Active and past projects

In this industrial partnership with RailEvo we study the traffic capacity of their innovative public transportation system, focusing both on the main line and on the stations, with the aim of showing the benefits compared to classic railway.

Cooperative driving systems can improve future transportation and human well-being from several different perspectives which include safety, traffic efficiency, and reduced emissions. This will enable new mobility paradigms such as Mobility as a Service which, in our vision, will lead to the advent of cooperative autonomous taxis (CATs). CATs will reduce the need for private ownership and thus the total number of vehicles in our streets.

Implementing such novel transportation paradigms requires communication between vehicles for autonomous coordination, and between passengers and the system to collect requirements such as pick-up positions and destinations. This poses challenges in terms of security and privacy. With respect to security, control information needs to be verified and trustworthy, whereas with respect to privacy it is necessary to protect users’ personal information.

PRECODE aims at tackling these issues by using novel communication technologies, like reconfigurable intelligent surfaces and directional antennas. Within the project, we plan to explore how such technology can be used to guarantee security and privacy. Moreover, PRECODE will develop algorithms that can optimize traffic flow and passenger assignment using partial knowledge, with the aim of improving users’ privacy.

The project focuses on next-generation mobility services, in particular by considering Mobility-as-a-Service enhanced by means of cooperative driving. Our final goal, which goes beyond SELF4COOP, is the reduction of traffic congestion and road-related accidents, which costs in the EU amount to 1% and 3% of the GDP, respectively. In our vision, cooperative autonomous vehicles (CAVs) will become an integrated system, realizing complex intelligent driving applications. This will minimize the environmental footprint of the vehicles, streamline the use of CAVs by multiple users and reducing the need for private vehicle ownership.

Realizing the above vision will require some fundamental building blocks which we will develop throughout SELF4COOP, which include the realization of advanced cooperative driving maneuvers, where intelligent systems gather data both from vehicles and from the infrastructure, and instruct CAVs to implement optimal actions. In turn, this will require to seamlessly blend different communication technologies, exploit Multi-access Edge Computing (MEC) resources flexibly and adaptively, and develop robust optimization and control techniques.

Plexe is an extension of the popular Veins vehicular network simulator which permits the realistic simulation of platooning (i.e., automated car-following) systems. It features realistic vehicle dynamics and several cruise control models, permitting the analysis of control systems, large-scale and mixed scenario, as well as networking protocols and cooperative maneuvers. It is free to download and easy to extend.

We developed this simulator as part of my PhD thesis and made it free to use for the community. On the website you find instructions for downloading, building, and running the simulator, as well as some tutorials on how to use it.

Robust and reliable wireless communication is one of the most important requirements in many application domains. This includes industrial networks, health care applications, as well as Vehicular Ad Hoc Networks where safety applications are seen as a main driver for their introduction. However, due to highly dynamic network structures and rapidly varying channels these networks are particularly challenging from a communications perspective.

In a first step, we investigated possibilities to detect and to mitigate wireless interferers. In order to gain deeper insights into the channel dynamics and the impact of different receive algorithms we create a framework for measuring and experimentation with IEEE 802.11[a/g/p] but also for IEEE 802.15.4 networks. The framework is SDR-based and can be used for simulations as well as over-the-air measurements. We release all source code under an Open Source license to make it accessible to other researchers and to allow reproduction of the results.

For more information about the software please see the software site.

Veins is an open source framework for running vehicular network simulations. It is based on two well-established simulators: OMNeT++, an event-based network simulator, and SUMO, a road traffic simulator. It extends these to offer a comprehensive suite of models for IVC simulation.

I personally contributed to the project by improving the communication model, in particular by increasing the realism and by adding new features to the PHY and MAC layers.

