Implementacija napada na neuronsku mrežu analizom napajanja FPGA sklopa

undergraduate thesis

• Title (english): Implementation of a power supply side-channel attack on FPGA-based neural network
• Author: Mirko Galić
• Mentor: Adrijan Barić (mentor)
• Committee member: Adrijan Barić (predsjednik povjerenstva)
• Committee member: Jurica Kundrata (član povjerenstva)
• Committee member: Vladimir Čeperić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-09-12, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Digitalni sustavi mogu biti sigurnosno ranjivi na razini programske i sklopovske podrške. Napadi koji koriste ranjivosti sklopovske podrške su tzv. side – channel napadi. U ovom radu na FPGA pločicu je implementirana jednoslojna neuronska mreža s linearnom funkcijom. Izvedeno je mjerenje napona napajanja FPGA sklopa. Napisana je programska podrška za obradu rezultata mjerenja te za njihovo daljnje korištenje. Izvršena je korelacijska analiza napajanja te su korištenjem koeficijenta korelacije pronađeni tajni ključevi, odnosno težine, neuronske mreže. Tim korakom je završen postupak reverznog inženjerstva neuronske mreže

Abstract (english)

Digital systems can be vulnerable at the software and hardware level. Attacks that exploit hardware support vulnerabilities are side - channel attacks. In this paper, a single-layer neural network with a linear function is implemented on an FPGA board. FPGA supply voltage measurement was performed. Software has been written for processing measurement results and their further use. A power correlation analysis was performed and secret keys, ie weights, of the neural network were found using the correlation coefficient. This step completed the process of reverse neural network engineering.

• Keywords: Neuronske mreže; side – channel napadi; FPGA; napad analizom napajanja; korelacijska analiza napajanja; Pearsonov koeficijent korelacije; Hammingova težina; reverzno inženjerstvo
• Keywords (english): Neural networks; side-channel attack; FPGA; power analysis; correlation power analysis; Pearson correlation coefficient; Hamming weight; reverse engineering
• Language: croatian
• URN:NBN: urn:nbn:hr:168:317414
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-14 08:47:41