Return-path: Received: from he.sipsolutions.net ([78.46.109.217]:42092 "EHLO sipsolutions.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753677Ab0CQQFR (ORCPT ); Wed, 17 Mar 2010 12:05:17 -0400 Subject: Re: [RFC PATCHv2 0/2] mac80211: cfg80211: Roam trigger support From: Johannes Berg To: Juuso Oikarinen Cc: "linux-wireless@vger.kernel.org" In-Reply-To: <1268805772.10120.603.camel@wimaxnb.nmp.nokia.com> References: <1268657751-1042-1-git-send-email-juuso.oikarinen@nokia.com> <1268776349.8918.9.camel@jlt3.sipsolutions.net> <1268805772.10120.603.camel@wimaxnb.nmp.nokia.com> Content-Type: text/plain; charset="UTF-8" Date: Wed, 17 Mar 2010 09:05:16 -0700 Message-ID: <1268841917.5989.7.camel@jlt3.sipsolutions.net> Mime-Version: 1.0 Sender: linux-wireless-owner@vger.kernel.org List-ID: On Wed, 2010-03-17 at 08:02 +0200, Juuso Oikarinen wrote: > > I'm curious. How does the HW/firmware do this? > > > In my testing on some hw, the beacon RSSI fluctuated wildly, so you'd > > have to set the hysteresis pretty high. > > The HW/firmware can be configured with parameters related to averaging, > which will smoothen the worst of the fluctuation at least between single > frames. Ok, cool. Thanks for taking the time to answer :) > Still, there is the WLAN tendency of a very steep RSSI curve, so we'll > probably end up with a relatively high hysteresis. Figuring out good > values for the config is still to-be-done. Right. And it will be a tradeoff between roaming decision speed and battery life. johannes