The aim of SECEDA is the small-scale, real-world development and testing of cooperative driving maneuvers and vehicular communication protocols through a cooperative driving fleet composed of model cars. SECEDA will work the development and the testing of new Multi-access Edge Computing (MEC)-based remote control. This idea exploits MEC cloud computing to bring the control of the vehicles to the edge, bringing great adaptation potentials to variable road traffic and communication conditions. The testbed will also pave the road to other research studies, spanning from collective data sensing and distributed embedded machine learning.

The aim of the project is to continue the work started in the past cooperation, in particular by working on the classification of IoT devices.

In 2019 I have been contacted by the Global Connectivity & Technology branch of the Electrolux group. In this industrial project we are going to design a system to perform compliance testing of IoT communication devices mounted inside Electrolux appliances. Even though the official project is concluded, the cooperation is still ongoing.

The Border Gateway Protocol (BGP) is the only Inter-AS (Autonomous System) protocol of the Internet, i.e., it is the glue that binds pieces of the Internet and keeps global communications in tune. BGP has a slow convergence: two configuration parameters (timers) have a high impact on BGP convergence speed, but there is no consensus on how to change their default value to a better one.

Centrality-based networking is a new paradigm that exploits graph centrality metrics to improve the scalability of routing protocols, with a technique we called Pop-Routing. Pop-Routing automatically tunes the frequency of the protocol messages to guarantee the optimal trade-off between overhead and convergence speed. In Infocom 2018 we presented the first fully distributed, exact algorithm to compute centrality using Bellman-Ford (BF): Pop-Routing can now be ported to Distance/Path-Vector protocols.

BGP is a Path-Vector protocol, uses the Bellman-Ford algorithm, and adds further information to compute the full path between two nodes. Our distributed centrality computation can effectively be used in BGP, as we have shown in a recent study. To prove that the convergence of Internet can be speed-up exploiting Pop-Routing, we need to test the distributed centrality computation on the BGP extension supporting Pop-Routing and measure the actual speed-up on a real implementation on sufficiently large topologies.

To achieve this goal Internet on FIRE will pursue three goals, corresponding to three Work Packages:

• Finalise, implement and test the distributed centrality computation of Pop-Routing on the BIRD daemon, the most popular open source implementation of BGP;
• Perform emulations of large-scale networks (hundreds of nodes, more than one node can be set on one device of the testbeds) to verify the convergence properties of the distributed centrality computation;
• Devise a strategy to optimize BGP parameters using Pop-Routing and testing it on the test-bed.

The goal of the project is to test and enhance “Pop-Routing”, a technique for wireless mesh link-state routing protocols that tunes the generation frequency of control messages independently for each node of a wireless mesh network as the result of real-time graph analysis performed on the network topology. Pop-Routing is backward-compatible and allows the reduction of the routing tables convergence time after a failure by a factor of up to 60%, or, conversely, it can keep the same convergence speed and reduce the total amount of control messages, thus reducing overhead and increasing the scalability of the protocol.

Pop-Routing theory was presented at IEEE Infocom 2016 and evaluated via emulation, the basic prototypes for a real implementation have been developed, and functional tests of prototypes are under review, OLSRv2 integration has been proven feasible. Further research is needed to fully understand Pop-Routing potentials.

The WiSHFUL infrastructure enables testing Pop-Routing in real conditions, and in particular, in scenarios in which we can fully control and modify the underlying physical and logical connectivity. This is a key task since Pop-Routing is dynamic: it changes the frequency with which control messages are generated on a per-node basis depending on the modification of the network topology. The great flexibility, total control and the monitoring functions that UPIs provide will empower a multi-layer analysis where we can compare the evolution of the physical network (which we control) with the evolution of the routing graph (which we monitor) and check how fast and consistent are the updates generated by Pop-Routing.

