skip to content

Wi-Fi

How IEEE 802.11 turns shared air into a LAN: the generations and bands, how stations contend for the medium and associate, and how the link is secured. Mostly everyone has debugged it.

on this pageshow

explore

questions

99 · 8 sections

In 802.11 Wi-Fi, what do the 2.4, 5 and 6 GHz bands trade in range, capacity and client support, and which generations use each?

level: juniorimportance: must knowfreq 68%
basics
~20 s

2.4 GHz reaches farthest but is narrow and crowded; 5 GHz trades some range for many more channels; 6 GHz, added in the Wi-Fi 6 era, is the widest and cleanest but serves only Wi-Fi 6E and Wi-Fi 7 clients.

open as a page

Why does a Wi-Fi client whose adapter reports a multi-gigabit PHY link rate measure only a fraction of it in a speed test?

level: juniorimportance: must knowfreq 60%
basics
~20 s

The PHY rate is how fast one frame's payload is modulated in one direction; airtime is shared and half duplex, and preambles, acknowledgements, contention, retries, rate adaptation and the wired path all take their cut before a speed test sees data.

open as a page

In Wi-Fi 6, how does OFDMA differ from MU-MIMO, and which kind of traffic does each one help most?

level: middleimportance: should knowfreq 34%
basics
~20 s

OFDMA divides one channel's subcarriers into resource units so several clients share a single transmission in frequency; MU-MIMO sends separate spatial streams to several clients over the whole channel. OFDMA suits many small frames; MU-MIMO suits large transfers to well-separated clients.

open as a page

A Wi-Fi 7 AP supports 4096-QAM, yet only clients a few metres away ever reach it; why, and why can a wider channel lower a distant client's rate?

level: seniorimportance: should knowfreq 18%
basics
~20 s

Each QAM step adds two bits per symbol and needs roughly 6 dB more SNR, so 4096-QAM survives only at short range; a wider channel spreads fixed transmit power over more spectrum, lowering SNR and the MCS at the cell edge.

open as a page

For an AP and laptop refresh where most clients are still Wi-Fi 6, how would you decide whether to standardise the new access points on Wi-Fi 7 and 6 GHz now?

level: principalimportance: should knowfreq 26%
basics
~20 s

Buy access points for their service life, not today's clients: Wi-Fi 7 APs serve Wi-Fi 6 clients at Wi-Fi 6 speed, and their 6 GHz and multi-link gains arrive as laptops turn over; client roadmap, 6 GHz rules, uplinks and power decide.

open as a page

On an 802.11 network, why do stations use CSMA/CA to avoid collisions instead of detecting them as classic shared Ethernet did?

level: middleimportance: must knowfreq 42%
basics
~20 s

An 802.11 radio cannot listen while it transmits, and a collision happens at the receiver, which the sender may not hear. So Wi-Fi senses the medium, defers, waits a random backoff and treats a missing ACK as failure.

open as a page

In a warehouse, 802.11 handheld scanners at opposite ends of an aisle both reach the AP but not each other, and retries climb; what is happening, and how does RTS/CTS help?

level: seniorimportance: must knowfreq 34%
basics
~20 s

It is the hidden-node problem: each scanner senses an idle channel and transmits, and their frames collide at the AP. With RTS/CTS, the AP's CTS reaches both scanners and sets their NAV, so the other scanner defers.

open as a page

On an 802.11 channel, how do SIFS, DIFS and the random backoff decide who transmits next, and what changes after a missing ACK?

level: middleimportance: should knowfreq 26%
basics
~20 s

SIFS, the shortest gap, lets an ACK or CTS follow its frame before anyone else can contend; a new transmission waits DIFS plus a random backoff. A missing ACK roughly doubles the contention window, up to CWmax.

open as a page

In the 802.11 MAC, how do the Duration field and the NAV give stations virtual carrier sense, and why is physical sensing alone not enough?

level: middleimportance: should knowfreq 30%
basics
~20 s

Most 802.11 frames carry a Duration field announcing how long the medium stays reserved after them. Stations that decode it load their NAV timer and treat the medium as busy until it expires, even when they sense no energy.

open as a page

As the lead setting 802.11 radio defaults for many sites, would you standardise an RTS/CTS threshold or leave it off, and what evidence would change that?

