico di Milano, Italy] , Antonio Capone [DEIB, Politecnico di Milano, Italy] , Luca Gianoli [Polytechnique de Montreal, Canada] , Sara Mattia [IASI, CNR, Roma, Italy] , Brunide Sansò [Polytechnique de Montreal, Canada] .

The focus of our research is to minimize the energy consumption of the network through a management strategy that selectively switches off devices according to the traffic level. We consider a set of traffic scenarios and jointly optimize their energy consumption assuming a per-flow routing. We propose a traffic engineering mathematical programming formulation based on integer linear programming that includes constraints on the changes of the device states and routing paths to limit the impact on quality of service and the signaling overhead.

A very important issue that may be affected by green networking techniques is resilience to node and link failures. We thus extended the optimization models to guarantee network survivability. Results show that significant savings, up to 30%, may be achieved even when both survivability and robustness are fully guaranteed.

Computational cost of proposed models can be very high when dealing with large size instances (network size and/or number of demands). For this reason, we proposed and tested different problem formulations with the aim of solving larger size instances at optimality. We focus on a particular form of shared protection mechanism, where energy consumption is associated only to active devices during normal functioning. We propose a standard and a projected formulation, with additions of cuts and valid inequalities. Computational results show that the projected formulation is very effective [20] . We plan to extend the work to consider multiperiod scenarios.

Virtual Network Functions Placement and Routing Optimization

Participants : Bernardetta Addis [contact] , Dallal Belabed [LIP6, Univ Paris 06, France] , Mathieu Bouet [Thales Communications & Security, France] , Stefano Secci [LIP6, Univ Paris 06, France] .

Network Functions Virtualization (NFV) is incrementally deployed by Internet Service Providers (ISPs) in their carrier networks, by means of Virtual Network Function (VNF) chains, to address customers' demands. The motivation is the increasing manageability, reliability and performance of NFV systems, the gains in energy and space granted by virtualization, at a cost that becomes competitive with respect to legacy physical network function nodes. From a network optimization perspective, the routing of VNF chains across a carrier network implies key novelties making the VNF chain routing problem unique with respect to the state of the art: the bitrate of each demand flow can change along a VNF chain, the VNF processing latency and computing load can be a function of the demands traffic, VNFs can be shared among demands, etc. We started our work providing an integer linear programming model for Virtual Network Functions Placement and demand rerouting. By extensive simulation on realistic ISP topologies, we draw conclusions on the trade-offs achievable between legacy Traffic Engineering (TE) ISP goals and novel combined TE-NFV goals [19] .

Composing IoT protocols with Named-Data Networking

Participants : Salvatore Signorello [University of Luxembourg] , Olivier Festor [contact] , Radu State [University of Luxembourg] .

With the emergence of IoT, many layer 2 protocols have been proposed with each of them its own characteristics, advantages and drawbacks. Choosing a protocol often depends on the global context, as for example number of users, time of the day... Although devices can now be fitted with multiple interfaces, using always the same specific layer 2 protocol is not efficient, in particular if we assume that connected devices are retrieving or exchanging similar contents. For example, assuming that WiFi is the most usable interface to download some files in Internet through an access point may not be ideal if a close-by device accessible by Bluetooth already has it. To accommodate so multiple layer 2 protocols, we propose to leverage the Named-Data Networking (NDN) paradigm which allows to explore in parallel multiple paths for retrieving content independently of the underlying protocol. Our first results [46] show that such a theoretical solution cannot work practically. Indeed, applying NDN in a blind mode over multiple layer 2 protocols does not assume the corresponding specificities like for example various collision rates depending on the underlying protocols, which have to be taken into account.