Digital systems and computer networks

group leader

assist. prof. Ante Kristić, PhD

associates

assoc. prof. Duje Čoko, PhD
assist. prof. Vesna Pekić, PhD
Marina Prvan, PhD

Research topics

  1. Flow control in packet networks
  2. Quality of Service (QoS)
  3. MAC protocols in wireless networks
  4. Protocol modeling
  5. Logic design for programmable logic architectures
  6. Logic design for application specific circuits
  7. Machine learning based data analysis

Description of laboratory and equipment

Computer Networks Laboratory:

  • Cisco router 1841-SEC/K9
  • Cisco switch WS-C2960-24TT-L
  • HP switch ProCurve 2626 J4900B
  • D-link firewall DFL-200
  • Planet switch WSD-800
  • Planet AP/router WRT-414
  • Planet WLAN WL-u356a

Digital Systems Laboratory:

  • Xilinx Virtex UltraScale+ FPGA VCU118 Evaluation Kit
  • Xilinx Spartan-3e FPGA
  • Xilinx CoolRunner-ii CPLD
  • Analog Devices Blackfin BF533 DSP
  • Analog Devices Blackfin BF537 DSP
  • Tektronix digital oscilloscope
  • HP 8012 pulse generator
  • ALL-100 programmer
  • ALL-11 P2 programmer
  • ALL-03 programmer
  • DELAB1 laboratory model
  • 25 academic licenses for XILINX Vivado IDE
  • ASIC development system Cadence (Europractice IC package)

Contacts with academic and other institutions

  • University of Zagreb, Faculty of Electrical Engineering and Computing (FER), Croatia
  • Polytechnic Institute of Paris, École politechnique, Palaiseau, France
  • Fermi National Accelerator Laboratory (Fermilab), Batavia, Illinois, USA
  • European Organization for Nuclear Research (CERN), Geneva, Switzerland

 

project title

Wireless Media Access Mechanisms Modeling (MAMM)

Description of research in a 5-year term

In the five-year period, work is planned to continue in the proposed areas, with an emphasis on building the system and further training of the Specific Integrated Circuits (ASIC) design team in line with the FESB development strategy. Special attention is paid to the development of ASIC systems for operation under high radiation. Work is underway to develop FPGA algorithms at the L1 and L2 trigger levels in collaboration with LLR Paris, with an emphasis on deployment of machine learning algorithms for detection of regions of interests (ROI) in HGCAL detector. Because of specific hardware constraints, the research will include methods of neural network quantization and input data quantization, combined with development of innovative architectures that take into consideration spatial and temporal signal characteristics.

In the wireless computer networks area, the research of medium access protocols will be continued, with the new MAC protocol as a result. The protocol should enable high throughput in congested network, while retaining good throughput characteristics in unconstrained networks. Additionally, the protocol should enable traffic differentiation and ensure prompt delivery of time-critical data. Such a protocol would be well suited for use in various networks that employ distributed medium contention, from standard 802.11 WLANs to modern sensor networks and vehicular networks.

Work on the project over a five-year period should result in the publication of about five articles in Category A journals, and about five articles in international conferences and Category B journals. Several visits of team members related to current research are planned (LLR Paris, France and CERN, Switzerland).