Long-range link question (long)

[Bubba]

Archived from groups: alt.internet.wireless (More info?)

I want to link two wired networks. The principal uses for the link
will be remotely monitoring security cameras at one location and
sharing a high-speed internet link that is only available at one site.

I need to have fairly decent security, but it doesn't have to pass DOD
standards.

I'm looking at the D-Link DWL-2100AP for use at both sites, mainly
because of the features it claims to offer.

The distance will be around 2 miles, and I will need about 40 foot
masts at both ends to clear some obstructions.

I plan on using 24 dB gain parabolic antennas, both for their gain, as
well as for the narrow beam-width (which I assume should help limit
"outside" contact with the link). I also plan on placing the AP's in
a weatherproof box on the antenna mast adjacent to the antenna and
using POE to get power to the units, and just a short pigtail between
the AP and antenna.

Does anyone know if there is a hack for the D-Link 2100 AP (for
increasing the power)?

According to the formulas I've been pointed to, the D-link units at
32mw should handle the link, even including consideration of fresnel
zones, but I want the links to be solid (even in rain if possible),
and I'd like enough signal strength to get better than "b" throughput
speeds since real-time monitoring of multiple cameras will take a bit
of bandwidth. I'd prefer to avoid the extra expense, complexity, and
likely illegality of adding amplifiers to these units.

One of the main reasons I'm looking at the DWL-21000 is the fact that
it handles WPA (or so they say), as well as the link type I want to
use. There is a remote possibility that I will eventually want to use
one of the sites as point-to-multipoint (with a different antenna),
although the other site will never be used for anything other than
point-to-point. I've never used D-link wireless equipment, but have
always had good luck with their wired products.

Are their any other units I should be considering before the D-link
units? Any brands/models anyone would recommend?

I figure if I eventually want to use vpn tunneling between the 2 sites
I will still need routers at each end that support this? The networks
at each end are a mish-mash of different versions of Windows,
including ME, W2K, and XP, and due to certain software requirements,
need to remain in their current bastardized configurations.

Any info is appreciated, and thanks for reading this tome.

Bob

 

[Guest]

Archived from groups: alt.internet.wireless (More info?)

On Wed, 07 Apr 2004 09:26:12 -0500, Bubba wrote:


> Any info is appreciated, and thanks for reading this tome.
>
> Bob

If this link is of any importance whatsoever, praytell, why in the world
are you even contemplating doing it with off-the-shelf consumer "stuff"?

vg

 

[Guest]

Archived from groups: alt.internet.wireless (More info?)

Bubba <vjdqlwlj02@sneakemail.com> wrote in
news:tc2870luptmhoaqujv0jj30lfre48fvljm@4ax.com:

> Does anyone know if there is a hack for the D-Link 2100 AP (for
> increasing the power)?

Not really a hack, but a WiFi Amplifier:

http://www.rflinx.com/

Not sure if these products are any good.


> Are their any other units I should be considering before the D-link
> units? Any brands/models anyone would recommend?

Netgate makes high powered WiFi equipment (200mw!):

http://www.netgate.com/

> I figure if I eventually want to use vpn tunneling between the 2 sites
> I will still need routers at each end that support this?

Yes, you'll need a router capable of VPN tunnels, or you can run a
software VPN tunnel in Windows 2000/XP.

Dlink does have a Wireless Router/AP + VPN built into one unit:

http://www.dlink.com/products/?pid=274


--
Lucas Tam (REMOVEnntp@rogers.com)
Please delete "REMOVE" from the e-mail address when replying.
http://members.ebay.com/aboutme/coolspot18/

 

[Bubba]

Archived from groups: alt.internet.wireless (More info?)

In one word, cost.

I'm entirely open to suggestions in regards to professional solutions.

In our little, isolated neck of the woods, the best I've been able to
find thus far in terms of pro equipment roughly follows:

Professional site survey -- minimum of $1,500 to tell me what I
already know.

Professional equipment AND install around $8,000 per site (multiply
this times 2). The "Pros" around here aren't even willing to use
customer installed masts and insist on installing their own preferred
version of tower.

