Lets make wifi fast again!
 help / color / mirror / Atom feed
From: bob.mcmahon@umbernetworks.com
To: Make-Wifi-fast <make-wifi-fast@lists.bufferbloat.net>
Cc: Jiml <jiml@quicksmart.com>,
	Igor Aleinikov <igor.aleinikov@adnacom.com>,
	Martycieslak <martycieslak@yahoo.com>,
	"Dave.seddon Ca" <dave.seddon.ca@gmail.com>,
	Robin Jarry <rjarry@redhat.com>, Thomas <thomas@monjalon.net>,
	Morten <morten@broerup.com>,
	Tim Odriscoll <tim.odriscoll@intel.com>,
	William Fisher <zzyzxr99@gmail.com>, Jim <jim@iniholdings.com>,
	Hawkinsw <hawkinsw@obs.cr>, Sebastian Moeller <moeller0@gmx.de>,
	robmoser@gmail.com, Koen DS <koen0607@gmail.com>,
	Firas Shaari <Firas.Shaari@shaariconsultancy.com>,
	Sblanchard <sblanchard@vaunix.com>,
	Frantisek Borsik <frantisek.borsik@gmail.com>,
	Jim Stewart <jstewart95066@gmail.com>
Subject: [Make-wifi-fast] Review request: Fi-Wi 802.11 development test rig
Date: Thu, 25 Jun 2026 21:30:53 -0700	[thread overview]
Message-ID: <b3368e0533a8ff57dd7b608c7d051045@umbernetworks.com> (raw)

Hi All,

I'm considering building a new 802.11 development test rig and writing 
the control and measurement software for it. Much of the software 
foundation is already in place.

The goal is to create a reproducible platform for testing whether 
centralized MAC scheduling and MCS selection can improve dense 
in-building Wi-Fi. The analogy I have in mind is the shift from Ethernet 
hubs to Ethernet switches: moving from distributed contention to a more 
controlled forwarding architecture.

The design and parts document is here:

https://www.umbernetworks.com/umber-devel-test-rig.html [1]

The rig is intended to support controlled A/B testing between a Fi-Wi 
RRH and a conventional AP under matched 2x2 MIMO channel conditions and 
repeatable dense-BSS interference. The first version validates the 
single-RRH RF/MAC-control boundary using GPS-synchronized one-way delay, 
throughput, and MAC-layer telemetry. A later multi-cell version would 
extend this to multiple RRHs, multiple STAs, and inter-RRH coordination.

The baseline AP should not be a strawman. It will be configured with the 
best available distributed Wi-Fi controls, including FQ-CoDel, CAKE, 
AQL, AC_N15 EDCA parameters, and L4S AQM where supported. The point is 
to test what remains after modern queue management and airtime queue 
limiting are applied: EDCA contention, CCA deferral, retries, rate 
adaptation, TXOP behavior, aggregation, and local AP-only MCS decisions 
under dense-BSS load.

The important point is that the rig is designed to be falsifiable. It 
should be capable of proving Fi-Wi wrong under controlled conditions. 
That is what would make positive results credible.

Reviews and comments are welcome, especially on the RF path, calibration 
method, dense-BSS load model, baseline AP configuration, queue/AQM 
setup, AQL settings, and measurement software.

Thanks,
Bob

Links:
------
[1] https://www.umbernetworks.com/umber-devel-test-rig.html

                 reply	other threads:[~2026-06-26  4:30 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://lists.bufferbloat.net/postorius/lists/make-wifi-fast.lists.bufferbloat.net/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=b3368e0533a8ff57dd7b608c7d051045@umbernetworks.com \
    --to=bob.mcmahon@umbernetworks.com \
    --cc=Firas.Shaari@shaariconsultancy.com \
    --cc=dave.seddon.ca@gmail.com \
    --cc=frantisek.borsik@gmail.com \
    --cc=hawkinsw@obs.cr \
    --cc=igor.aleinikov@adnacom.com \
    --cc=jim@iniholdings.com \
    --cc=jiml@quicksmart.com \
    --cc=jstewart95066@gmail.com \
    --cc=koen0607@gmail.com \
    --cc=make-wifi-fast@lists.bufferbloat.net \
    --cc=martycieslak@yahoo.com \
    --cc=moeller0@gmx.de \
    --cc=morten@broerup.com \
    --cc=rjarry@redhat.com \
    --cc=robmoser@gmail.com \
    --cc=sblanchard@vaunix.com \
    --cc=thomas@monjalon.net \
    --cc=tim.odriscoll@intel.com \
    --cc=zzyzxr99@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox