Community discussions

MikroTik App
 
huntah
Member Candidate
Member Candidate
Topic Author
Posts: 287
Joined: Tue Sep 09, 2008 3:24 pm

1km link no Line Of Sight

Mon Aug 01, 2011 9:00 pm

Hello,

I am atempting to link two sites via Wireless (2x RB433 with R52Hn and 19dBi Cyberbajt Giga Eter antenas).
The problem is I do not have clear line of sight (LoS) because of the steep descend in front of me (coastal area).

The link is working so/so. I only get -80 to -85 dBm and hence the link is dropping.
I know I should get in -65 dBm area but is this even achievable with better antennas and no clear LoS?

Would it be better to use 5GHz 23 DBi antenas with only 10 degree angle as opposed to 2,4GHz same degree angle.
The current antennas have 20 degree angle.

The current configuration is:
Bridge -> Pseudo Station Bridge and Nstreme Enabled
The frequency is 2437 and the frequence usage is almost zero (no other Wireless signals)

Any suggestions?
You do not have the required permissions to view the files attached to this post.
 
n21roadie
Forum Guru
Forum Guru
Posts: 1949
Joined: Fri Aug 07, 2009 10:36 pm
Location: Limerick,Ireland

Re: 1km link no Line Of Sight

Mon Aug 01, 2011 10:00 pm

If you don't have line of sight then consider picking a location which has LOS to point A (internet Wlan) and point B (local Ap) as a relay or pick a new location for the local AP on high ground which is illustrated on your drawing, extra height is always good for a AP.

The link is working so/so. I only get -80 to -85 dBm and hence the link is dropping.
I know I should get in -65 dBm area but is this even achievable with better antennas and no clear LoS?
No antenna is going to help with no LOS all it will do is reduce the amount of time the link drops?
 
User avatar
MCT
Member Candidate
Member Candidate
Posts: 158
Joined: Wed Mar 03, 2010 5:53 pm

Re: 1km link no Line Of Sight

Mon Aug 01, 2011 11:15 pm

+1 to above

You need a relay that both points can see. I don't know if it's feasible for you to get the points higher up.

900mhz is more tolerant to that type of setup.

The higher frequency you use the more it will require a clear line of sight so going to 5Ghz will just make it worse.
 
InoX
Forum Guru
Forum Guru
Posts: 1966
Joined: Tue Jan 09, 2007 6:44 pm

Re: 1km link no Line Of Sight

Tue Aug 02, 2011 1:32 am

2x2 mimo should help a little...maybe 2x2 and 2.4Ghz.
 
huntah
Member Candidate
Member Candidate
Topic Author
Posts: 287
Joined: Tue Sep 09, 2008 3:24 pm

Re: 1km link no Line Of Sight

Wed Aug 03, 2011 12:04 am

thanx for your replies, even though it is not what I wanted to hear :)

2x2 mimo --> You mean 2x 19dBi Antena on each side (HT TX in HT RX Chains 0 and 1?) - one for TX and one for RX..
Or do I need special MIMO Antenas? Which yould you suggest.

As answered above better results will be with lower frequencies (2.4 GHz better than 5Ghz)
 
n21roadie
Forum Guru
Forum Guru
Posts: 1949
Joined: Fri Aug 07, 2009 10:36 pm
Location: Limerick,Ireland

Re: 1km link no Line Of Sight

Wed Aug 03, 2011 5:13 pm

thanx for your replies, even though it is not what I wanted to hear :)

2x2 mimo --> You mean 2x 19dBi Antena on each side (HT TX in HT RX Chains 0 and 1?) - one for TX and one for RX..
Or do I need special MIMO Antenas? Which yould you suggest.

As answered above better results will be with lower frequencies (2.4 GHz better than 5Ghz)
I would before you purchase any addition equipment use a program like ligowave http://www.ligowave.com/linkcalc/

I used this program for user http://forum.mikrotik.com/viewtopic.php?f=2&t=53922

Put in your coordinates, antenna height, TX power, etc. and post back the results then we would be in a better position to suggest solutions, as it stands you have a drawing which is really not detailed enough and you could discover a lot of time wasted and money spent on the wrong type of equipment to link to the target AP.
 
huntah
Member Candidate
Member Candidate
Topic Author
Posts: 287
Joined: Tue Sep 09, 2008 3:24 pm

Re: 1km link no Line Of Sight

Fri Aug 05, 2011 12:42 am

Ok the attached picture is from Ligo Calc..

I am not excalty sure what RX-Tresholds means.. I put -80 dBm is this OK?
Noise floor is over -100 dBm (i think 105)..

