Improving reception

My daughters moved out of college for the summer into a student hostel. They have a wireless network but her room is furthest away and the reception is very poor. She can only just see the network to log on but it keeps dropping out when ever she tries to do anything - her laptop works fine closer to the transmitter. Her laptop does not have a SMA plug so she can't use a higher gain antenna. Obviously she can't put in a repeater so what are her options to improve reception?

I've got a spare netgear DG834G router, would it be likely to get better reception? Or how about the DG834 with a better antenna?

TIA Slatts

Reply to
Sla#s
Loading thread data ...

"Sla#s" hath wroth:

If the DG834G has a wireless client mode, it can be made to work. However, I don't think it does. That won't help.

Neither will an antenna reflector as the antenna is inside the laptop. However, if you're curious, see:

It's also possible to butcher the laptop and add an external antenna:

I've done a variation on this theme with a U-FL to SMA pigtail cable. The U-FL end goes into the internal MiniPCI wireless card. The SMA end dangles out the back of the laptop. Ugly, but effective.

Therefore, it has to be an external radio and directional antenna. Anything that claims to be an "ethernet client bridge" or "game adapter" will work. See a list at:

However, be careful with this list as there are some devices on the list that will NOT work as a client.

You can also use a USB wireless device, at the end of a 16ft USB cable to get a better location. There are also PCMCIA cards with external antenna connectors.

Forget the repeater, range extender, and such. Deployed in a dorm, it's an invitation for it to walk away. They don't work all that well anyway.

Reply to
Jeff Liebermann

That can really make a big difference, cheaply. The little Zydas

1211 variants can be had for as little as $5.99 at Fry's if you watch the sales, and out on the end of a long lead they can really get better signal with good positioning, compared to those dreadful mini-PCI internal units that usually come in notebooks. With one of those so-called "active" USB cables (actually a hub) you can get even more reach.

Yes, and quickly!

How do you mean? At the moment, I'm using a cheapie repeater and it makes the difference between no (0) connectivity to the AP with my notebook and a full-speed connection with it.

This repeater is certainly no distinguished piece of equipment, one of those Airlink 101 junkers from Fry's at $19.99, but it does get the paint out of the bucket and onto the wall.

What's not to like?

Reply to
Bezmotivnik

"Bezmotivnik" hath wroth:

Problems with repeaters, range extenders, and range expanders:

  1. Compatibility issues. They just don't work with all routers. Manufacturers of repeaters publish compatibility lists in their support pages. Linksys has come closest to a "universal repeater" with the WRE54Gv2, but there are still compatibility problems. A scan of the titles in the Linksys "range expanders" forum should offer a clue as to their overall compatibility:

"Can't connect to internet" is the most common symptom.

Linksys has added WDS compatibility which is their solution to dealing with other manufacturers. In effect, that puts the compatibility burden on the other manufacturer.

Note that the Wi-Fi Alliance does NOT certify repeaters for Wi-Fi compatibilty.

  1. Topology and layout are critical. I recommend the following exercise in frustration as a good example of why it's important for the end points of the system to *NOT* see each other.

- Setup an AP and a client in one room.

- Measure the thruput using IPerf.

- Add a repeater to the room.

- Measure the thruput again using IPerf.

What usually happens is that the thruput drops to perhaps 10% of the value without the repeater. However, if you somehow isolate the AP from the client, with the repeater roughly half way in between, the thruput will only drop to about 50%. That's because there are far too many packets flying around with the added repeater to transfer data efficiently.

  1. Even under ideal conditions, repeaters reduce maximum thruput. Note the word "maximum". Each hop cuts the max thruput in half. However, if you have a 1Mbit/sec DSL and are running multiple 802.11g repeaters, you probably will not notice the max thruput reduction. If it goes from 54Mbits/sec to 25, and then to 12.5, it's still much more than 1Mbit/sec. The problem is that nobody uses range extenders with the strong signals required to get a 54Mbit/sec association. Usually, it's with seriously weak signals running at the slower 802.11b speeds. It might work with a direct link, but with just one repeater, the wireless is now going slower than the DSL line. One could probably live with that for sharing DSL or satellite, but not the much faster cable modem connection. Those run at 6Mbits/sec and up. A single wireless hop can usually keep up with that. Putting a repeater in the line will require that both hops have at least a 12Mbit/sec connection. Most weak signals can't do that.

