Site menu:

Real-Time Systems and Cyber Physical Systems

Real-time systems are computer systems where the correctness depends not only on computing the right result but also on computing the result at the right time. This creates new dependencies. For example, consider two processes that do not share data and do not exchange messages and where there are no real-time requirements. Then, it is possible to prove the functional correctness of one process without considering the other. Suppose now that they have timing requirements and the two processes are scheduled on a single processor. From a timing verification perspective, the two processes are now dependent. For this reason, it is necessary to develop algorithms for scheduling processes so that the timing requirements are considered when making scheduling decisions. This is called real-time scheduling. For a given scheduler and for a given software system, it is also necessary to prove that for all possible interleavings that can be generated at run-time, all timing requirements are met. This is called schedulability analysis.

Cyber-Physical Systems are computer systems where the computer system and its physical world are linked, either intentionally (the system was designed with this linking to attain better performance) or unintentionally (the designer thought that there was no linking but there actually was and this causes new types of faults).

Research Interest

About real-time systems: I have worked on real-time scheduling for multiprocessors, especially multicore processors. I am interested in schedulability analyses that are more expressive than the ones that are available in the research literature right now. About cyber-physical systems: I have developed WiDom --- a medium access protocol for wireless channels --- which makes it possible to apply the fixed-priority scheduling theory to wireless channels.

Community report about certification

Upcoming Events where I serve as TPC chair

Upcoming Conferences and Events where I serve on TPC

News

R. Martins, M. McCall, D. de Niz, A. Vasudevan, B. Andersson, M. Klein, J. Lehoczky, and H. Kim, ``Formal Verification of a Mixed-Trust Synchronization Protocol,'' Real-Time Network and Systems, 2021.

D. de Niz, B. Andersson, H. Kim, M. Klein and J. Lehoczky, ``Resilient Mixed-Trust Scheduling,'' RTSS 2021.

H. Kim, D. de Niz, B. Andersson, M. Klein and J. Lehoczky, ``Addressing Multi-Core Timing Interference using Co-Runner Locking,'' RTSS 2021.

V. Petrucci, B. Andersson, E. Massa, G. Lima, ``Heterogeneous Quasi-Partitioned Scheduling,'' RTSS 2021.

B. Andersson, D. de Niz, and M. Klein, ``Satisfying Real-Time Requirements of Multicore Software on ARINC 653: The Issue of Undocumented Hardware,'' DASC 2022.

B. Andersson and D. de Niz, ``Which Future Trends Motivate Real-Time Research?,'' ECRTS-RTP 2022.

B. Andersson, D. de Niz, and M. Klein, ``A Tool for Satisfying Real-Time Requirements of Software Executing on ARINC 653 with Undocumented Multicore,'' DASC 2023.

Bjorn Andersson, Dionisio de Niz, William Vance, John Ross, Mark Wotell, and Tuan Bui, ``Methodology of Combining Empirical Stress Testing and Formal-Methods Based Schedulability Analysis for Real-Time Multicore Software,'' DASC 2023.

Gave presentation at the panel ``Promises and Challenges of AI-enabled Software Development Tools for Safety-Critical Applications'' at the 27th Ada-Europe International Conference on Reliable Software Technologies (AEiC 2023).

FAA has published (as DOT/FAA/TC-23/06) our report ``Assessing the Use of Machine Learning to Find the Worst-Case Execution Time of Avionics Software''.

Dionisio de Niz, Bjorn Andersson, Mark Klein, John Lehoczky, Amit Vasudevan, Hyoseung Kim, and Gabriel Moreno, ``Mixed-Trust Computing: Safe and Secure Real-Time Systems,'' ACM Transactions on Cyber-Physical Systems, to appear (probably in 2024).