Human-interactive iot microcontroller with silent background-mode CRC protection unit for runtime code memory integrity verification

Yongtae Kim, Jeonghun Cho, Daejin Park

Research output: Contribution to journalArticlepeer-review


This paper presents a safe microcontroller (MCU) architecture and code memoryintegrity verification method in background mode for human-interactive Internet-ofThings (IoT) applications. The MCU is now being widely used as a main brainexecuting software in human-wearable, human-monitoring, human-observingapplications. System designers have to provide methods for detecting unsafeconditions by software-driven or hardware support because unsafe operations resultin human damage. Small, fast, safety-conscious operations are critical in designingsafe MCUs that provide human-interactive services, due to additional hardware andsoftware overhead. Our proposed work is based on silent integrity verification byconsidering human-activity for fast monitoring of unsafe conditions with smallhardware/software overhead. The proposed silent background mode protection unitmonitors the code memory access patterns of the interrupt service routines, which areinvoked by human interrupt. This unit seamlessly detects abnormal code executions,which may affect unsafe operations to the human body without any CPU wait cost.The 64KB ROM integrity verification in the experimental 8051 MCU requires onlyabout 24ms of execution time at an 8MHz clock speed without any CPU wait statebecause it runs silently in background mode, and it requires an additional 1650 gates.

Original languageEnglish
Pages (from-to)149-156
Number of pages8
JournalInternational Journal of Mechanical Engineering and Technology
Issue number10
StatePublished - Oct 2018


  • Human-interactive Internet-of-Things services
  • Safe operations and Runtime verification


Dive into the research topics of 'Human-interactive iot microcontroller with silent background-mode CRC protection unit for runtime code memory integrity verification'. Together they form a unique fingerprint.

Cite this