Our research activities are organized into the five main following themes:
Security in infrastructure-less and constrained networks
The Internet was not designed to operate in a completely open and hostile environment. It was designed by researchers that trust each other and security was not an issue. The situation is quite different today and the Internet community has drastically expanded. The Internet is now composed of more than 300 millions computers worldwide and the trust relationship has disappeared. One of the reasons of the Internet success is that it provides ubiquitous inter-connectivity. This is also one of its main weaknesses since it allows to launch attacks and to exploit vulnerabilities in a large-scale basis. The Internet is vulnerable to many different attacks, for example, distributed Denial-of Service (DDoS) attacks, epidemic attacks (Virus/Worm), spam/phishing and intrusions attacks. The Internet is not only insecure but it also infringes users’ privacy. Those breaches are due to the Internet protocols but also to new applications that are being deployed (VoIP, RFID,...).
A lot of research is required to improve the Internet security and privacy. For example, more research work is required to understand, model, quantify and hopefully eliminate (or at least mitigate) existing attacks. Furthermore, more and more small devices (RFIDs or sensors) are being connected to the Internet. Current security/cryptographic solutions are too expensive and current trust models are not appropriate. New protocols and solutions are required: security and privacy must be considered in the Internet architecture as an essential component. The whole Internet architecture must be reconsidered with security and privacy in mind.
Our current activities in this domain on security in wireless, ad hoc and sensor networks, mainly the design of new key exchange protocols and of secured routing protocols. We work also on location privacy techniques and authentication cryptographic protocols and opportunistic encryption.
Rapid advances in microelectronics are making it possible to mass-produce tiny inexpensive devices, such as processors, RFIDs, sensors, and actuators. These devices are already, or soon will be, deployed in many different settings for a variety of purposes, which typically involve tracking (e.g., of hospital patients, military/rescue personnel, wildlife/livestock and inventory in stores/warehouses) or monitoring (e.g., of seismic activity, border/perimeter control, atmospheric or oceanic conditions). In fact, it is widely believed that, in the future, sensors will permeate the environment and will be truly ubiquitous in clothing, cars, tickets, food packaging and other goods. Simultaneously, ad hoc networks are gaining more and more interest in the research community. An ad hoc network is a ”spontaneous” network of wireless devices/users that does not rely on any fixed infrastructure. In such a network, each node is also a router, i.e., it routes/forwards packets for other nodes.
Ad hoc networks can be categorized into two main groups: Mobile Ad Hoc networks (MANET) and Wireless Sensor Networks (WSN). MANETs are used to provide a communication infrastructure to end-users when a fixed infrastructure is unavailable. MANETs are typically used in emergency/rescue situations, i.e., following an earthquake, when infrastructure is destroyed. They can be also used to provide relatively cheap and flexible wireless access to network backbones. In contrast to MANETs, WSNs are not meant to provide a communication infrastructure to end-users, but rather to reach a collective conclusion regarding the environment. A WSN is typically composed of a base station (sink) and many small sensors. Communication is often one-way, i.e. only from sensors to the base stations. Even though MANETs and WSNs are closely related, they have quite different characteristics. WSNs are usually much larger than MANETs, by at least an order of magnitude. Also, WSNs act under severe technological constraints: they have severely limited computation and communication abilities. Furthermore, their power (battery) resources are limited, i.e. if a node runs out of battery power, it essentially becomes permanently non-operational. These new highly networked environments create many new exciting security and privacy challenges. Our goals are to understand and tackle some of them.
are also interested in the particular case of RFID tag security. An RFID
(Radio-Frequency IDentification) tag is a small circuit attached to a small
antenna, capable of transmitting data to a distance of several meters to a
reader device (reader) in response to a query. Most RFID tags are passive,
meaning that they are battery-less, and obtain their power from the query
signal. They are already attached to almost anything: clothing, foods, access
cards and so on. Unfortunately, the ubiquity of RFID tags poses many security
threats: denial of service, tag impersonation, malicious traceability, and
information leakage. We focus in this work on this latter point that arises
when tags send sensitive information, which could be eavesdropped by an
adversary. In the framework of a library, for example, the information openly
communicated by the tagged book could be its title or author, which may not
please some readers. More worryingly, marked pharmaceutical products, as
advocated by the US Food and Drug Administration, could reveal a person’s
pathology. For example, an employer or an insurer could find out which
medicines a person is taking and thus work out his state of health. Large
scale applications like the next generation of passports are also subject to
such an issue. Avoiding eavesdropping can be done by establishing a secure
channel between the tag and the reader. This requires the establishment of a
session secret key, which is not always an easy task considering the very
limited devices’ capacities. This difficulty is reinforced by the fact that
tags and reader do not share a master key in most of the applications. In the
future, implementing a key establishment protocol may become a mandatory
feature. For example Californian Bill 682 requires such a technical measure
to be implemented in ID-cards deployed in
RFID deployment creates many new exciting security and privacy challenges. Our goals are again to understand and tackle some of them.
New dissemination paradigms
The future Internet will be even more heterogeneous and should provide a scalable support for seamless information dissemination, whatever the underlying support. A lot of work has already been done on the efficient support of group communications on the Internet, both at routing, transport and application levels. These works gave birth to content broadcasting services (e.g. in DVB-H networks) as well as some content dissemination peer-to-peer systems (e.g. BitTorrent). Mastering scalable communications requires to deal with a wide range of networking components and techniques, like reliable multicast, FEC codes, multicast routing and alternative group communication techniques, audio and video coding, announcement and control protocols.