level: principalimportance: should knowfreq 12%
basics
~20 s

Leave RTS/CTS off by default and enable it per site on evidence: uplink retries from clients that cannot hear each other. It adds two control frames per protected exchange and pays off only where hidden transmitters collide.

open as a page

What is an 802.11 beacon frame, what does it carry, and how often does an access point send one?

level: juniorimportance: must knowfreq 40%
basics
~20 s

A beacon is a broadcast 802.11 management frame an access point sends, by default about ten times a second per BSSID, announcing its SSID, timestamp, capabilities, channel, supported rates, security suites (RSN element) and which dozing stations have buffered traffic (TIM).

open as a page

What are the three 802.11 frame types, what job does each do, and how does the Frame Control field tell them apart?

level: middleimportance: must knowfreq 34%
basics
~20 s

802.11 uses management frames (beacons, probes, authentication, association) to build and run the network, control frames (RTS, CTS, ACK, Block Ack) to coordinate delivery, and data frames to carry payload. The Frame Control field's type and subtype bits identify each.

open as a page

In an 802.11 data frame, how do the To DS and From DS bits decide what each of the four address fields holds?

level: middleimportance: should knowfreq 16%
basics
~20 s

Address 1 is always the receiver and Address 2 the transmitter; To DS and From DS place the BSSID and end points. Uplink: BSSID, source, destination. Downlink: destination, BSSID, source. Both bits set adds Address 4 for AP-to-AP or mesh.

open as a page

In 802.11, how do an MSDU, an MPDU, an A-MSDU and an A-MPDU differ, and why does block acknowledgement make aggregation pay?

level: middleimportance: should knowfreq 13%
basics
~20 s

An MSDU is one packet handed to the MAC; an MPDU is one 802.11 frame. An A-MSDU packs MSDUs into one MPDU; an A-MPDU packs whole MPDUs into one transmission. Block Ack confirms many MPDUs at once, so aggregation saves airtime.

open as a page

In an 802.11 capture of a laptop that keeps dropping, how do the Retry bit, sequence numbers and Block Ack bitmaps show frame loss on the air?

level: seniorimportance: should knowfreq 11%
basics
~20 s

Retransmissions repeat a sequence number with the Retry bit set, and Block Ack bitmaps leave bits clear for MPDUs that never arrived; many of either in one direction means loss on the air, while null-data power-save frames or deauthentication point elsewhere.

open as a page

On an 802.11 network, which frame exchanges does a laptop complete between seeing an SSID and sending its first data, and in what order?

level: juniorimportance: must knowfreq 52%
basics
~20 s

An 802.11 station finds the network from a beacon or probe response, completes 802.11 authentication (two open-system frames, or SAE on WPA3-Personal), then an association request and response; on a protected network a key handshake follows before user data flows.

open as a page

On an 802.11 WLAN, one laptop sees the corp-example SSID but never gets an IP address; how do you find which join exchange failed?

level: seniorimportance: must knowfreq 36%
basics
~20 s

Walk the join in order using AP logs, client logs or a capture: no probe response, a non-zero status in the authentication or association response, a teardown right after association during key setup, or a completed join with no DHCP reply.

open as a page

In 802.11, how does passive scanning on beacons differ from active scanning with probe requests, and why does a hidden SSID still answer some probes?

level: middleimportance: should knowfreq 34%
basics
~20 s

Passive scanning listens on each channel for beacons; active scanning sends probe requests, wildcard or naming an SSID, and collects probe responses. A hidden SSID blanks the name in beacons and ignores wildcard probes, but still answers probes that name it.

open as a page

In 802.11 association, what does the station's request carry, what does the access point check before answering, and how does reassociation differ?

level: middleimportance: should knowfreq 24%
basics
~20 s

An 802.11 Association Request carries capabilities, listen interval, SSID, supported rates and, when protected, the RSN element with the chosen suites; the AP validates them and replies with a status code and association ID. Reassociation adds the current AP's address.

open as a page

As WLAN lead, teams keep requesting their own SSIDs, some hidden; what SSID policy would you set, weighing what each extra SSID costs on the air?