On the bottom (coast) there is also another Wlan card witch 18dBi Gain Omni antenna

Should I use NV2 protocol?
Would the results be better using 802.11b rather than 802.11g?
You do not have the required permissions to view the files attached to this post.
 
User avatar
mramos
Member Candidate
Member Candidate
Posts: 231
Joined: Sun Nov 23, 2008 1:05 am
Location: S. B do Campo - SP - Brazil

Re: 1km link no Line Of Sight

Fri Aug 05, 2011 6:00 am

Hi ...

A cheap & fun test is a "passive refletor", if you have free access to the middle of the path.

A pair of 25dBi grids, back to back by a short piece of cable at a small mast may be do the job.

I did't took a look at the numbers but calculation goes more or less like this:

AP = 23dBm
AP Antenna gain = 25dBi
500m free space loss (first leg) = 95dB
Passive 1 ant gain = 25dBi
Power at Passive 1 = -21dBm
Cable loss = 3dB
Passive 2 ant gain = 25dBi
Power delivered by passive 1 = -21-3+25 = 1dBm
500m free space loss (second leg) = 95dB
Power at client side = -94dBm
Client Antenna gain = 25dBi
Power at client = 25-94 = -69dBm

Re-run this numbers at some trusteable tool and play with frequencies & avialable antennas.

I never tried this on 2.4 / 5.8 but at 8, 18 and 23GHz. We used this turn around where one side of the ptp link was blocked by some building that was - maximum - 600 meter away. And where there was not antenna size limitations (e.g. a pair of 1.2m dishes at a 8GHz link ends, a pair of 1.8 or 2.4m passive, ~8Km between end points, passive at 500m from one side). Those links was 100M (ETH) or 155M (STM1). Connection between passive antennas was coax cable jumpers at 8GHz and waveguides jumpers above 15GHz.

In this cases, one lag at vertical, another at horizontal since its a co-channel scenario so better to improve rejection.

Regards;
 
n21roadie
Forum Guru
Forum Guru
Posts: 1949
Joined: Fri Aug 07, 2009 10:36 pm
Location: Limerick,Ireland

Re: 1km link no Line Of Sight

Sat Aug 06, 2011 4:46 pm

The Ligowave terrain picture says it all, bearing in mind this is ground level terrain no allowance for houses or trees which could further reduce LOS, in short you will not have a working link 24/7 for a AP maybe after a lot of time spent on trying different antenna's + frequencies enough throughput for a single cpe but not for a AP ( if u do please let us know?), you need a relay or move your AP to higher ground, forget about wireless protocols NV2, 802.11 get LOS for the AP first the rest will follow afterwards.
 
User avatar
saintofinternet
Forum Veteran
Forum Veteran
Posts: 768
Joined: Thu Oct 15, 2009 3:52 am

Re: 1km link no Line Of Sight

Sun Apr 13, 2014 5:09 am

Hi ...

A cheap & fun test is a "passive refletor", if you have free access to the middle of the path.

A pair of 25dBi grids, back to back by a short piece of cable at a small mast may be do the job.

I did't took a look at the numbers but calculation goes more or less like this:

AP = 23dBm
AP Antenna gain = 25dBi
500m free space loss (first leg) = 95dB
Passive 1 ant gain = 25dBi
Power at Passive 1 = -21dBm
Cable loss = 3dB
Passive 2 ant gain = 25dBi
Power delivered by passive 1 = -21-3+25 = 1dBm
500m free space loss (second leg) = 95dB
Power at client side = -94dBm
Client Antenna gain = 25dBi
Power at client = 25-94 = -69dBm

Re-run this numbers at some trusteable tool and play with frequencies & avialable antennas.

I never tried this on 2.4 / 5.8 but at 8, 18 and 23GHz. We used this turn around where one side of the ptp link was blocked by some building that was - maximum - 600 meter away. And where there was not antenna size limitations (e.g. a pair of 1.2m dishes at a 8GHz link ends, a pair of 1.8 or 2.4m passive, ~8Km between end points, passive at 500m from one side). Those links was 100M (ETH) or 155M (STM1). Connection between passive antennas was coax cable jumpers at 8GHz and waveguides jumpers above 15GHz.

In this cases, one lag at vertical, another at horizontal since its a co-channel scenario so better to improve rejection.

Regards;

hi,

your passive reflector theory is great.... but this is work practically?? was really interested.... :)
 
n21roadie
Forum Guru
Forum Guru
Posts: 1949
Joined: Fri Aug 07, 2009 10:36 pm
Location: Limerick,Ireland

Re: 1km link no Line Of Sight

Sun Apr 13, 2014 4:36 pm

Hi ...