  1. There are ways to misconfigure repeaters that wreck connectivity for everyone around. Setting the SSID to "Any" will cause repeaters to regurgitate everything they hear. Setting it to "Linksys" is only a little better.

  2. Repeaters gobble airtime. Basically, each one roughly doubles the number of packets floating around in the air. Since only one transmitter can use a given channel at a time, everyone, including the neighbors, has to wait on for a opening to transmit. In really crowded areas, that can be a challenge. That's also why mesh networks, a form of store and forward repeating, tend to have lousy performance. Too many packets flying around. If you want performance, just reduce the number of transmitters on the air.

  1. I also hate omnidirectional antennas. Repeaters use those so that they can get coverage in all directions. That's nice but they also pickup (and generate) interference in all directions. It would be nice if repeaters would use two directional antennas, that only covered the two trarget areas, but that's difficult to impliment.

  2. I also have some issues with how broadcasts are handled by repeaters, rewriting MAC addresses for transparent bridging, and how repeaters don't work with technologies beyond 802.11g such as MIMO, Afterburner, Turbo-G, and beam forming. I'll save these for later.
Reply to
Jeff Liebermann

What level of security can you reliably get with the repeater? Tried WPA on it?

Steve

Reply to
seaweedsteve

WPA is supported, but only if the AP does. In this case, the AP (over which I have no administrative control) is unencrypted.

Reply to
Bezmotivnik

OK, interesting & educational! I have experienced some of those problems, though not fatally, and did some workarounds -- such as using a directional antenna and I power down the repeater when not in use, as a matter of principle.

Ultimately, though, there was just no other solution that was practicable and in this case the results were more than adequate for the purpose.

I wouldn't suggest a repeater when it wasn't absolutely necessary, but when used intelligently it can be the difference between no connectivity and pretty good connectivity -- a quick & dirty fix, in other words.

Reply to
Bezmotivnik

With limited experience, I found that though WPA was supported in a repeater setup, it caused a lot of problems. To the point where I had to ask what's more important? repeating with a single radio or a WPA secure system? I chose secure system.

Steve

Reply to
seaweedsteve

"Bezmotivnik" hath wroth:

Welcome to Learn By Destroying(tm). Some of those things had to be learned the hard way. Some things just don't work very well.

You're one of the few. I spent some time trying to figure out how to identify a repeater with a wireless sniffer. The plan was to do some Netstumbler style data collection to see how common they were in the area. I ran out of time (and interest) and never finished. I suspect they're not as common as I fear, mostly because they don't work very well.

My issue is with a store and forward repeater. I have fewer problems with two radios, back to back, on different channels. They have their own collection of issues, none of which are as bad as store and forward. The cost is nominal and they come stock with 2 antennas (one per radio). 2 radios is also the way all the mesh vendors are going.

In addition, there are plenty of wire and fiber based datacomm alternatives.

Power Line networking:

Phone line networking:

CATV coax sharing:

Ethernet extenders:

Also, I totally forgot about the WPA encryption issues. Some vendors claim WPA compatibilty with repeaters and WDS service. I have yet to be able to make any of them work. I don't know what they're doing, but I certainly am having problems. Even DD-WRT v23 sp2, which claims WPA support, only seems to work between two identical routers, and then only as a bridge, not when one is a repeater. I might be doing something wrong, but I sure can't see what.

Reply to
Jeff Liebermann

Thanks for the info.

Update. It seems that a lot of the rooms are having reception problems and the authorities are "looking in to it"!

Slatts

Reply to
Sla#s

Try the PepWave AP200 Access Point with DHCP. It has a sensitivity down to -94dbm and works great.

Chris Werner

Reply to
Chris Werner

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.