Licensing fees (they aren't telling at this point, but if I want or
need anything above the unlicensed 32mw units, they tell me there will
be additional licensing fees (AND equipment costs). I can get 32 to
200 mw units on my own, sans licensing fee, and although of
questionable legality considering the antennas I will be using, I can
also get the amps for much less than the "pros" will charge for them.

Sooo, for something like 17.5K, plus any licensing fees, I should have
a functional link, but the "pros" responsibility stops at the cable
(or ethernet) entrance, and they don't even consider making the link
work with the networks, just guaranteeing the presence of a functional
link.

I've considered the "enterprise" level outdoor AP's that run around 1k
apiece, but the ones I've seen are "b" units" (and if possible, I want
more than "b" throughput), and the ones that I've seen don't yet offer
WPA, and I don't think I've even seen one with VPN tunneling. Still
need masts and antennas for these, and there is no guarantee that they
will work for my purposes.

The link is important, but NOT critical, and I've been a confirmed
(and generally very successful) tinkerer for most of my life. While
important to me, the link is not more important in terms of cost than
both of the networks that it will be connecting, which would be the
case using the professional solutions I've been able to find in my
region thus far.

Bob

On Wed, 07 Apr 2004 11:26:25 -0600, Valentín Guillén
<usenet1@myrealbox.com> wrote:

>On Wed, 07 Apr 2004 09:26:12 -0500, Bubba wrote:
>
>
>> Any info is appreciated, and thanks for reading this tome.
>>
>> Bob
>
>If this link is of any importance whatsoever, praytell, why in the world
>are you even contemplating doing it with off-the-shelf consumer "stuff"?
>
>vg

 

[Guest]

Archived from groups: alt.internet.wireless (More info?)

With the system you described you don't need any amplifiers or
anything else. I will explain as follows.

2 mile distance path loss equals about 110.4 dB
32 mW power equals about +15 dB
Antenna gain equals 24 dB
Cable loss with access point mounted at antenna equals no more then 2
dB

total equals something like this.

+15 -2 +24 -110.4 +24 -2 = -51.6 RSSI

With everything aligned properly you should receive a signal strength
of about -51.6 dBm at your receiver. Even poor consumer equipment can
function very well at this level. Cisco equipment has a receive spec
of about -86 for 11 mbits and - 95 for 1 mbit performance.




Bubba <vjdqlwlj02@sneakemail.com> wrote in message news:<tc2870luptmhoaqujv0jj30lfre48fvljm@4ax.com>...
> I want to link two wired networks. The principal uses for the link
> will be remotely monitoring security cameras at one location and
> sharing a high-speed internet link that is only available at one site.
>
> I need to have fairly decent security, but it doesn't have to pass DOD
> standards.
>
> I'm looking at the D-Link DWL-2100AP for use at both sites, mainly
> because of the features it claims to offer.
>
> The distance will be around 2 miles, and I will need about 40 foot
> masts at both ends to clear some obstructions.
>
> I plan on using 24 dB gain parabolic antennas, both for their gain, as
> well as for the narrow beam-width (which I assume should help limit
> "outside" contact with the link). I also plan on placing the AP's in
> a weatherproof box on the antenna mast adjacent to the antenna and
> using POE to get power to the units, and just a short pigtail between
> the AP and antenna.
>
> Does anyone know if there is a hack for the D-Link 2100 AP (for
> increasing the power)?
>
> According to the formulas I've been pointed to, the D-link units at
> 32mw should handle the link, even including consideration of fresnel
> zones, but I want the links to be solid (even in rain if possible),
> and I'd like enough signal strength to get better than "b" throughput
> speeds since real-time monitoring of multiple cameras will take a bit
> of bandwidth. I'd prefer to avoid the extra expense, complexity, and
> likely illegality of adding amplifiers to these units.
>
> One of the main reasons I'm looking at the DWL-21000 is the fact that
> it handles WPA (or so they say), as well as the link type I want to
> use. There is a remote possibility that I will eventually want to use
> one of the sites as point-to-multipoint (with a different antenna),
> although the other site will never be used for anything other than
> point-to-point. I've never used D-link wireless equipment, but have
> always had good luck with their wired products.
>
> Are their any other units I should be considering before the D-link
> units? Any brands/models anyone would recommend?
>
> I figure if I eventually want to use vpn tunneling between the 2 sites
> I will still need routers at each end that support this? The networks
> at each end are a mish-mash of different versions of Windows,
> including ME, W2K, and XP, and due to certain software requirements,
> need to remain in their current bastardized configurations.
>
> Any info is appreciated, and thanks for reading this tome.
>
> Bob

 

[Bubba]

Archived from groups: alt.internet.wireless (More info?)

