Real-Time Secure/Unsecure Video Latency Measurement/Analysis with FPGA-Based Bump-in-the-Wire Security

Area of Science:

• Computer Science
• Electrical Engineering
• Cybersecurity

Background:

• The Internet of Things (IoT) presents significant security and privacy challenges.
• The Internet of Video Things (IoVT) specifically requires robust security for sensitive video streams.
• Remotely Operated Vehicles (ROVs) equipped with cameras are critical IoVT nodes demanding secure, real-time data transmission.

Purpose of the Study:

• To evaluate the latency impact of implementing Advanced Encryption Standard (AES) encryption on IoVT video streams.
• To assess the feasibility of real-time secure video communication for marine ROV applications.
• To compare latency under different AES block lengths and transmission protocols.

Main Methods:

• Utilized a LabVIEW Field Programmable Gate Arrays (FPGAs)-based bump-in-the-wire (BITW) secure core for AES encryption.
• Implemented a software architecture allowing dynamic enabling/disabling of AES encryption for video transmission.
• Measured video latency using H.264 and MJPEG encoding over TCP and UDP protocols with varying AES block sizes (128, 256, 512 bytes).

Main Results:

• The integration of AES encryption showed minimal influence on the video feedback performance of the IoVT node.
• Latency measurements confirmed that real-time constraints critical for ROV piloting are maintained with encryption enabled.
• Performance was evaluated across different AES block lengths, encoding schemes, and transmission protocols.

Conclusions:

• AES encryption is a viable solution for securing IoVT video streams without significantly degrading real-time performance.
• The findings support the deployment of secure communication protocols for ROV remote control and piloting.
• This research contributes to addressing the critical information security challenges within the growing IoVT landscape.