BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
X-LIC-LOCATION:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20240626T180034Z
LOCATION:Level 2 Lobby
DTSTART;TZID=America/Los_Angeles:20240626T180000
DTEND;TZID=America/Los_Angeles:20240626T190000
UID:dac_DAC 2024_sess237_RESEARCH333@linklings.com
SUMMARY:Fully Automated Implementation of Reservoir Computing Models on FP
 GAs for Nanosecond Inference Times
DESCRIPTION:Work-in-Progress Poster\n\nFabian Legl (Ingenieurbuero fuer Th
 ermoakustik GmbH) and Jonas Kantic (Technical University of Munich)\n\nWe 
 propose an efficient and generic Field-Programmable Gate Array (FPGA) impl
 ementation of Reservoir Computing using Linear Cellular Automata models fo
 r the application of time series processing. Our implementation results fr
 om a fully automated model definition to FPGA bitstream design automation 
 process. Hence, it significantly reduces the design time and complexity. I
 t can be clocked with at least 475 MHz on a Xilinx Zynq Ultrascale+ FPGA\n
 while performing one prediction in every clock cycle. Since our implementa
 tion only uses lookup tables and registers, it is platform independent. It
  thus can not only be run on high-performance but also on low-cost FPGAs w
 ithout special hardware components. Being up to six orders of magnitudes f
 aster than other Reservoir Computing model implementations, our implementa
 tion enables intelligent real-time sensor signal processing for applicatio
 ns requiring MHz sampling rates, like structure-borne noise monitoring or 
 high-frequency oscillation analysis.\n\nTopic: AI, Autonomous Systems, Clo
 ud, Design, EDA, Embedded Systems, IP, Security
END:VEVENT
END:VCALENDAR