So, in theory at any rate, I should be able to achieve "G" speeds with
the DWL-2100AP I'm looking at??

Per specs at D-Link's site, this AP is supposed to give 54mbps at - 66
dBm. Assuming I am able to achieve something close to the calculated
-51.6 dBm with my installation, should the additional 15 dBm handle
"average" rain fade, etc., or am I nuts trying for more than "B"
speeds on this link?

Thanks,



On 8 Apr 2004 09:48:44 -0700, kbloch2001@hotmail.com (K Bloch) wrote:

>With the system you described you don't need any amplifiers or
>anything else. I will explain as follows.
>
>2 mile distance path loss equals about 110.4 dB
>32 mW power equals about +15 dB
>Antenna gain equals 24 dB
>Cable loss with access point mounted at antenna equals no more then 2
>dB
>
>total equals something like this.
>
>+15 -2 +24 -110.4 +24 -2 = -51.6 RSSI
>
>With everything aligned properly you should receive a signal strength
>of about -51.6 dBm at your receiver. Even poor consumer equipment can
>function very well at this level. Cisco equipment has a receive spec
>of about -86 for 11 mbits and - 95 for 1 mbit performance.
>
>
>
>
>Bubba <vjdqlwlj02@sneakemail.com> wrote in message news:<tc2870luptmhoaqujv0jj30lfre48fvljm@4ax.com>...
>> I want to link two wired networks. The principal uses for the link
>> will be remotely monitoring security cameras at one location and
>> sharing a high-speed internet link that is only available at one site.
>>
>> I need to have fairly decent security, but it doesn't have to pass DOD
>> standards.
>>
>> I'm looking at the D-Link DWL-2100AP for use at both sites, mainly
>> because of the features it claims to offer.
>>
>> The distance will be around 2 miles, and I will need about 40 foot
>> masts at both ends to clear some obstructions.
>>
>> I plan on using 24 dB gain parabolic antennas, both for their gain, as
>> well as for the narrow beam-width (which I assume should help limit
>> "outside" contact with the link). I also plan on placing the AP's in
>> a weatherproof box on the antenna mast adjacent to the antenna and
>> using POE to get power to the units, and just a short pigtail between
>> the AP and antenna.
>>
>> Does anyone know if there is a hack for the D-Link 2100 AP (for
>> increasing the power)?
>>
>> According to the formulas I've been pointed to, the D-link units at
>> 32mw should handle the link, even including consideration of fresnel
>> zones, but I want the links to be solid (even in rain if possible),
>> and I'd like enough signal strength to get better than "b" throughput
>> speeds since real-time monitoring of multiple cameras will take a bit
>> of bandwidth. I'd prefer to avoid the extra expense, complexity, and
>> likely illegality of adding amplifiers to these units.
>>
>> One of the main reasons I'm looking at the DWL-21000 is the fact that
>> it handles WPA (or so they say), as well as the link type I want to
>> use. There is a remote possibility that I will eventually want to use
>> one of the sites as point-to-multipoint (with a different antenna),
>> although the other site will never be used for anything other than
>> point-to-point. I've never used D-link wireless equipment, but have
>> always had good luck with their wired products.
>>
>> Are their any other units I should be considering before the D-link
>> units? Any brands/models anyone would recommend?
>>
>> I figure if I eventually want to use vpn tunneling between the 2 sites
>> I will still need routers at each end that support this? The networks
>> at each end are a mish-mash of different versions of Windows,
>> including ME, W2K, and XP, and due to certain software requirements,
>> need to remain in their current bastardized configurations.
>>
>> Any info is appreciated, and thanks for reading this tome.
>>
>> Bob

 

[Guest]

Archived from groups: alt.internet.wireless (More info?)

