ONJava.com -- The Independent Source for Enterprise Java
oreilly.comSafari Books Online.Conferences.

advertisement

AddThis Social Bookmark Button
Article:
  Wireless Mesh Networking
Subject:   NZ Mesh Project
Date:   2006-09-21 04:19:05
From:   MESHD
Response to: Large scale Meshnetworks

I of course agree, Single radio mesh is dead. Even Tropos who probably have the only decent routing and self healing protocols necessary for single radio to be feasible have shifted to multi radio. I have recently been compiling a lot of info on mesh networks, doing a lot of fact finding for a campus mesh. I am not looking at any single radio models. Most have 3 or 4 and strix has even more but Tropos and Strix prices are a bit silly around NZ$6K to NZ$12K per node depending on model/radios.


To me the strongest contenders as far a good product appear to be Strix, Tropos and for me, local company RoamAD, while others like PRoxim AP-4000MR-LR, SOhoware (both indoor MIMO and their new adaptive Wave routing node just released), Qorvus, 3eti, Trango, Locustworld , theres more.
Also firetide for Backhaul mesh only. With large scale deployments I think the important factors, (to me anyway) are fast handoff for VoIP/SIP traffic, scalability of nodes (adding more client& backhaul radios to nodes rather than more nodes) fully integrated unit (sory firetide) and also a strong, easy to use and stable management appliance or application to ensure minimal downtime and ease of administration rather than throughput over more than 4 hops as a fibre backbone can usually be found every so often in a metropolitan area. However multi radio backhaul mekes this possible anyway.


I'm looking really forward to the implementation phase of this project. I have made my recomedations and know what I want but I am not in control of the budget and have had to give all solutions a fair write up. RoamAD despite being cheaper than even Proxim gear for 2 radio version (and about the same for four radio as Proxim which only has 2), and that the management console is as good or better than both Tropos and Strix. Onjly upside to strix that I can see is that their atheros chipset radios may have a bit better ability for client load that the atheros radios of roamAD (but until I witness both side to side who knows). Tropos, to me, well, I looks very pretty, has powerful 700mw radios and just touches out New Zealand 36dBm / 1watt EIRP limit. But that price, I'll just buy 3 x 4 radio RoamAD nodes instead thanks.


Interesting to see what they go for when we go to tender. No matter what is chosen it'll be a lot more fun than playing with my wrt54g's loaded with Freifunk thats for sure.


Craigusto
Down Under
NZ


1 to 1 of 1
1 to 1 of 1