An Exploration of Authentication Protocols for IoT Edge Devices

Document Type

Event

Abstract

Authentication protocols for resource-constrained devices constitute a significant issue facing the Internet of Things community. We have explored authentication protocols for the ESP32 device. The ESP32 is an incredibly resource-constrained device with only 520 KB of SRAM. Initially, we explored different permutations of the binary strings associated with identifiable information for ESP32 device nodes, such as unique node identification and group identification. Binary digit permutation was efficient and minimal but ineffective because a brute force attempt will very quickly poison the network. This discovery called for an exploration of more robust digital signature algorithms. We explored the elliptic curve asymmetric encryption scheme and elliptic curve digital signature algorithm. The elliptic curve cryptography provides extremely powerful protection while being more efficient than other asymmetric encryption algorithms. The mathematical complexity of the elliptic curve over a finite field was the first challenge we overcame. However, the elliptic curve discrete logarithm complexity provided the optimal tradeoff between the desired security and ESP32 device capabilities. We analyzed different available software libraries for elliptic curves over a finite field implementation and secure hash algorithms suitable for extremely resource-constrained ESP32 devices. Our work aims to examine the effect on network performance of the ESP-NOW protocol for devices using the elliptic curve digital signature algorithm. We will examine the packet loss of a single receiving device with many nodes communicating to the receiver, and the impact of the authentication protocols on a busy network’s performance will also be measured and analyzed. Future research and expansion of this project will involve developing efficient and effective methods for key distribution, storage, and maintenance.

This document is currently not available here.

Share

COinS
 
Apr 8th, 2:00 PM Apr 8th, 2:15 PM

An Exploration of Authentication Protocols for IoT Edge Devices

Breakout Session A: Computer and Data Sciences

CASB 102

Authentication protocols for resource-constrained devices constitute a significant issue facing the Internet of Things community. We have explored authentication protocols for the ESP32 device. The ESP32 is an incredibly resource-constrained device with only 520 KB of SRAM. Initially, we explored different permutations of the binary strings associated with identifiable information for ESP32 device nodes, such as unique node identification and group identification. Binary digit permutation was efficient and minimal but ineffective because a brute force attempt will very quickly poison the network. This discovery called for an exploration of more robust digital signature algorithms. We explored the elliptic curve asymmetric encryption scheme and elliptic curve digital signature algorithm. The elliptic curve cryptography provides extremely powerful protection while being more efficient than other asymmetric encryption algorithms. The mathematical complexity of the elliptic curve over a finite field was the first challenge we overcame. However, the elliptic curve discrete logarithm complexity provided the optimal tradeoff between the desired security and ESP32 device capabilities. We analyzed different available software libraries for elliptic curves over a finite field implementation and secure hash algorithms suitable for extremely resource-constrained ESP32 devices. Our work aims to examine the effect on network performance of the ESP-NOW protocol for devices using the elliptic curve digital signature algorithm. We will examine the packet loss of a single receiving device with many nodes communicating to the receiver, and the impact of the authentication protocols on a busy network’s performance will also be measured and analyzed. Future research and expansion of this project will involve developing efficient and effective methods for key distribution, storage, and maintenance.