From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wm0-x22e.google.com (mail-wm0-x22e.google.com [IPv6:2a00:1450:400c:c09::22e]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 487273B260 for ; Mon, 27 Jun 2016 14:22:41 -0400 (EDT) Received: by mail-wm0-x22e.google.com with SMTP id a66so127044768wme.0 for ; Mon, 27 Jun 2016 11:22:41 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=broadcom.com; s=google; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc; bh=aTw/VDe7gYPcSqpHxg9uN5A5l20VZp9cq4kca3dRlUY=; b=TabjXPEgF7pglOD5/oLRXY8KsgkNqqc3aAfHVXMIKCs5qjHR0WeRaF7kb0wy3NOANx qudS5K2PZnjJI/K0ta6e8pp8AtJ06K9jBFzTuZwBV+co/AyYr9vpi3iZcmauFVg0z2G7 hU/FEYiubEQJ/wQGe6hNHPcaaV01z1E8MoGCY= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc; bh=aTw/VDe7gYPcSqpHxg9uN5A5l20VZp9cq4kca3dRlUY=; b=K4+Fc2rQ9CflOt/s0NGX3JV4GImrOSl3l6G3Lv0NI3lLEFN1wQlQAgjM9qDy4EwNbz iRdpYc+V9VPyIOAHFJMvpJWaTpGhOO9prqbJrweD6Tz7EiYq03mWsWoppGcXXVXBjoc8 GEnK1rg9IgD6dv6+rwi8qOYFdfwaL9UpyjYiJfd1a37eoKIWu7JZQykLMvL47xofLWNC KTq6zNu9ypDsYBLXmV0Z9Jvg8MA5E2qA8F6E+eUKUDaGdqiLgRTr3R4KgcUHgin4Ppct FzXFqZ1J1nP7zMI7jAPce22u1f+rufLuQmNrB+Jn6HyBUbR0vgfsN8EQc4MR39nDgJ9c O70A== X-Gm-Message-State: ALyK8tKMALbycPzpfKxc0hLz5ZVbQHK3A6XrbYOpln/ocot0JB/QhroXmw0KLquhdqsHFda0SVL4Slm2gYpcfjuC X-Received: by 10.28.191.90 with SMTP id p87mr11449904wmf.69.1467051759562; Mon, 27 Jun 2016 11:22:39 -0700 (PDT) MIME-Version: 1.0 Received: by 10.194.239.202 with HTTP; Mon, 27 Jun 2016 11:22:37 -0700 (PDT) In-Reply-To: References: <1466803464.927322699@mobile.rackspace.com> From: Bob McMahon Message-ID: To: Dave Taht Cc: David Lang , make-wifi-fast@lists.bufferbloat.net, "cerowrt-devel@lists.bufferbloat.net" Content-Type: multipart/alternative; boundary=94eb2c06ddb2ff9ade0536469a91 X-Mailman-Approved-At: Mon, 30 Mar 2020 07:21:14 -0400 Subject: Re: [Cerowrt-devel] [Make-wifi-fast] more well funded attempts showing market demandfor better wifi X-BeenThere: cerowrt-devel@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: Development issues regarding the cerowrt test router project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Date: Mon, 27 Jun 2016 18:22:41 -0000 X-Original-Date: Mon, 27 Jun 2016 11:22:37 -0700 X-List-Received-Date: Mon, 27 Jun 2016 18:22:41 -0000 --94eb2c06ddb2ff9ade0536469a91 Content-Type: text/plain; charset=UTF-8 Hi All, This is a very interesting thread - thanks to all for taking the time to respond. (Personally, I now have better understanding of the difficulties associated with a PHY subsystem that supports a wide 1GHz.) Not to derail the current discussion, but I am curious to ideas on addressing the overhead associated with media access per collision avoidance. This overhead seems to be limiting transmits to about 10K per second (even when a link has no competition for access.) Is there a way, maybe using another dedicated radio, to achieve near instantaneous collision detect (where the CD is driven by the receiver state) such that mobile devices can sample RF energy to get theses states and state changes much more quickly? Thanks, Bob On Mon, Jun 27, 2016 at 10:55 AM, Dave Taht wrote: > In terms of wifi history... since I go back to the 70s... > > was that we did not know how to do it - 73 we had aloha, which begat > ethernet... and for years progress was slow. Even as late as 91 or so > a "good" microwave link cost something like 40k an end, and required > special cooling and permits on so on. Wifi was started as an ipx/spx > bridge tech that didn't start to get anywhere until the mid 90s. > > It was far from obvious at any point that the cost reductions would > take place that did, there was so much work in the analog domain that > looked (at the time) resistant to moores law. As for spectrum, finding > ways to leverage 2.4ghz cost metricom's backers in particular more > money than I care to think about, and I'm always pointing at how the > discovery that a more centralized clock and a retransmit at the mac > layer is what eventually made 802.11b viable. Many other wireless > ideas have been tried and died - wimax, for example, UWB, for another. > Bluetooth has evolved into adding "discovery prototol", which was kind > of unexpected... (there is even 6lopan over bluetooth now). The 5ghz > spectrum users have tended to adopt their own mac, as has some other > less popular bands. > > While I'm pretty happy that we've got much of the queuing theory for > fixing 802.11n and 802.1ac nailed now, outstanding problems include > the hidden station problem, the rise in the background noise levels, > insufficient channels, and increasingly proprietary standards and > chipsets, as well as transport, switching and routing protocols > layered on top originally designed for isochronus transports. > > I like to think (or possibly delude myself), that the solutions to > airtime fairness scheduling now emerging may one day lead to saner > scheduling around the hidden station problem in particular. Otherwise, > and elsewhere, there remain a lot of rocks to bang together, and a > long list of other issues we've captured elsewhere. > > I am still periodically reviewing and updating this as we go along, as > it remains the best central document we have on all that's wrong in > wifi with some hints as to how to go about fixing them. > > > https://docs.google.com/document/d/1Se36svYE1Uzpppe1HWnEyat_sAGghB3kE285LElJBW4/edit > --94eb2c06ddb2ff9ade0536469a91 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
Hi All,

