Help with Boosting wireless router signal

Please not that I did not recommend using an omni antenna. I only offered it as an option.

That's correct. However, I can't tell from the description where the access point is located relative to ALL the users. A 10dBi panel antenna will have a beamwidth of about 60 degrees which is more than enough to cover the entire house. However, if the garage user is located behind the antenna, there will be problems. I've done indoor directional antennas, but only when the radios are at the ends of house and the antenna is pointed inward. Working from the back of a panel antenna does not work very well.

True. I'll dig out my notes and run some calculations on the theoretical maximum sensitivity at 10^6 BER for various modulation schemes and data rates. What you'll find is that most of the chipsets advertise RX sensitivities that are very close to the noise floor. Unless someone comes out with a cryogenically cooled front end, the current crop of GaAsFET front ends are about as sensitive as can be. Incidentally, when I was doing bench tests for sensitivity and other specs, I found that the digital noise sprayed all over the circuit board by the non-RF parts of the circuit had a substantial effect on the receive sensivity. On the better boards, with little digital noise to ruin my tests, the variation in measured sensitivities over a production lot was as bad as 6dB. These were basically rejects as the bulk of the test boards were within a 2dB spread.

So, lets see if the numbers make sense. The worst case sensitivity deterioration that I saw was 6dB (in dBm not dBv). Yet the suggested power amplifier will go from +17dBm to +30dB or an increase in 13dB TX gain. However, the receiver sensitivity only varies 6dB so we have anywhere between 7 and 13dB of overkill in one direction. I don't think that cranking up the tx power to compensate for a crappy receiver is going to do much good.

I like your list better except for the repeater and range extender issue. Incidentally, if you had specified a WDS repeater, I might be a bit less venement about my hatred of repeaters.

A variation of this is to seperate the access point (radio) from the router/modem section. The problem is that the conglomerated modem/router/access-point devices want to live where all the wires tend to come together. That's usually in the basement, in a closet, under a desk, or other places where hiding the wires is practical.

However, the radio does not work very well buried in these locations. Therefore, by seperating the radio section (access point) from the rest of the boxes, there's quite a bit more versatility in locating the box and antenna. Up high, on a bookshelf is a good place. There are only two cables (power and ethernet) required so wiring isn't as bad as to a modem/router which has many more cables involved. It also has the advantage of allowing easy replacement of the radio section when the next generation of acronyms appear on the market.

Linksys makes SM01 mounting plate for their plastic boxes. Costs about $2/ea. |

As you noted with the WG511, it's somewhat difficult to differentiate between the good stuff and the junk. I'm in the biz and I get fooled a bit too often.

I would repeaters and range extenders in the desperation category.

That's what I would do first or 2nd. Not because it's the best solution but because it's one of the easiest to try and probably the cheapest.

A repeater will also piss them off if they're within range by doubling the amount of traffic floating in the air. However, at least the repeater only trashes one channel. However, the 2nd access point could easily be on the same channel as the main wireless router if (and only if) they are sufficiently seperated and isolated from each other so as not to interfere with each other.

Yep. Also impossible to add external antennas.

Well, they are a pain but not bad it you're willing to tolerate some directionality. Most users will NOT tolerate directionality on devices like cell phones, where a change in user position can easily result in a dropped call. If the directionality is really obvious, like a panel antenna plastered on the back of a laptop lid, methinks most users can figure out where to aim it. If the laptop or PDA is too small for a decent directional gain antenna, then I agree, it's impractical.

(I have some antenna design ideas and few prototypes that methinks may be practical, but nothing that will work for every possible client radio or attached computer/PDA).

Thanks. At least we agree on one out of nine items.

If I remember the bad old days of CB radio burners were quite popular, wonder if anybody has come up with one for wireless networks yet?

[POSTED TO alt.internet.wireless - REPLY ON USENET PLEASE]

Yes. As I wrote in my follow-up, I've gotten good results without the downside of using a different channel. To be fair, both cases were 802.11g networks running in 802.11g mode only.

As noted in my earlier replies, I select hardware carefully. ;)

My experience: In the real world they often (usually?) won't be, and on different channels (as you recommended in your earlier post), they will pollute an additional primary channel, which I personally think is the greater sin.

Indeed, in that special case. Where they overlap, as they often do, interference can be a big issue.

[POSTED TO alt.internet.wireless - REPLY ON USENET PLEASE] :

In indoor applications, directional antennas can provide better performance than omnidirectional antennas. This is not due to the gain increase typically associated with directional antennas, but rather to backside and off-axis rejection that can reduce multipath cancellation. Not all directional antennas (e.g., Yagi antennas) have much backside rejection.

Antenna gain results from focusing transmitted energy into a smaller cross-sectional area. Ideal radiators, called isotropic radiators, radiate energy in all directions from a point source at equal intensity. Limiting the radiated energy to a portion of this ideal sphere increases signal intensity in the focal area.