We will pursue two goals:

• Improve, tune and extend Pop-Routing algorithms in order to achieve scientific breakthrough that will produce top-level scientific publications;
• Stabilize Pop-Routing open source code in order to move one crucial step towards the implementation of Pop-Routing in real networks.

netCommons is a Horizon2020 research project, which follows a novel transdisciplinary methodology on treating network infrastructure as commons, for resiliency, sustainability, self-determination, and social integration. Project partners have expertise in engineering, computer science, economics, law, political science, urban, media, and social studies; and close links with successful Community Networks like guifi.net, ninux.org, and sarantaporo.gr.

EN-ACT aims to drastically reduce energy consumption and CO2 emissions of applications and systems based on Information and Communication Technology (ICT). The main goal of the project is to define specifications for the design of energy-aware software (that specifically concerns about the energy consumption, i.e., “green software”). The project will be characterized in terms of energy performance at application level, based on the implementation of platform-independent metrics.

Intelligent Transportation Systems (ITS) aim mainly at the reduction of traffic-related accidents, the improvement of the infrastructure exploitation, and the reduction of energy consumption and pollution. The key components of ITS are inter-vehicle and vehicle-to-infrastructure communication (IVC). Without proper communication means that empower the distribution of data so that informed and intelligent decisions can be taken, any application of ITC technologies and methodologies at ITS remains vain.

Within this broad framework this PhD proposal is focused mainly on understanding, modeling, exploiting and possibly improving the IEEE 802.11 standard amendment 802.11p or WAVE (Wireless Access in Vehicular Environment). Several lines of research can be identified within this focus, which are intertwined and require a cross-layer approach.

First of all, the behavior and characteristics of WAVE communications must be understood at a higher level than it has been analyzed so far: the presence of hundreds of communicating devices all moving at high speed and using the same frequency channels makes even the definition of the communication channel itself extremely difficult. Novel work is needed to find appropriate channel models to be used in conjunction with safety-related application, where approximate or wrong modeling of the communication part may lead to malfunctioning systems, which is unacceptable in these cases. The novel models must be validated, possibly against the first real implementations of IEEE 802.11p and not only against simulation tools.

Secondly, proper IVC protocols must be designed to properly support safety and non-safety applications. In particular, it is envisaged that vehicles will be equipped with multiple radios to exploit the multiple available channels, specifically to guarantee that the safety-related channel is continuously monitored. This scenario requires adequate coordination. If we can assume that infotainment applications will run on orthogonal channels from safety, other applications, such as cooperative driving, may require the use of dedicated channels, but are tightly intertwined with basic safety applications, so that proper joint management is required.

The information exchange among vehicles on our streets is becoming a key building block for a variety of applications ranging from safety critical warning systems to information systems and even entertainment applications. Besides the use of cellular networks, short range broadcast is an extremely efficient way for inter-vehicle communication. In this project we investigate several aspects of such IVC approaches:

• Beacon-based information exchange: periodic beacon messages build the basis for so-called cooperative awareness of driving vehicles. It has been shown that the periodicity of such beacons strongly depends on several factors including the vehicles’ density, possible obstacles blocking the radio communication, and the importance of messages to be transmitted. Based on our ATB (Adaptive Beacon Protocol) concept, we investigate adaptive and self-organizing options for improving the beacon efficiency with a strong focus on safety related applications.

• Distributed information management: besides the direct exchange of information between vehicles, the use of infrastructure elements strongly improves the efficiency of information management in a massively distributed system such as a vehicular network. We exploit both the use of road side units as well as the use of parked vehicles to optimize the communication efficiency. Dynamically grouping vehicles allows to share information between this group and passing vehicles. We see special advantages of using parked vehicles as a distributed information storage as well as to bridge communication gaps in safety critical applications.

• Simulation techniques and field tests: performance evaluation of IVC solutions typically depends on extensive simulation experiments. We continuously extend our Veins simulation toolkit that integrates network simulation and road traffic micro simulation. In particular, we are developing very realistic radio attenuation and shadowing models based on empirical observations from field tests and experimental measurements. These models are fully based on the standard DSRC/WAVE protocol suite.