Feeds
|
Location:
Fri, 30 Jan 2004
Fun with WiFi in the office
The second i turned on a Cisco 1200 to configure it, two people from down the office called me, giving me signal reports. Its like people are watching their laptops, waiting eagerly for this mystical corp wifi network to power up. Im going to put three APs around the area, and do a more thorough survey. I need to make sure the spots I picked will cover this office seemlessly or not. Observer doesnt seem to be reporting dBm, just some lousy percentage, so I think im going ot have to install netstumbler to supliment this $10k software package...rofl. Tue, 27 Jan 2004
Fun stuff with the 1200
Signal propagation is a little better than i remember in the office, im suspecting its the 100mW radio vs. the 30mW orinoco i used earlier. I am getting a bit of signal leakage outside on the ground, somewhere around -85 to -95dBm, which means that someone with a laptop near the building (or a high gain antenna at a distance) can access our network. We are throwing this network on a leg off our border (public addresses, outgoing connections only, no access to corporate network). In order for people to access corp resources, they would simply VPN in to our VPN server located on our public segment. Anyone could jump on the WLAN if they wanted to, but with all the bandwidth available at our office, no one is particularly concerned. I'm going to need to remember to block outgoing 25 on the ACLs, just to prevent drive by spamming. I am going to put a BS WEP key on the APs, just to prevent accidental associations and people looking for drive-by easy-access. They could just go down to the next high rise and jump on NEC's network anyway...
Wed, 21 Jan 2004
Crazy Idea: Electronicless Internet
I propose an interesting question, try not to lose your lunch, but: Could the concept of the Internet be created in pre-electronic days? I define the "concept of the internet" being any system, automated, which allows the communication of multiple end terminals to a centralized network of servers. The system would not be limited or expected to be any particular size. It could function only in the confines of a building or room, or stretch for miles. Two applications would need to be deployed for basic productivity:
This system would obviously be based off of some mechcanical means. Such a system wouldn't need to be based off of any elaborate protocols, but be able to serve information between various hosts in an effective matter. Modes of serial/parallel data communication:
Modes of Display: The display device would display information downloaded. There would be no requirements to store data, and the display method is welcome to have significant impact on protocol design.
Modes of Information Storage There are no limits to how many pages or files this system needs to store. We will set this at a soft limit of at least 5 pages, so some sort of browsing experience is created.
Modes of Input
Other Stuff I guess thats ok for now ... Tue, 20 Jan 2004
Hack Night
I thought it was a pretty slick package -- a nice tripod mount with a waterproof enclosure bolted to the side. Inside was some sort of routerboard, cant remember the name. It had a couple ethernet interfaces, PCI slot, pcmcia slots, etc. They popped in two Senao cards...one was the old 100mW version, the other is a newer 200mW. To be mounted to this mast will be a 24dBi parabolic (aimed at Matt's node on Capital Hill), and a 15dBi omni. This omni should provide about -75dBm signal to any pcmcia card (-4dBm) within 2 miles. Talk about lighting up the entire north part of downtown! Near the end of Hack Night was some really weird talk of future "projects". The drift of it was more along the lines of building an application rather than just building a network to ping back and forth with. The application's needs would then drive the requirements and motivation to install specific wireless links and hardware in specific locations. Sounds like a better approch to me. When the personal computer was first invented, it was a solution without a problem. Same goes for the internet (and it still is the case depending on your outlook) Maybe we can learn from his and actually come up with some problems to drive network development.
Passive Repeaters
You may have seen one of these bellow -- a weird panel, with no wires, on the side of towers. What is it? A passive repeater! They are good from 1.7GHz-40GHz, and they are all over the place. Its already installed and free to use.
It might be worth the effort to catalog the locations of these passive repeaters, and run them against our node databases to see if they can be put to good use.
[/seattlewireless]
permanent link
Fri, 16 Jan 2004
LOS to Seattle??
I have a little bit of tree-age in the way of this 20 mile shot, but a couple of trees hasn't stopped be before. I'll have to get on the roof in the next couple of days with a 24dBi to see if I can hit that cortland radio (blackberry) or at least something. Wed, 14 Jan 2004
Hack Night
I've been toying around with shooting some sort of segment for SeattleWireless TV, kind of a Myth Busters / Bill Nye deal on wireless HowTos and common myths about wifi. Some of the guys down there offered to help me shoot it and use some of the video gear. I was thinking a couple of myths that are endlessly debated on mailing lists:
Passive repeaters A lot of people are dumbfounded that something that doesn't plug into the wall, has no radio or amplifier could possibly work. This demo would consist of a passive repeater shot. First, do the math. Second, the shot without the passive repeater (showing marginal or no signal). Last, the shot with the passive repeater in place. Show the signal near full strength. ...and for the end, we'll take I, Cringley's shot and prove it with my nifty formula i have for passive repeaters. There's a rumor going around that Cringley's repeater was a bunch of BS, it never was installed...that needs to be stopped. ISM devices vs. WiFi Its always been a running debate whether or not these devices a) interfere at all b) degrade service c) completely make the wifi signal unusable until it turns off. We would do throughput and signal tests without any devices. Then, we'd bring an onslaught of random 2.4GHz devices -- bluetooth, a video sender, a nasty FHSS portable phone, microwave oven, etc. WAP11/WRT54G There are a lot of power hacks out there for the linksys gear. One proved to be rather dirty. Prove the WAP11 once and for all, show the new WRT54G hack to see what it actually does. Would need to borrow the spectrum analyzer from mike or work. 802.11 MAC timing Do a really long shot (maybe use baldi or some other temp arrangement) and see if we can hit some nasty timing issues. Use a 802.11 MAC analyzer (i got the software at work) to evaluate what happens. Pringles Can Take all these ghetto antennas and use test gear to determine how much real gain we get, along with possible pattern issues, SWR figures, and maybe some math to prove that these antennas suck ass.... 2.4GHz emission study I think this would be the most interesting. Matt says he can feel when one of my big ol' nasty antennas are bleching out tons of power. I have some great ideas for this one...
Tue, 13 Jan 2004
New toys
First is a really cool handheld spectrum analyzer. It works from DC to 3.3GHz, and has a series of antennas tuned to various frequencies.
[/seattlewireless]
permanent link
WiSPoT updates
The plan is to have every CWN, big and small, publish their node databases in XML/RDF format. This way people have an official source of data for CWN hotspots in a particular area. Node databases and hotspot finders can then scrape this data for whatever purpose, rather than scouring the web and looking at unreliable wikis, rumors, and heresay. All of this data will then be published in my master wispot.rdf file. This makes applications like my cellular phone app, WiSPoT Mobile, work on a global scale.
[/seattlewireless]
permanent link
|
Make some extra cash with your blog too: |