On Thu, 08 Apr 2004 18:22:16 -0500, Bubba wrote:

> Per specs at D-Link's site, this AP is supposed to give 54mbps at - 66
> dBm. Assuming I am able to achieve something close to the calculated
> -51.6 dBm with my installation, should the additional 15 dBm handle
> "average" rain fade, etc., or am I nuts trying for more than "B"
> speeds on this link?
>
> Thanks,

With all due respect, I think you *still* don't get it!

Regardless of what's "doable in theory", if you're not willing to spend to
do it right, you'll regret spending the money you DO spend.

What about wind-loading factors; rain-fade; equipment aging and frequency
drifting; moisture infiltration into cables; and perhaps the bigest bugabo
of all -- signal inferference? If you're not using dedicated frequencies,
but instead are using the ISM band, you HAVE NO FREQENCY &
FREQENCY-INTERFERENCE PROTECTION!!

Now I should be perfectly clear here: I'm NOT saying you shouldn't do it,
nor am I saying it can't be done. What I'm saying is to eveluate with the
principles involved, exactly how important the link IS. If you're still
insistent that YOU can do it, hire a professional consultant to design the
system. You probably won't get a written guarantee of performance unless
they purchase and install!

The most likely outcome here is that you'll insist upon doing it, it will
underperform/not perform, and we'll see your posts here in a month with
pleas for assistance. Are you both an RF AND networking engineer? Leave
it to the professionals, unless this link is nothing more than a hobby and
a learning experience. In the latter case, have at it!

I regularly shoot WiFi 802.11x signals in excess of ten miles, but I'm
both an RF and networking engineer, and I know what I'm doing. So I'm not
saying that this is not doable. I'm saying to sit back with decision and
budget makers, and decide how vital this link is. Then proceed in the
appropriate manner.......

Best Regards,

vg

 

[Guest]

Archived from groups: alt.internet.wireless (More info?)

What he asked for is what else he may need. For 802.11b speeds the
distance/loss/antenna combo is more then adequate. Adding amplifiers
as others suggested usually causes more problems then its worth.
Unless the Amp is very very good it cannot switch fast enough from TX
to RX in the presence of noise to operate well.

Now a link speed of 11 mbits per second should be able to yield about
5 to 6 mbits of total throughput. For most home and business users
this is a more then adequate link speed. The questions is still what
is the link going to be used for. If it is to provide internet and
email access on a netowrk with a T1 to the internet it should be more
then adequate. If you are trying to use it for VOIP services it will
probably have trouble being that it is a half duplex media with a high
rate of packet retries (anything above 1/1000 is high in my opinion)

Valentín Guillén <usenet1@myrealbox.com> wrote in message news:<pan.2004.04.09.00.42.07.564828@myrealbox.com>...
> On Thu, 08 Apr 2004 18:22:16 -0500, Bubba wrote:
>
> > Per specs at D-Link's site, this AP is supposed to give 54mbps at - 66
> > dBm. Assuming I am able to achieve something close to the calculated
> > -51.6 dBm with my installation, should the additional 15 dBm handle
> > "average" rain fade, etc., or am I nuts trying for more than "B"
> > speeds on this link?
> >
> > Thanks,
>
> With all due respect, I think you *still* don't get it!
>
> Regardless of what's "doable in theory", if you're not willing to spend to
> do it right, you'll regret spending the money you DO spend.
>
> What about wind-loading factors; rain-fade; equipment aging and frequency
> drifting; moisture infiltration into cables; and perhaps the bigest bugabo
> of all -- signal inferference? If you're not using dedicated frequencies,
> but instead are using the ISM band, you HAVE NO FREQENCY &
> FREQENCY-INTERFERENCE PROTECTION!!
>
> Now I should be perfectly clear here: I'm NOT saying you shouldn't do it,
> nor am I saying it can't be done. What I'm saying is to eveluate with the
> principles involved, exactly how important the link IS. If you're still
> insistent that YOU can do it, hire a professional consultant to design the
> system. You probably won't get a written guarantee of performance unless
> they purchase and install!
>
> The most likely outcome here is that you'll insist upon doing it, it will
> underperform/not perform, and we'll see your posts here in a month with
> pleas for assistance. Are you both an RF AND networking engineer? Leave
> it to the professionals, unless this link is nothing more than a hobby and
> a learning experience. In the latter case, have at it!
>
> I regularly shoot WiFi 802.11x signals in excess of ten miles, but I'm
> both an RF and networking engineer, and I know what I'm doing. So I'm not
> saying that this is not doable. I'm saying to sit back with decision and
> budget makers, and decide how vital this link is. Then proceed in the
> appropriate manner.......
>
> Best Regards,
>
> vg

