×

QR-Buddy: HMAC-Based Dynamic Attendance Tokens with Atomic Replay Control and Geofencing

Author : R. Vigneshwaran, N. Mageshwaran, K. Ramkumar, M. N. Vasu, M. Lavanya Journa Name: International Journal of Science, Engineering and Technology Volume: 14 issue: 6 Year: Volume-14-issue-6 Views : 5
Abstract:
QR-based attendance systems can reduce manual record keeping, but a scanned code alone does not establish that a submitted credential is current, unused, or associated with an acceptable location. This paper presents QR-Buddy, a prototype attendance-verification system that combines rotating QR credentials, HMAC-SHA-256-based token construction, server-side expiry checks, conditional single-use token consumption, student-registration validation, and a geofencing rule. The documented configuration refreshes the displayed QR value every 30 seconds and sets a maximum token-record validity of 120 seconds. The token construction uses department and session identifiers, a time-window value, and a 128-bit random nonce. The documented software components comprise a React/TypeScript client, a Node.js/Express verification service, a Python/Flask administration portal, and MongoDB Atlas persistence. The location rule calculates great-circle distance from the coordinates submitted by the client using the Haversine formula and accepts a request when the calculated distance is no greater than the configured 1,500 m threshold. Project records report tests involving 600 adversarial payloads distributed across six categories; all 600 were recorded as rejected in that selected test set. A separate synthetic API benchmark is reported as 500 requests with 10 concurrent workers, an average latency of 1.85 ms, a median latency of 2 ms, a 99th-percentile latency of 13 ms, a maximum latency of 26 ms, and peak synthetic throughput of 4,545.45 requests per second. These values are reported project measurements: the underlying test scripts, raw logs, hardware details, software versions, database tier, and workload specification were not available for independent reproduction during manuscript preparation. Consequently, the reported rejection outcome does not establish resistance to all attack classes, and the API benchmark must not be interpreted as end-to-end classroom capacity. In particular, a Haversine calculation does not authenticate the source of client-reported coordinates or independently prevent GPS spoofing or live QR relaying. The paper describes the architecture and verification logic, reviews closely related attendance-security approaches, reports the available measurements with their evidence limitations, and identifies reproducibility, concurrency, failure-handling, and location-manipulation tests required for a stronger evaluation.

Related Indexing Platform

Indexed

Leave Your Comment

Related Reviewers

Chat with Expert