level: principalimportance: should knowfreq 18%
basics
~20 s

Keep a small set of SSIDs, one per distinct security model rather than per team, because every SSID on every radio beacons and answers probes; segment users behind the SSID instead, and refuse hidden SSIDs, which cost compatibility and buy no security.

open as a page

In 802.11 roaming, who decides when a client moves to another access point, and what do 802.11k and 802.11v add?

level: middleimportance: must knowfreq 42%
basics
~20 s

The client decides when to roam and which access point to join; the network can only influence it by suggesting, refusing or disconnecting. 802.11k neighbor reports shorten the client's scan, and 802.11v BSS Transition Management requests recommend target APs.

open as a page

In WMM, what are the four access categories, and how do EDCA parameters give voice frames priority over bulk traffic on the air?

level: middleimportance: must knowfreq 37%
basics
~20 s

WMM sorts traffic into AC_VO (voice), AC_VI (video), AC_BE (best effort) and AC_BK (background). Under EDCA each category contends with its own AIFSN, contention window and TXOP limit, so voice usually, but not always, wins the medium first.

open as a page

On a dual-band Wi-Fi network, what is band steering, why do operators enable it, and what are its limits?

level: juniorimportance: should knowfreq 30%
basics
~20 s

Band steering is an access point nudging dual-band clients off crowded 2.4 GHz onto 5 GHz, by withholding or delaying 2.4 GHz replies or suggesting another BSS. It only influences the client, which still picks the band itself.

open as a page

How does 802.11r Fast BSS Transition let a client roam without repeating 802.1X, and what do the PMK-R0 and PMK-R1 key holders hold?

level: middleimportance: should knowfreq 27%
basics
~20 s

802.11r runs the full 802.1X login once per mobility domain and derives a PMK-R0, from which each access point gets its own PMK-R1. On a roam, client and target AP derive a fresh PTK from that PMK-R1 during reassociation, skipping EAP.

open as a page

Hospital voice handsets on an 802.1X Wi-Fi network drop calls as nurses walk between wards, and each full re-authentication takes hundreds of milliseconds; how do you diagnose and fix it?

level: seniorimportance: should knowfreq 22%
basics
~20 s

Time each phase of the roam: scan, reassociation, 802.1X and key handshake. A RADIUS exchange on every roam shows full EAP each time; enable Fast BSS Transition (802.11r) across the wards, add 802.11k and 802.11v, and confirm voice is sent as AC_VO.

open as a page

In the 2.4 GHz Wi-Fi band, why are access points planned on channels 1, 6 and 11 rather than on any free channel?

level: juniorimportance: must knowfreq 62%
basics
~20 s

2.4 GHz channel centres sit 5 MHz apart but signals are about 20 MHz wide, so only 1, 6 and 11 clear each other. Partial overlap is worse than sharing a channel: radios cannot decode it and take turns.

open as a page

In a dense 5 GHz Wi-Fi deployment, why can 80 MHz channels deliver less total capacity than 20 or 40 MHz channels?

level: middleimportance: must knowfreq 42%
basics
~20 s

Each doubling of width roughly halves the non-overlapping channels, so in a dense site more APs share each channel and defer to each other. Shared airtime, lower signal-to-noise and busy secondary channels can outweigh the wider pipe.

open as a page

On 5 GHz Wi-Fi channels 52-144, what does DFS require of an access point, and what happens to its clients when radar is detected?

level: middleimportance: should knowfreq 36%
basics
~20 s

DFS makes the AP listen for radar for 60 seconds before using a channel and keep monitoring; on detection it must leave within 10 seconds, announcing the new channel to clients, and avoid that channel for at least 30 minutes.

open as a page

In Wi-Fi site surveys, how do predictive, passive and active surveys differ, and when in a project does each belong?

level: middleimportance: should knowfreq 34%
basics
~20 s

A predictive survey models coverage from floor plans before installation; a passive survey listens to every AP on site without associating; an active survey associates and measures throughput, retries and roaming. Validation surveys check the installed result.

open as a page

For a 300-seat lecture hall Wi-Fi design, how does planning for capacity differ from planning for coverage in AP count, transmit power and channel width?

