Novel security paradigms for heterogeneous networks

Problem statement

The past decades have seen the development of various wireless technologies and mobile network architectures including Wi-Fi, 3G and WiMax.

With the arrival of 4G technologies and the heterogeneous networks, new questions have been raised about security such as the inter-operability of the various security architectures already available and the processing of vertical handovers.

As of today, classical recipes have applied to these different architectures, leading to non-optimal patchwork of protocols exhibiting a growing complexity. Therefore, a serious discussion has to be conducted about the overall system security and its cryptographic properties.

In this research, several novel techniques are presented in order to specifically protect mobile networks and mobile terminals.

Problem statement

Software Defined Networking (SDN) is a new approach to designing, building and managing networks. The basic concept is that SDN separates the network’s control (routing decisions) and forwarding (switches) planes to make it easier to optimize each. SDN allows a virtualization of the networking functions.

In the context of 5G communications, our research aims at delivering SDN controllers suitable to provide end-to-end communications across multiple distinct wireless technologies, such as 3G, 4G, Wi-Fi, etc. For instance, Logically centralized control allows for efficient base-station coordination and traffic grooming, which is particularly useful for off-loading while preserving QoS requirements.

Finally, telecommunications network must exhibit high-availability and high-reliability while being extremely secure, all at the same time. Hence, we aim at developing a carrier-grade SDN architecture.

Problem statement

Experimental validation is often under-evaluated in Wireless Sensors Network (WSN) research. Moreover, several recent papers have shown that there exists a significant cross-layering between the routing and the MAC protocols, especially in presence of very low duty cycles, i.e., deep sleeping sensors.

In this research, we study experimentally the interactions between the Radio Duty Cycling (RDC), the MAC protocols, and the new Low power and Lossy Networks (RPL) protocol for IPv6 sensor networks. The objective is to provide new routing metrics and objective function (OF) that can accommodate very low duty cycles and validate the corresponding performance parameters in-field.