Our goal in this domain is to design and implement such components to ensure efficient and scalable group communications. To realize this goal, we investigate several key services and building blocks: first, the efficient application-level Forward Error Correction (AL-FEC) codes that are needed to improve the transmission reliability and application efficiency; secondly the security services (e.g. content integrity, source authentication, confidentiality) whose importance will become more and more acute especially in heterogeneous networking/broadcasting environments; and finally scalable session-level control tools that will be required to control at a high abstraction level the operational aspects of the underlying dissemination systems.
One the other hand, peer-to-peer technology is widely widespread and highly studied. However, the dynamics of a peer-to-peer network is still not fully understood. Indeed, we observe significant differences in service capacities among the different peer-to-peer protocols. These differences are due to small protocols specificities. It is of major importance to understand why and how these specificities impact the dynamics of a peer-to-peer network.
Our goal, with this activity, is to gain a deep understanding of these dynamics in order to propose improvements for the next generation of peer-to-peer protocols.
The tremendous success of the wireless access technologies and their great diversity has further increased the heterogeneity of the Internet. The miniaturization of electronic components gave birth to a large number of new applications such as RFID, wireless sensors/nanosensors for medical applications, all kinds of wireless sensors that for example are able to avoid or forecast natural disasters, etc. Each of these new applications has particular needs and requires specific optimizations (e.g., battery life, power control to limit interferences, optimal multihop routing). These new miniaturized circuits and applications have launched the beginning of the new era of ambient networks, where the heterogeneity is more and more present. All these new applications have very different characteristics, with multiple standards, all with the same target to communicate.
It is therefore important to address management and control of wireless networks including support for auto-configuration and self-organization under policy and security constraints; creation of survivable systems in the face of the challenges of the wireless environment; issues in wireless networks from a systems perspective such as the interactions of protocol layers and different access networks including cross-layer optimizations and feedback/control mechanisms; and realistic and affordable means for carrying out representative, repeatable, and verifiable experiments to validate research on wireless networks including open tools and simulation models, as well as experimental facilities to access realistic environments and map experimental results to simulation models.
We work also on how to efficiently support audio and video applications in heterogeneous wired and wireless environments. Here we focus on congestion control for multicast layered video transmission, scalable protocols for large scale virtual environments and on performance improvements and quality of service support for wireless LANs. We also consider the impact of new transmission media on the TCP protocol performance. Our goal is to provide each end-user the best quality possible taking into account its varying capacities and characteristics of multimedia flows, and to propose adaptation to the TCP protocol to make it fully profit from the available resources in a heterogeneous environment.
Understanding the Internet behavior
One topic in this area is to develop mathematically rigorous models to study and analyze the dynamics and properties of large-scale networks. One of the goals is to understand the fundamental performance limits of networks and to design algorithms that allow us to approach these limits. Another topic is to address the fundamental methodological barriers that make it hard to reproduce experiments or to validate simulations in real world systems. The goal here is to understand network behaviors for varying time-scales, a range of spatial topologies, and a range of protocol interactions. One of the major challenges with the future Internet will be how to monitor it in a scalable and distributed way. This requires designing intelligent sampling methods that provide good network coverage while reducing overhead. Another challenge will be in the characterizing of traffic sources by network operators and the detection of anomalies. The challenge for network operators in the future will be in providing an attack free Internet connectivity to their end-users and in prohibiting malicious users from using their premises. A third challenge is to understand the issues related to transport and peer-to-peer protocols dynamics on a very large scale with the current Internet, and to propose efficient solutions for the future Internet. We describe briefly in the following our main activities in this domain.
An important objective in this domain is a better monitoring of the Internet and a better control of its resources. On one side, we focus on new measurement techniques that scale with the fast increase in Internet traffic. Among others, we use the results of measurements to infer the topology of the Internet and to localize its distributed resources. The inference of Internet topology and the localization of its resources is a building block that serves for the optimization of distributed applications and group communications. We cite in particular replicated web servers, peer-to-peer protocols and overlay routing technologies.
On the other side, we focus on solutions that optimize the utilization of network resources. Our solutions are usually based on mathematical modeling of the underlying problem and an optimization using analytical and numerical tools. This optimization is meant to provide insights on how to tune protocols and dimension networks. As examples of activities in this direction one can find the optimization of routing and its mapping to underlying layers, the dimensioning of wireless mesh networks, the clustering of network entities for the purpose for traffic collecting and monitoring, etc.
Experimental environment for future Internet architecture
It is important to have an experimental environment that increases the quality and quantity of experimental research outcomes in networking, and to accelerate the transition of these outcomes into products and services. These experimental platforms should be designed to support both research and deployment, effectively filling the gap between small-scale experiments “in the lab”, and mature technology that is ready for commercial deployment. In terms of experimental platforms, the well-known PlanetLab testbed is gaining ground as a secure, highly manageable, cost-effective world-wide platform, especially well fitted for experiments around New Generation Internet paradigms like overlay networks. The current trends in this field, as illustrated by the germinal successor known as GENI, are to address the following new challenges. Firstly, a more modular design will allow achieving federation, i.e. a model where reasonably independent Management Authorities can handle their respective sub-part of the platform, while preserving the integrity of the whole.
Secondly, there is a consensus on the necessity to support various access and physical technologies, such as the whole range of wireless or optical links. It is also important to develop realistic simulators taking into account the tremendous growth in wireless networking, so to include the many variants of IEEE 802.11 networking, emerging IEEE standards such as WiMax (802.16), and cellular data services (GPRS, CDMA). While simulation is not the only tool used for data networking research, it is extremely useful because it often allows research questions and prototypes to be explored at many orders-of-magnitude less cost and time than that required to experiment with real implementations and networks.