level: seniorimportance: should knowfreq 28%
basics
~20 s

Coverage design uses the fewest APs at high power to reach a signal target everywhere; capacity design starts from concurrent devices and their airtime, then uses more APs, lower power and narrower channels so cells stay independent.

open as a page

In enterprise Wi-Fi, how do autonomous, controller-based and cloud-managed access points differ in where configuration, control and user traffic are handled?

level: juniorimportance: must knowfreq 58%
basics
~20 s

An autonomous AP keeps configuration, control and forwarding to itself. A controller-based AP takes its configuration and control decisions from a controller over CAPWAP. A cloud-managed AP is configured from a hosted service but normally controls and forwards locally.

open as a page

In CAPWAP, what do the control channel on UDP 5246 and the data channel on UDP 5247 carry, and which is protected by DTLS?

level: middleimportance: must knowfreq 34%
basics
~20 s

CAPWAP's control channel, UDP 5246 at the controller, carries join, configuration, firmware and station management inside DTLS; only discovery messages go in the clear. The data channel, UDP 5247, carries tunnelled frames, and DTLS on it is optional.

open as a page

In CAPWAP's IEEE 802.11 binding, how do Split MAC and Local MAC divide 802.11 functions between the access point and the controller?

level: middleimportance: should knowfreq 24%
basics
~20 s

In Split MAC the AP keeps real-time work such as beacons and probe responses, while the controller processes associations and receives all user data through the tunnel. In Local MAC the AP terminates the 802.11 MAC and may bridge traffic locally.

open as a page

After a WLAN controller's radio resource management is enabled, clients drop briefly several times a day; how can automatic channel and power changes cause this, and what do you tune?

level: seniorimportance: should knowfreq 22%
basics
~20 s

Radio resource management changes AP channels and transmit power from a controller-wide view. Each channel change makes an AP's clients follow or reconnect, and coverage-hole logic can raise power for one weak client; restrict changes to quiet windows and bound power.

open as a page

When store APs lose their WAN link to a head-office CAPWAP controller, what keeps working on tunnelled versus locally bridged WLANs, and why?

level: seniorimportance: should knowfreq 27%
basics
~20 s

Tunnelled WLANs stop at once, because every user frame must reach the controller. Locally bridged WLANs can keep forwarding store traffic, but controller-run steps such as 802.1X and key management fail unless the AP has a local fallback.

open as a page

In WEP, what do the 24-bit IV and the ICV in each frame do, and why is WEP retired?

level: juniorimportance: must knowfreq 62%
basics
~20 s

In WEP each frame's RC4 key is a 24-bit IV, sent in the clear, followed by the shared key, and a CRC-32 ICV is appended. A small IV space repeats and a CRC cannot authenticate, so WEP is retired.

open as a page

How does a rogue access point differ from an evil twin, and why can't a client tell a twin by SSID and security type alone?

level: juniorimportance: must knowfreq 55%
basics
~20 s

A rogue AP is any unauthorised access point, possibly wired into your network; an evil twin impersonates an authorised SSID. A client cannot tell a twin by name and security type, because both are advertised labels, not proof of identity.

open as a page

Across WEP, WPA, WPA2 and WPA3, what did each Wi-Fi security generation fix, and which should a WLAN still offer?

level: juniorimportance: must knowfreq 68%
basics
~20 s

WEP used one static RC4 key; WPA's TKIP retrofitted per-packet keys and integrity onto that hardware; WPA2 made AES-based CCMP the baseline; WPA3 adds SAE, required PMF and a 192-bit mode. Offer WPA3, WPA2 only for stragglers.

open as a page

On a WPA2 network, what does the "AES" encryption setting actually refer to, and what does it protect in each data frame?

level: juniorimportance: must knowfreq 52%
basics
~20 s

In WPA2, "AES" means CCMP: AES in CCM mode with a 128-bit per-session temporal key. It encrypts each data frame's body and adds an 8-octet MIC that detects tampering, while the MAC header travels in the clear.

open as a page

In WPA2-Personal, how do the client and access point prove they share the passphrase without ever sending it over the air?

level: juniorimportance: must knowfreq 48%
basics
~20 s

Both sides turn the passphrase into the same Pairwise Master Key locally, swap random nonces in the four-way handshake, derive a session key from them, and prove possession with a MIC only the correct key can produce.

open as a page