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:20240626T180033Z
LOCATION:3001\, 3rd Floor
DTSTART;TZID=America/Los_Angeles:20240626T143000
DTEND;TZID=America/Los_Angeles:20240626T144500
UID:dac_DAC 2024_sess108_RESEARCH895@linklings.com
SUMMARY:Fake Node-Based Perception Poisoning Attacks against Federated Obj
 ect Detection Learning in Mobile Computing Networks
DESCRIPTION:Research Manuscript\n\nXiong Xiao, Mingxing Duan, Yingjie Song
 , and Zhuo Tang (Hunan University) and Wenjing Yang (National University o
 f Defense Technology)\n\nFederated learning (FL) supports massive edge dev
 ices to collaboratively train object detection models in mobile computing 
 scenarios. However, the distributed nature of FL exposes significant secur
 ity vulnerabilities. Existing attack methods either require considerable c
 osts to compromise the majority of participants, or suffer from poor attac
 k success rates. Inspired by this, we devise an efficient fake node-based 
 perception poisoning attacks strategy (FNPPA) to target such weaknesses. I
 n particular, FNPPA poisons local data and injects multiple fake nodes to 
 participate in aggregation, aiming to make the local poisoning model more 
 likely to overwrite clean updates. Moreover, it can achieve greater malici
 ous influence on target objects at a lower cost without affecting the norm
 al detection of other objects. We demonstrate through exhaustive experimen
 ts that FNPPA exhibits superior attack impact than the state-of-the-art in
  terms of average precision and aggregation effect.\n\nTopic: Autonomous S
 ystems\n\nKeyword: Autonomous Systems (Automotive, Robotics, Drones)\n\nSe
 ssion Chair: Hokeun Kim (Arizona State University)
END:VEVENT
END:VCALENDAR