Not necessarily -- depends on which device is operating closer to its noise floor. If that's the client, then an increase in access point transmit power can help, and vice versa.

Fair enough -- we disagree.

Except for our different feelings about repeaters, I don't think we all that far apart, except that I'm perhaps a bit more pragmatic and sensitive to creating problems for neighbors. My order and sequence would be (in order).

1. Move the wireless router closer to the clients. Ideal, but often not possible/practical.
2. Better access point antenna orientation. Easy fix. Lock down access point with tough Velcro. Glue antenna in place.
3. Better hardware. Way too much junk out there.
4. Repeater or range extender. Watch out for compatibility issues. Agreed.
5. Install a better antenna system on the wireless router end. Tradeoffs (loss of coverage in another area) may make that impractical.
6. Add a 2nd access point on a different channel. Piss off the neighbors.
7. Replace everything including clients with MIMO technology. High cost.
8. Add directional or better antennas on the client radios. Impractical on mobile computers.
9. Power amplifier (don't bother). Agreed.

Traffic amount most certainly does have an effect on interference. Double the traffic by repeating everything twice and you have twice the chance of a collision and 1/2 the available airtime during which to transmit. The speed adjustment mechanism actually makes things worse. When the access point detects errors and corrupted frames, it intentionally slows down the connection speed in order to improve the signal to noise ratio. As the speed goes slower and slower toward the

1Mbit/sec lower limit, the amount of air time necessary to send a single packet increased proportionally. The longer the air time, the less time another station has to send their traffic. I've had some fun with an old Teletronics 802.11 (1-2Mbits/sec only) PCMCIA card. When I start to move traffic, everything else just about stops because there's very little available air time left.

For intereference testing, I use streaming UDP audio from some station on Shoutcast. The traffic rate is well controlled and there are few ACKS/NACKS floating around. As I switch to higher and higher UDP bandwidth sources, the interference level (as measured by the sustained thruput) of a nearby access points drops rapidly. As I said, increased traffic means increased interference.

Incidentally, you can defeat the speed reduction system by fixing the speed of the access point. For high interference environments, I usually pick 6, 9, or 12Mbit/sec OFDM as well as turn on CTS/RTS flow control. The resultant thruput is nothing spectacular, but is far less susceptible to interference induced problems. Lowering the fragmentation threshold, in order to send smaller packets doesn't seem to help much so I just leave it alone.

I was going to ask you what you thought of my idea. Violent opposition usually implies that it's a fabulous idea. Perhaps I'll work on it some more.

I have a semi-working prototype. It's about 6" x 6" x 0.35" thick. It's a patch antenna made from foam board, aluminium foil elements, and a vacuum formed cover. In order to keep the thickness down, I have to live with about 8dBi gain instead of the 10dBi that a thicker antenna would produce. For additional gain, I was thinking of replacing the patch with 4ea phased folded dipoles in an array, but that will require an expensive circuit board.

There's a flange around the edge which allows mounting with Velcro, scotch tape, chewing gum, and threaded fasteners. I'm working on a spring loaded clip on mount that's designed for temporary attachment to the laptop lid. Surely the added gain and directionality will compensate for any aesthetic considerations.

I also have an inflatable version that works amazingly well.

[POSTED TO alt.internet.wireless - REPLY ON USENET PLEASE]

I don't see how that follows. The amount of traffic on a given channel shouldn't be a significant factor in the amount of interference.

No way would I put up with plastering a directional antenna onto my ThinkPad. :)

[POSTED TO alt.internet.wireless - REPLY ON USENET PLEASE]

With all due respect, that's just relative amount of sucking, which I don't think is in the same ballpark as consuming another primary channel -- in my experience, a repeater on the same primary channel is usually much less of an overall interference problem than hogging two of only three primary channels with two access points would be.

A caveat is that many consumers and access points are dumb enough to use only channel 6, which in my not so humble opinion is a big (if not the biggest) reason for WiFi dissatisfaction. (There's really no excuse for an access point not being smart enough to sniff signals and pick the clearest available primary channel by default, with of course the ability to override auto selection.) In this case adding a repeater will tend to make things worse than putting a separate access point on a different primary channel, although again that's relative amounts of sucking -- it would be better still to put both an access point and a repeater on the same different primary channel, thus eliminating all interference on channel 6, and minimizing consumption of the band.

I suspect there are cases where that might be an acceptable tradeoff, but not in most cases I can think of, in part because of the need to orient the computer in a particular direction. If and when I really need a better antenna, I'd rather have something as compact as possible that just plugs into my PC Card so that I can position it independently of my computer. But then I'd also have to get a good PC Card with an antenna connector that also works well without an external antenna. I just can't get excited about all that.

As an additional data point, I've done some careful testing of the Netgear WG511 against a Toshiba notebook with integrated 802.11g WiFi and an (omnidirectional) antenna in the lid, and haven't found significant differences.

Just centralise the router.

Put it in the centre of the house.