Research of prof. Massimo Franceschetti
The endless enigma - Salvador Dali.
Complex information systems
Today, technological advances have made it possible to develop massively large communication systems composed of small and relatively simple devices that can be randomly deployed and `ad-hoc' organize into a complex communication network. These networks can be used for human communication, as well as for sensing the environment and collecting and exchanging data for a variety of applications, such as environmental and habitat monitoring, industrial process control, security and surveillance, and structural health monitoring.
The objective of our research is to develop the foundations for the theory of these complex communication systems that will increasingly constitute the infrastructure of modern information society.
Network seen as a random aggregation of particles.
One of the most challenging problems in the development of this theory is to manage complexity. The key is to develop the right abstractions to reason about complex systems. Using the same approach that has been successful in describing natural systems of interacting particles in physics, we manage complexity by describing local interactions probabilistically, and then derive global system properties by averaging out the microscopic effects. Hence, by using probabilistic tools of statistical physics, such as random geometric graphs and percolation-theoretic models, we reveal global structural properties of the system that can can then be used to study how information can be processed, stored, and transferred in the system.
Within this framework, we focus specifically on: radom strucure discovery and routing of information, interference limited communication models, network information theory and network coding, feedback loop analysis over (time varying) communication channels. The research is highly interdisciplinary and is a blend of information theory, control and dynamical systems, applied probability, and stochastic analysis.
Interference process in a random network due to radio waves.
Physics of wireless communication
Many modern communication systems rely on electromagnetic propagation to convey information. As wave propagation is a complex process that occurs through line of sight, multiple reflection, diffraction, and scattering, precise characterization of the wave's information content is a challenging task. The amount of information that can be transported by waves depends on the power, and on the spectrum of the radiated field. The spectrum has two components that are mutually coupled: space and frequency. When communication is performed using a multiple antenna system, both of these components pose fundamental limits on the amount of information that can be resolved at the receiver.
Our approach on the one hand is to develop stochastic models of propagation that can capture the essential features of the field that can be measured at the receiver, such as the path loss, delay spread, and the coherence bandwidth. On the other hand, we investigate the fundamental limits on the information content that follow directly from the laws of physics.
Physical view of wave propagation and communication system model.
One of the objectives is to reveal information conservation principles in space-time. This research relies on tools from electromagnetics, information theory, communications, as well as functional analysis.
Control over communication channels
In a distributed control system the controller elements are not in a central in location but are distributed throughout the system with each component sub-system under the control of one or more controllers. The entire system may be networked for communication and monitoring. Such a system typically uses sensors for estimation, computers as controllers, and interconnections and protocols for communication. Stabilization and control become challenging tasks due to the time varying nature of the channel conditions. The achievable information rate over the channel must be high enough compared to the sytem's unstable modes to guarantee stability, but this rate is also likely to change unpredictably over time. Our aim is to develop a theory for control over communication channels that explicitly accounts for the randomness in the channel.
This research is a blend of information theory and control and dynamical systems.