Welcome to the
System Level Design Group

Everything we do is about understanding and harnessing computation and communication in technological and natural systems. We believe in thinking out-of-the-box and discovering the path not taken when discovering "new things".

We challenge the status quo by bringing science and engineering together. Specifically, we develop methods and tools to explore various angles in system design. Our current topics of interest include, but are not limited to, embedded, biological, and social applications. Our topics target computational platforms of the future and go beyond traditional silicon platforms.

We illustrate our vision by exploring embedded, biological, and social applications. Our research results are highly visible, widely cited, and consistently funded. Would you like to join us?

NEW Recent Publications

  • Biological Systems
  • Paper on inferring microbial association network in PLOS COMP. BIO (Dec. 2017). here
  • Paper on bacteria-based drug delivery system in Elsevier Nano Communication Networks (Feb. 2017). here
  • Paper on impacts of quorum sensing inhibitions (QSI) on strain competition and development of QSI resistance in Nature Scientific Reports (Oct. 2016). here
  • Computational Sustainability
  • Paper on discovering hidden knowledge in carbon emissions data in international conference on discovery science (2017). here
  • Paper on Non-stationary Bayesian Learning for Global Sustainability in IEEE Transactions on Sustainable Computing (Jun., 2017). here
  • Silicon Systems
  • On-Chip Communication Network for Efficient Training of Deep Convolutional Networks on Heterogeneous Manycore Systems (Nov. 2017). here
  • Paper on imitation learning for dynamic VFI control in IEEE Transactions on Very Large Scale Integration (VLSI) Systems (Sept. 2017). here
  • Paper on Wireless NoC and Dynamic VFI Co-Design published in IEEE Trans. VLSI (2016). here ECE News Article
  • Social Networks
  • Paper on Data-Driven Enginnering of Social Dynamics published in PLOS ONE (2016). here
  • Paper on Opinion Formation and Diffusion in Social Networks published in PeerJ Computer Science (2016). here

  • Recent and Upcoming Talks

  • NEW Keynote talk @ SocialSens 2018 (April 17th, Orlando) here
  • NEW Talk @ Intelligent Systems for Molecular Biology (ISMB) (July 12, Orlando) Special Session on Molecular Communication and Networking with Applications to Precision Medicine
  • Distinguished Lecture @ ESE UPenn 2015 (April 28) Entering the Labyrinth of (Inexact) Computer Science: A Cyber-Physical Approach
    Read the Abstract

    Abstract: During high school, I was fascinated by science. I used to read stories about Nobel Prize laureates and dream about their discoveries. Despite this, I became an engineer and, over the years, got to appreciate the power and transformative nature of computing in all human endeavors. In recent years, due to the cyber-physical systems (CPS) advent, I have found renewed interest and (somehow unexpected) opportunities to revisit some long forgotten topics in math and physics realizing that they can actually offer a much deeper understanding of CPS modeling and optimization. Truth being told, designing cyber-physical systems still feels more like an art rather than science but, in order to harness their huge potential, we need to reach beyond the established confines of computer systems design and (re)define a new science of CPS design. Starting from these overarching ideas, I discuss the theoretical foundations and practical implications of using a network approach to developing new mathematical models and tools needed to guide the CPS optimization ranging from hardware, all the way up to software, and (user-aware) application development. In other words, this talk is precisely about the subtle interplay between science and engineering and the joy of seeing things come full circle.

    Recent Press

  • NEW Our paper on Wireless NoC and Dynamic VFI Co-Design wins T-VLSI Best Paper Award! ECE News Article Link to Paper

  • NEW Our paper is featured by the IEEE Transactions on Computers, May 2018 issue: On-Chip Communication Network for Efficient Training of Deep Convolutional Networks on Heterogeneous Manycore Systems

  • NEW 2018 Marculescu receives Kavčić-Moura Professorship: here

  • 2016 CMU Today: Battling Bacterial Infections

  • 2015 Scientific American: A broader understanding of bacterial social networks might help scientists combat antibiotic resistance

  • Revolutionising Multicore Architecture Design

    Recent Awards

  •  The 2013 CIT Outstanding Research Award from the College of Engineering for seminal contributions to the modeling, design and optimization of network-on-chip (NoC) communication architectures for integrated systems 
  •  2013 Most Influential Paper Award: 10-Year Retrospective from the Asia and South Pacific Design Automation Conference - J. Hu, R. Marculescu 
  •  2013 Asia and South Pacific Design Automation Conference Best Paper Award - Hiroki Matsutani, Paul Bogdan, Radu Marculescu, Yasuhiro Take, Daisuke Sasaki, Hao Zhang, Michihiro Koibuchi, Tadahiro Kuroda, Hideharu Amano photo