This is a very interesting thre= ad - thanks to all for taking the time to respond. =C2=A0 (Personally, I no= w have better understanding of the difficulties associated with a PHY subsy= stem that supports a wide 1GHz.)=C2=A0

Not to dera= il the current discussion, but I am curious to ideas on addressing the over= head associated with media access per collision avoidance.=C2=A0 This overh= ead seems to be limiting transmits to about 10K per second (even when a lin= k has no competition for access.) =C2=A0 Is there a way, maybe using anothe= r dedicated radio, to achieve near instantaneous collision detect (where th= e CD is driven by the receiver state) such that mobile devices can sample R= F energy to get theses states and state changes much more quickly?

Thanks,
Bob

=C2=A0 =C2= =A0



On Mon, Jun 27, 2016 at 10:55 AM, Dave Taht <= ;dave.taht@gmail.c= om> wrote:
In terms of wifi= history... since I go back to the 70s...

was that we did not know how to do it - 73 we had aloha, which begat
ethernet... and for years progress was slow. Even as late as 91 or so
a "good" microwave link cost something like 40k an end, and requi= red
special cooling and permits on so on. Wifi was started as an ipx/spx
bridge tech that didn't start to get anywhere until the mid 90s.

It was far from obvious at any point that the cost reductions would
take place that did, there was so much work in the analog domain that
looked (at the time) resistant to moores law. As for spectrum, finding
ways to leverage 2.4ghz cost metricom's backers in particular more
money than I care to think about, and I'm always pointing at how the discovery that a more centralized clock and a retransmit at the mac
layer is what eventually made 802.11b viable. Many other wireless
ideas have been tried and died - wimax, for example, UWB, for another.
Bluetooth has evolved into adding "discovery prototol", which was= kind
of unexpected... (there is even 6lopan over bluetooth now). The 5ghz
spectrum users have tended to adopt their own mac, as has some other
less popular bands.

While I'm pretty happy that we've got much of the queuing theory fo= r
fixing 802.11n and 802.1ac nailed now, outstanding problems include
the hidden station problem, the rise in the background noise levels,
insufficient channels, and increasingly proprietary standards and
chipsets, as well as transport, switching and routing protocols
layered on top originally designed for isochronus transports.

I like to think (or possibly delude myself), that the solutions to
airtime fairness scheduling now emerging may one day lead to saner
scheduling around the hidden station problem in particular. Otherwise,
and elsewhere, there remain a lot of rocks to bang together, and a
long list of other issues we've captured elsewhere.

I am still periodically reviewing and updating this as we go along, as
it remains the best central document we have on all that's wrong in
wifi with some hints as to how to go about fixing them.

https://docs.goo= gle.com/document/d/1Se36svYE1Uzpppe1HWnEyat_sAGghB3kE285LElJBW4/edit

--94eb2c06ddb2ff9ade0536469a91--