 

[Guest]

Archived from groups: alt.internet.wireless (More info?)

Fresnel Fadermargini wrote:

> Rainfade is not bad at 2.4 either since it is such a large wavelength.. it
> is when you exceed
> 10Ghz that the water molecules exceed the size of the wavelength.
>

The wavelength at 10 GHZ is 3 cm (1.2"). I haven't seen many water
molecules that big.

--

Fundamentalism is fundamentally wrong.

To reply to this message, replace everything to the left of "@" with
james.knott.

 

[Guest]

Archived from groups: alt.internet.wireless (More info?)

Mistake!!! Should not have been water molecule... Water Drop
Hence why you can run UNII/ISM 5-5.8Ghz links 6Ghz Licensed links
for huge distances, but when you get to 10Ghz or 23Ghz or 38Ghz microwave,
you have to blast the power and still have limited range...

-Guy

"James Knott" <bit_bucket@rogers.com> wrote in message
news:GJ8fc.1820$2oI1.1052@twister01.bloor.is.net.cable.rogers.com...
> Fresnel Fadermargini wrote:
>
> > Rainfade is not bad at 2.4 either since it is such a large wavelength..
it
> > is when you exceed
> > 10Ghz that the water molecules exceed the size of the wavelength.
> >
>
> The wavelength at 10 GHZ is 3 cm (1.2"). I haven't seen many water
> molecules that big.
>
> --
>
> Fundamentalism is fundamentally wrong.
>
> To reply to this message, replace everything to the left of "@" with
> james.knott.
>

 

[Guest]

Archived from groups: alt.internet.wireless (More info?)

Fresnel Fadermargini wrote:

> Mistake!!! Should not have been water molecule... Water Drop
> Hence why you can run UNII/ISM 5-5.8Ghz links 6Ghz Licensed links
> for huge distances, but when you get to 10Ghz or 23Ghz or 38Ghz microwave,
> you have to blast the power and still have limited range...
>

Even still, it's the humidity level, not drop size that causes the problems.
For example, at those frequencies, fog will have a significant effect, even
though the "drops" are still much smaller than a wavelength.

--

Fundamentalism is fundamentally wrong.

To reply to this message, replace everything to the left of "@" with
james.knott.

 

[Guest]

Archived from groups: alt.internet.wireless (More info?)

You are referring to the climate factor! That is a totally different
thing...
You are correct where the humidity level will effect the availability of the
Link.
This can be compensated with additional db's.. If this weren't the case,
then all of Verizon's Cell Phone Backhaul links in South Florida would not
work..
They use Lynx Radios at 2.4Ghz in ranges from 1 to 30 miles... and a climate
factor
of .5



"James Knott" <bit_bucket@rogers.com> wrote in message
news:a3vgc.112126$2oI1.53036@twister01.bloor.is.net.cable.rogers.com...
> Fresnel Fadermargini wrote:
>
> > Mistake!!! Should not have been water molecule... Water Drop
> > Hence why you can run UNII/ISM 5-5.8Ghz links 6Ghz Licensed links
> > for huge distances, but when you get to 10Ghz or 23Ghz or 38Ghz
microwave,
> > you have to blast the power and still have limited range...
> >
>
> Even still, it's the humidity level, not drop size that causes the
problems.
> For example, at those frequencies, fog will have a significant effect,
even
> though the "drops" are still much smaller than a wavelength.
>
> --
>
> Fundamentalism is fundamentally wrong.
>
> To reply to this message, replace everything to the left of "@" with
> james.knott.
>