Massive Information Scavenging with Intelligent Transportation Systems
The goal of the MISC project is to devise efficient, secure and dependable system architectures for a massively distributed urban scanner composed of private cars, taxis, city buses and trucks, which, while pursuing their usual routes at various speeds in a scattered fashion, are capable of gathering, storing, processing, and disseminating massive amounts of data. Depending on the sensing equipment installed in the vehicle, the captured data sets can be extremely rich and varied, including information about road conditions, traffic characteristics, commercial services, environmental parameters, and sampled measurements of critical infrastructures (such as the water distribution network and the local power grid). As an important addition, the envisioned massive data gathering system shall take the human factor into consideration, exploiting novel wearable sensing technology to collect vital biomedical data about drivers and passengers on urban transportation.
Vehicles may act upon the received data by disseminating vital information about the environment or critical infrastructures, or storing the collected data about commercial services and points of interest in a distributed manner, making it available to everyone in a peer-to-peer fashion. Decision makers can exploit the wealth of collected data for urban planning, risk assessment or adaptive transportation scheduling. As an example, real-time data about driver fatigue could be explored by public transportation authorities to improve job assignment and driver rotation schedules, thus increasing the safety of the transportation system. To avoid traffic disruption, leakage of private data, misinformation or chaotic behavior due to active attacks by means of bogus data, the integrity of the stored data and of the resulting information flows must meet the highest security standards.
The MISC project has the following key objectives:
- Scientific goals: devise adequate models for network information flow in vehicular networks and urban environments; quantify the benefits of network coding for massive data gathering; characterise human stress in traffic situations; provide solid guidelines for the design secure and dependable system architectures;
- Technical goals: Develop secure network coding protocols that leverage geographic positioning information to bring the right information to the right place at the right time; combine real-time biomedical signals with vehicle and urban context information;
- Social-economic goals: increase the safety and efficiency of public transportation systems and road networks by means of massive data gathering technology; transfer technology to local and global companies.
- Education goals: attract the brightest students for a career in R&D, use the developed data gathering tools for lab training, adopt some of MIT's best practices.
The project achieved the following results:
- Realistic Vehicular Network Models;
- Geographically Aware Network Coding Protocols;
- Evaluation of the Human Factor: VitalJacket for measuring stress levels of individual drivers and passengers;
- Context-Aware Security Countermeasures against Malicious Attacks;
- Large-Scale Deployment and Testing: joint utilisation of sensing, locating, bio-monitoring and disseminating techniques for massive data acquisition in public buses.
See a visualization of MISC data overlaid on Google Earth depicting a voyage from Porto to Aveiro at https://youtu.be/fZmm9RR8Iz8
 L. Lima, F. Zhao, J. Barros, M. Medard, R. Koetter, T. Kalker, and K.Han. On Counteracting Byzantine Attacks in Network Coded Peer-to-Peer Networks, To Appear in the IEEE Journal of Selected Areas in Communications, 2010.
 G. Maierbacher, J. Barros and M. Medard, Practical Source-Network Decoding, IEEE International Symposium on Wireless Communication Systems (ISWCS'09), Siena, Italy, September 2009
 MinJi Kim, M. Medard, J. Barros, R. Koetter, An Algebraic Watchdog for Wireless Network Coding, IEEE International Symposium on Information Theory, Seoul, Korea, June/July, 2009.
 L. Lima, J. Barros, R. Koetter, Byzantine Attacks against Network Coding in Peer to Peer Distributed Storage, IEEE International Symposium on Information Theory, Seoul, Korea, June/July, 2009.
 L. Lima, D. Ferreira and J. Barros, Topology Matters in Network Coding, To Appear in Springer Telecommunication Systems Journal, Special Issue on Future Internet Architectures, 2010.
 João Rodrigues, Fausto Vieira, Tiago T. V. Vinhoza and João Barros, A Non-Intrusive Multi-Sensor System for Characterizing Driver Behavior, submitted to IEEE Symposium on Intelligent Transportation Systems, Madeira, Portugal 2010.
Institutions/Research Centers involved: FEUP, FCUP, UA, IT
Companies involved: Biodevices, BAE, MacLaren Electronics, Petratex, STCP