From mboxrd@z Thu Jan 1 00:00:00 1970 Authentication-Results: mail.toke.dk; spf=pass smtp.mailfrom=; dkim=fail; arc=none (Message is not ARC signed); dmarc=fail (Used From Domain Record) header.from=umbernetworks.com policy.dmarc=quarantine Received: from mail.umbernetworks.com (mail.umbernetworks.com [198.74.51.139]) by mail.toke.dk (Postfix) with ESMTPS id E13FD11D747F; Mon, 08 Jun 2026 07:18:05 +0200 (CEST) Received: from webmail.umbernetworks.com (files.umbernetwork.com [198.74.51.139]) by mail.umbernetworks.com (Postfix) with ESMTPA id EC1EA214A5F; Mon, 08 Jun 2026 05:18:03 +0000 (UTC) MIME-Version: 1.0 Date: Sun, 07 Jun 2026 22:18:03 -0700 From: bob.mcmahon@umbernetworks.com To: David Lang Cc: Sebastian Moeller , Frantisek Borsik , codel@lists.bufferbloat.net, dan , Cake List , Make-Wifi-fast , bloat , Jiml , William Fisher , Thomas , Tim Odriscoll , Koen DS In-Reply-To: <04196260-1r1n-3qpr-39qs-7r40p1183s5o@ynat.uz> References: <8e14c6935753c6263351ad00ec59b9cb@umbernetworks.com> <055e42685cddfa4c1a4ff4da089996eb@umbernetworks.com> <7455E3B4-7FB4-4D40-A900-B31151D12F6F@gmx.de> <52CE3DA7-EC5A-4FF8-A88E-26A7A6661983@gmx.de> <8qr7qons-5sp9-o30o-49qr-02p67prss6rr@ynat.uz> <04196260-1r1n-3qpr-39qs-7r40p1183s5o@ynat.uz> Message-ID: <96b49cf26ef92dc866b072f1f455b648@umbernetworks.com> X-Sender: bob.mcmahon@umbernetworks.com Content-Type: text/plain; charset=US-ASCII; format=flowed Content-Transfer-Encoding: 7bit X-MailFrom: bob.mcmahon@umbernetworks.com X-Mailman-Rule-Hits: nonmember-moderation X-Mailman-Rule-Misses: dmarc-mitigation; no-senders; approved; loop; banned-address; emergency; member-moderation Message-ID-Hash: 3WS5L4IGFDHO3JLE7XVMR4YBIFBVGLZN X-Message-ID-Hash: 3WS5L4IGFDHO3JLE7XVMR4YBIFBVGLZN X-Mailman-Approved-At: Mon, 08 Jun 2026 12:36:18 +0200 X-Mailman-Version: 3.3.10 Precedence: list Subject: [Make-wifi-fast] Re: [Bloat] Re: [Codel] [Rpm] Re: Re: [Cake] "Fi-Wi is a new forwarding plane for wireless" - Bob McMahon List-Id: Lets make wifi fast again! Archived-At: List-Archive: List-Help: List-Owner: List-Post: List-Subscribe: List-Unsubscribe: > you may be able to make a dent in the AP use of airtime, but how are > you going to coordinate the wifi devices you are talking to that are > using so much of it, and doing most of the hidden transmitter stomping > on each other? you can do channel allocation and lower the AP power to > reduce (and if you can use DFS channels almost eliminate) the APs > stepping on each other, but doesn't solve the mobile device > coordination. > David, The first answer is RRH density. With many RRHs distributed through the building, each RRH serves a smaller number of STAs in a smaller physical area. STAs associated to the same RRH are much more likely to hear each other, so the hidden-node precondition becomes less common by construction. Hidden node requires two STAs that cannot hear each other but share the same AP. Shrink the cell and the geometry works in your favor before any other mechanism is applied. After that, the toolkit splits into three categories. For all devices, regardless of 802.11 generation, the concentrator can use building-wide PHY visibility to manage association, channel, power, EDCA policy, and protection windows across RRHs. A conventional AP can only do this locally. Fi-Wi can coordinate those choices across the building. That includes coordinated use of standard protection mechanisms, per-RRH EDCA parameters, transmit-power control to reduce interference radius, and TCP ACK pacing to limit uplink window growth without requiring STA cooperation. For devices with L4S-aware transport, the concentrator can mark ECN based on WLAN service time as well as WAN behavior. That gives the sender a congestion signal tied to the radio bottleneck, not just the internet edge. Legacy devices do not get that benefit, but L4S-capable senders can react directly. For 802.11ax and later clients, uplink OFDMA, trigger-based access, MU-RTS, and TWT give the infrastructure more explicit control over uplink timing and resource allocation. That does not make every client a proprietary scheduled endpoint, but it moves cooperative clients much closer to scheduled uplink behavior. There is also a compounding effect from receive quality. Distributed receive combining across multiple RRHs improves uplink SNR, which drives higher MCS. Higher MCS shortens each TXOP. Shorter TXOPs return the medium faster, reducing collision exposure for other STAs. Lower PER from better SNR eliminates retransmissions, which are pure airtime waste. And once phase coherency is established across RRHs, more spatial streams can be driven to the same STA, multiplying throughput per TXOP without consuming more airtime. The net effect is that the total airtime required to serve a given load shrinks, which directly reduces contention pressure on legacy devices. The uplink problem gets easier not just from coordination mechanisms but from making each transmission more efficient. The honest answer is that legacy hidden-node behavior is mitigated, not eliminated. The architectural advantage is that Fi-Wi applies all of these mechanisms from one concentrator with building-wide PHY visibility and a shared cost function. Conventional APs apply them locally, independently, and through vendor-specific firmware. Bob