CWT common mistakes
CWT · Understand the traps and how to avoid them.
Computing received power by subtracting cable and connector loss from transmit power but forgetting that the free-space path loss term grows logarithmically with distance, then choosing an antenna gain that cannot close the link.
Candidates memorize the link-budget formula as a single line and treat dBm and dB as interchangeable, so they add an antenna gain expressed in dBi directly to a milliwatt value and drop the 20log(d) path-loss term. The arithmetic looks plausible because every term is a small number.
What to do instead
Write the budget as a chain in dB: transmitter output in dBm, minus cable/connector losses in dB, plus transmit antenna gain in dBi, minus free-space path loss (32.44 + 20log10(f in MHz) + 20log10(d in km)), plus receive antenna gain, and compare the result to the receiver sensitivity in dBm. Keep a margin of at least 10 dB for fading. Practice by converting one 2.4 GHz link at 100 m by hand until the units discipline is automatic.
Mixing up which 2.4 GHz channels actually overlap and recommending a three-channel plan of 1, 5, and 9 or 1, 6, and 11 without checking that only 1, 6, and 11 are non-overlapping in a 20 MHz plan.
The channel numbering looks evenly spaced, so candidates assume spacing of five channels is always safe. In 2.4 GHz each 20 MHz channel spans roughly five channel numbers, and the non-overlapping set depends on regulatory domain and whether 40 MHz bonding is in use.
What to do instead
Memorize that in North America the practical non-overlapping 20 MHz set is 1, 6, and 11, and that any 40 MHz use in 2.4 GHz consumes most of the band and should generally be disabled in high-density designs. When a question gives a channel list, count the 5 MHz spacing between channel centers and flag any pair closer than five channels apart.
Selecting an access point mounting location based on where cabling is easiest, such as above a suspended ceiling with foil-backed tiles or inside a metal telecommunications closet, and then expecting normal coverage.
Installation questions often describe a convenient cable path, and candidates accept it without evaluating the RF environment. Metal ceiling grids, foil vapor barriers, concrete with rebar, and elevator shafts attenuate 2.4 and 5 GHz signals severely, and mounting inside a closet can cost 20 dB or more.
What to do instead
Before choosing a location, evaluate the obstruction materials between the AP and the intended coverage area, prefer below-ceiling or plenum-rated mounting with the antenna oriented as designed, keep APs away from metal ductwork and fluorescent ballasts, and verify with a post-install survey rather than assuming the cable path is acceptable.
Confusing PoE standards and power budgets, for example assuming an 802.3af source can power an 802.3at access point that requires more than 15.4 W at the port, or ignoring the cable's power loss on long runs.
The standards share the same connector and the wattage numbers are close, so candidates treat PoE as a binary feature. They also forget that the PSE budget, the PD class, and the cable resistance all interact, and that long or thin cable runs dissipate power.
What to do instead
Learn the pairs explicitly: 802.3af delivers up to about 15.4 W at the PSE and roughly 12.95 W at the device; 802.3at (PoE+) delivers about 30 W at the PSE and 25.5 W at the device; 802.3bt adds higher Type 3 and Type 4 tiers. Match the AP's stated maximum draw to the switch port's class, verify the switch's total power budget across all ports, and prefer short runs of Cat5e or better.
Treating a wireless LAN controller as a mandatory component and describing every deployment as controller-based, then failing questions about autonomous or cloud-managed access points.
Vendor marketing and study material emphasize controller architectures, so candidates collapse all WLANs into one model and cannot answer questions about standalone APs, split-MAC operation, or CAPWAP control and data tunnels.
What to do instead
Build a small comparison table: autonomous AP (all functions local, managed individually), controller-based with split MAC (CAPWAP control and data tunnels, lightweight APs), and cloud or locally managed with embedded controller. Note which functions move to the controller and which stay in the AP, and practice identifying the architecture from a short scenario description.
Configuring a service set identifier and security settings but leaving the regulatory domain, channel width, and transmit power at defaults, then being surprised by coverage holes or co-channel interference.
Candidates focus on the visible SSID and password fields because those are what users notice, and treat radio settings as automatic. Defaults often include maximum transmit power and 40 or 80 MHz widths that are wrong for a dense environment.
What to do instead
Treat radio configuration as part of the build: set the correct country code so legal channels and power limits apply, choose 20 MHz in 2.4 GHz and plan 5 GHz widths deliberately, and set transmit power from the survey rather than leaving it at maximum. Verify the resulting channel plan across neighboring APs before going live.
Diagnosing a client that associates but cannot pass traffic by re-checking the Wi-Fi password, instead of following a layered path from association, authentication, VLAN assignment, and DHCP through to routing.
Association problems and addressing problems look similar to an end user, and candidates anchor on the first plausible cause. They also skip the distinction between open, WPA2-Personal, and WPA2/WPA3-Enterprise authentication failures, which fail at different stages.
What to do instead
Work the layers in order: does the client see the SSID, associate, complete the security exchange, receive an IP address, and resolve and reach a gateway? Check the controller or AP logs for the specific failure reason, confirm the client VLAN and DHCP scope, and verify that the RADIUS exchange succeeded when Enterprise authentication is in use. Name the failing layer in the answer rather than the symptom.
Reading a site survey as a single pass and reporting coverage from one measurement location, without accounting for the difference between a predictive survey, a passive survey, and an active or AP-on-a-stick validation.
Survey terminology overlaps and candidates treat any walk-around with a laptop as equivalent. They also forget that a survey taken with the AP transmitting at full power overstates the design that will be deployed at reduced power.
What to do instead
Separate the phases: predictive or planning survey from a floor plan and attenuation assumptions, passive survey measuring existing RF, active survey measuring real client association and throughput, and validation after installation. Survey at the transmit power and channel plan that will actually be deployed, sample multiple locations including edges and stairwells, and record both signal strength and signal-to-noise ratio.
Choosing WEP or WPA with TKIP as an acceptable answer for a new deployment, or describing WPA2-Personal as suitable for an enterprise campus without noting the shared-key limitation.
Older security options remain in configuration menus and in legacy interoperability discussions, so candidates pick the option that maximizes compatibility. They also conflate encryption strength with key management, missing that a pre-shared key cannot be revoked per user.
What to do instead
For new deployments require WPA2 or WPA3 with AES-CCMP or GCMP, and reserve WPA2/WPA3-Enterprise with 802.1X and a RADIUS server for organizational networks where per-user credentials and revocation matter. Treat WEP and TKIP as deprecated and note explicitly why: known cryptographic weaknesses and, for the pre-shared key model, no individual accountability.
Answering rogue access point and interference questions by describing only the security threat, ignoring the non-malicious interferer category such as microwave ovens, cordless phones, Bluetooth, and neighboring networks.
The word rogue triggers a security frame in candidates' minds, and they forget that most interference in real deployments is unintentional. They also miss that a rogue AP connected to the corporate network is a different risk from an external AP broadcasting a similar SSID.
What to do instead
Classify the problem first: authorized AP, neighbor's AP, malicious rogue, or non-Wi-Fi interferer. For each, state the detection method (WIPS or spectrum analysis), the risk (data exfiltration versus performance degradation), and the response (containment, locate and remove, or channel and power adjustment). Include spectrum analysis as the tool for non-802.11 interference.