Internet connected computing devices as Internet of Things (IoT) are increasingly used for many applications including safety critical applications such as healthcare devices, autonomous vehicles and drones. Such computing systems are typically designed as Multi-Processor Systems in Chip (MPSoC) to provide more opportunities to investigate the design trade-offs and optimize the target execution platform based on the application needs. Addressing the security vulnerabilities of these systems have become significantly important. Security issues for MPSoCs can be considered as vulnerabilities of different system components including the processing elements, memory system, hardware accelerators and the interconnection network. Recent studies indicated that the security vulnerability of modern processor microarchitecture and the memory systems may result in side-channel attacks. Side channel attacks refer to retrieving security critical information from the device due to the weakness of physical implementation of the system.
The aim of this research is to investigate the security vulnerabilities of processors and their interaction with the memory system. Due to continuous changes of security threats, employing a run-time potential security threat detection is essential. Architectural and micro-architectural changes are used to improve the security of the processor. The impact of these changes on performance and energy efficiency will be investigate using an FPGA based implementation of the extended processor microarchitecture.
Undergraduate
None
Embedded Systems (405.760, Lab)