Persons

doc. Ing. Stanislav Vítek, Ph.D.

Dissertation topics

Enabling machine learning in embedded systems and IoT devices

  • Branch of study: Electrical Engineering and Communications
  • Department: Department of Radioelectronics
    • Description:
      The cloud approach to data processing in embedded/Internet of Things systems faces latency and network bandwidth constraints. In addition, it often requires persistent radio signal coverage. Moving some of the recognition and classification capabilities to embedded devices can solve some of these problems. New architectures and approaches can massively reduce network bandwidth usage and response latency and create truly autonomous devices capable of operating even in cases of network unavailability. It can also significantly improve performance and power consumption through more distributed hierarchical structures.

Methods of EMCCD image data processing

  • Branch of study: Electrical Engineering and Communications
  • Department: Department of Radioelectronics
    • Description:
      Advances in astronomy, like observation of the fast optical transients, require new technology. An example of new technology is EMCCD (Electron Multiplying CCD) camera, able to detect extremely low amounts of photons in very short exposure times. Main goal of the work is modeling of EMCCD detectors and design/optimization of the fast image processing algorithms.

Responsible person Ing. Mgr. Radovan Suk