Wi-Fi Validations: Cheap insurance on an expensive build

• Travel and air cargo logistics. 100,000 sq ft hangar, warehouse plus offices, ten-metre ceilings, nine access points, ten to fifteen agents on rugged tablets running the cargo platform against a live flight schedule.

They had just paid for a complete Wi-Fi build. Nobody had measured whether it actually worked.

• The client funded, cabled and installed the entire network before we were involved. Nine access points, bought specifically for the site.
• The only visible symptom was small enough to dismiss: cargo tablets kept drifting onto the guest router mid-shift and had to be manually forgotten, one device at a time.
• The real problem was that nothing had been measured. No way to know whether the network would hold up under live freight operations, and no baseline to argue from if it did not.
• Finding out the hard way would have been expensive. Troubleshooting Wi-Fi in a running cargo hangar means doing it against a live flight schedule, with freight already moving.

One short, scoped post-deployment assessment and validation engagement, run before the network was trusted.

• Full on-site survey of the warehouse and offices: primary coverage, secondary coverage and SNR on both bands.
• Spectrum analysis and rogue hunt across the hangar, to see what the design actually had to share the air with.
• Configuration and inventory audit: per-radio AP inventory, channel plan, channel widths, regulatory settings and roaming features.
• A remediation plan ranked by severity, so the team knew exactly what to fix first, what could wait, and what needed budget.

9

Problems found on a network that was already built and assumed working

8 of 9

Fixes that cost nothing but a configuration change

1

offline AP no one had yet noticed

Nine problems found and documented while they were still cheap to fix.

• Nine findings, three critical. Not one of them visible from the signal bars everyone was relying on.
• Two rogue consumer routers were eating the 5 GHz spectrum, one of them serving guest Wi-Fi inside the warehouse. Between them they blocked two of the three 5 GHz sub-bands, which is why the corporate access points had been squeezed into the remaining ones.
• The country code was set to the wrong country putting the network at risk of being outside channel and power behaviour.
• One access point was mounted and aimed perfectly, and silently offline, leaving a coverage hole nobody had noticed.
• Eight of the nine fixes were configuration changes. Exactly one additional access point was the only hardware recommended.