CCI vs CCC

Clear Channel Assessment

Before any transmission, a station performs CCA to determine whether the medium is idle. CCA has two independent detection paths running simultaneously. Both can declare the medium busy. Either declaration prevents transmission.

The 4 microsecond CCA window

CCA must complete within 4 microseconds before any transmission. Both SD and ED checks run in parallel. The outcomes determine the path:

Network Allocation Vector

Physical carrier sense (CCA-SD / CCA-ED) tells a station what is happening on the medium right now. Virtual carrier sense, implemented through the NAV, tells a station what is scheduled to happen.

Every 802.11 MAC frame carries a Duration field in bytes 2-3 of the MAC header. Any station that receives and decodes a frame reads this field and updates its NAV timer. The NAV counts down in microseconds. While NAV is greater than zero, the station treats the medium as busy regardless of what CCA says.

The CSMA/CA backoff sequence

In a healthy co-channel deployment, this is the ordered sequence every station follows. No frames are corrupted. Throughput is shared across all contending stations.

BSS Color - the 802.11ax hardware solution

BSS Color addresses the core CCC problem: pre-802.11ax stations defer to any OBSS frame above the SD threshold, even at -81 dBm from an AP 200 metres away. 802.11ax introduces a 6-bit BSS Color identifier carried in the HE SIG-A field of every HE PPDU. Each BSS is assigned a unique color (1-63, value 0 is reserved). Receiving stations use this to classify incoming frames as intra-BSS or inter-BSS before applying CCA rules.

The constant product rule

Raising the OBSS_PD threshold lets you ignore weaker OBSS signals and transmit more aggressively. But this creates a problem: the OBSS station you are now ignoring cannot know you are transmitting. If you hit it hard, its receiver will see interference. 802.11ax solves this with a mandatory TX power reduction constraint.

The constraint is designed so that as you raise your OBSS_PD threshold (ignoring weaker OBSS signals), you also lower your own TX power to match. The OBSS station you are ignoring will therefore hear your transmission at a level low enough not to corrupt its ongoing frame. This is the constant product constraint: ignore at the same level you reduce your impact.

Intra-BSS NAV vs OBSS NAV

802.11ax introduces separate NAV timers for intra-BSS and inter-BSS frames. Pre-802.11ax, all NAV was treated equally. In 802.11ax, a station performing Spatial Reuse can override the OBSS NAV if OBSS_PD conditions are met. The intra-BSS NAV remains mandatory - Spatial Reuse cannot override your own BSS's medium reservation.

BSS Color is the marking, Spatial Reuse is the action

These two terms get used interchangeably and they are not the same thing. BSS Color is a 6-bit label carried in every HE PPDU. On its own it does nothing except let a receiver tell intra-BSS frames from inter-BSS frames. Spatial Reuse is the behaviour that acts on that label: ignoring a low-power inter-BSS frame and transmitting anyway. The marking is inert. The action is what changes the airtime.

The cleanest analogy is QoS. Marking a frame with a DSCP value or an 802.11e access category does nothing by itself. It is the action taken on the marking, the policing, the queuing, the scheduling, that changes behaviour. BSS Color is the DSCP marking. Spatial Reuse is the queuing policy. Confusing the two is the single most common source of muddled Spatial Reuse discussion, even among experienced engineers.

3 to 5 dB, not 20 dB

The standard permits OBSS_PD to range from -82 dBm up to -62 dBm, a 20 dB span. That figure gets quoted as if it were the operating range. It is not. The 20 dB is the ceiling the standard allows, paired with a mandatory transmit power back-off that grows as you raise the threshold. Practitioners who actually run Spatial Reuse in production report that the useful gain sits in the first 3 to 5 dB of threshold raise. Nobody runs 10 dB or more, because the transmit power penalty cancels the benefit and the failure modes get sharp.

The decision is made at the wrong end

OBSS_PD-based Spatial Reuse has a built-in information problem. The station deciding whether to reuse the medium measures the inter-BSS signal it hears from the other transmitter. But the device that actually suffers if the reuse decision is wrong is the destination station of that other transmission, which may be somewhere else entirely.

Picture AP1, AP2, and a station STA1 sitting at the midpoint between them. AP2 hears AP1 weakly, so it decides the medium is reusable and transmits at reduced power. But STA1, the device AP1 was talking to, is close enough to AP2 that AP2's transmission lands on it as real interference. AP2 made a reasonable decision from its own vantage and still corrupted a frame at a station it could not properly account for. The reuse decision is taken at the transmitter while the consequence lands at the destination. That is the structural critique of OBSS_PD-based SR, and it is why aggressive thresholds bite.

OBSS_PD is the default, PSR and SRG are the rest

Does Spatial Reuse help downlink without client support?

Here is a question the wireless community is genuinely still debating. The AP makes the OBSS_PD CCA-override decision unilaterally on the downlink. That suggests Spatial Reuse could improve downlink throughput even when the clients are legacy and cannot read or set a BSS Color, in the same way RX-SOP improved downlink for years with zero client support. The counter-view is that meaningful SR with BSS coloring needs participating clients, and that ad-hoc testing without them has been inconclusive.

This is presented here as a live open debate rather than a resolved answer, because that is what it is. Engineers with deep field experience disagree, the measured evidence is thin, and it is exactly the kind of question a careful per-frame capture analysis could eventually settle. If you see a confident one-line answer either way, be skeptical.

Wi-Fi 8 Coordinated Spatial Reuse

802.11bn (Wi-Fi 8, Ultra High Reliability) is drafting Coordinated Spatial Reuse (Co-SR), which targets the destination-STA gap directly. Instead of each transmitter deciding alone, neighbouring APs coordinate their reuse decisions and power levels, so the device that would suffer interference is accounted for before the reuse happens. It is the standards body's answer to the structural critique above. The amendment is still in draft, so treat any specific Co-SR behaviour as provisional until TGbn publishes.

What CCC looks like in a PCAP

Co-channel contention leaves five distinct signatures in a PCAP capture. None of them prove the cause in isolation. Together they build a picture. The key diagnostic question is whether high retry rate correlates with visible OBSS traffic - that confirms contention rather than RF noise or driver issues.