Embracing Co-Channel Contention: AI Assisted Channel Planning for a 3,000 Student BYOD Digital Exam Venue with No Room for Failure | PRG 26

A stadium can tolerate a few unhappy users. A conference can survive a few dropped connections.

A 2,500-student BYOD digital exam cannot. Traditional Wi-Fi planning tools optimise channel plans by minimising co-channel interference.

In a single exam hall with 3,000 students starting simultaneously, that objective is unattainable. Rather than accepting a sub-optimal design, I used AI to solve a different optimisation problem by embracing co-channel contention while balancing airtime fairness, channel reuse and client distribution.

This talk explains the AI workflow, how the design was validated in the real world using traditional Wi-Fi planning and validation tools, and how it performed in a live digital exams