A cheap & fun test is a "passive refletor", if you have free access to the middle of the path.

A pair of 25dBi grids, back to back by a short piece of cable at a small mast may be do the job.

I did't took a look at the numbers but calculation goes more or less like this:

AP = 23dBm
AP Antenna gain = 25dBi
500m free space loss (first leg) = 95dB
Passive 1 ant gain = 25dBi
Power at Passive 1 = -21dBm
Cable loss = 3dB
Passive 2 ant gain = 25dBi
Power delivered by passive 1 = -21-3+25 = 1dBm
500m free space loss (second leg) = 95dB
Power at client side = -94dBm
Client Antenna gain = 25dBi
Power at client = 25-94 = -69dBm

Re-run this numbers at some trusteable tool and play with frequencies & avialable antennas.

I never tried this on 2.4 / 5.8 but at 8, 18 and 23GHz. We used this turn around where one side of the ptp link was blocked by some building that was - maximum - 600 meter away. And where there was not antenna size limitations (e.g. a pair of 1.2m dishes at a 8GHz link ends, a pair of 1.8 or 2.4m passive, ~8Km between end points, passive at 500m from one side). Those links was 100M (ETH) or 155M (STM1). Connection between passive antennas was coax cable jumpers at 8GHz and waveguides jumpers above 15GHz.

In this cases, one lag at vertical, another at horizontal since its a co-channel scenario so better to improve rejection.

Regards;

hi,

your passive reflector theory is great.... but this is work practically?? was really interested.... :)
If you going to setup a repeater, is not more practical to use a standard setup for a repeater, rather than passive?
 
plisken
Forum Guru
Forum Guru
Posts: 2509
Joined: Sun May 15, 2011 12:24 am
Location: Belgium
Contact:

Re: 1km link no Line Of Sight

Sun Apr 13, 2014 8:26 pm

With no perfect line of sight use circulair polariserd antennas but the link shal not be stable.
Test the traject with linktest and looks what signal you have
 
n21roadie
Forum Guru
Forum Guru
Posts: 1949
Joined: Fri Aug 07, 2009 10:36 pm
Location: Limerick,Ireland

Re: 1km link no Line Of Sight

Sun Apr 13, 2014 11:28 pm

With no perfect line of sight use circulair polariserd antennas but the link shal not be stable.
Test the traject with linktest and looks what signal you have
From viewing the link profile graph - no antenna will provide a reliable connection (I wish there was?)
 
WirelessRudy
Forum Guru
Forum Guru
Posts: 3119
Joined: Tue Aug 08, 2006 5:54 pm
Location: Spain

Re: 1km link no Line Of Sight

Mon Apr 14, 2014 1:25 am

Hi ...

A cheap & fun test is a "passive refletor", if you have free access to the middle of the path.

A pair of 25dBi grids, back to back by a short piece of cable at a small mast may be do the job.

I did't took a look at the numbers but calculation goes more or less like this:

AP = 23dBm
AP Antenna gain = 25dBi
500m free space loss (first leg) = 95dB
Passive 1 ant gain = 25dBi
Power at Passive 1 = -21dBm
Cable loss = 3dB
Passive 2 ant gain = 25dBi
Power delivered by passive 1 = -21-3+25 = 1dBm
500m free space loss (second leg) = 95dB
Power at client side = -94dBm
Client Antenna gain = 25dBi
Power at client = 25-94 = -69dBm

Re-run this numbers at some trusteable tool and play with frequencies & avialable antennas.

I never tried this on 2.4 / 5.8 but at 8, 18 and 23GHz. We used this turn around where one side of the ptp link was blocked by some building that was - maximum - 600 meter away. And where there was not antenna size limitations (e.g. a pair of 1.2m dishes at a 8GHz link ends, a pair of 1.8 or 2.4m passive, ~8Km between end points, passive at 500m from one side). Those links was 100M (ETH) or 155M (STM1). Connection between passive antennas was coax cable jumpers at 8GHz and waveguides jumpers above 15GHz.

In this cases, one lag at vertical, another at horizontal since its a co-channel scenario so better to improve rejection.

Regards;

hi,

your passive reflector theory is great.... but this is work practically?? was really interested.... :)
http://en.jirous.com/passive-relay/
 
WirelessRudy
Forum Guru
Forum Guru
Posts: 3119
Joined: Tue Aug 08, 2006 5:54 pm
Location: Spain

Re: 1km link no Line Of Sight

Mon Apr 14, 2014 1:26 am

Sorry; http://en.jirous.com/passive-relay/

Very interesting. Has anybody ever tried something like it?
 
n21roadie
Forum Guru
Forum Guru
Posts: 1949
Joined: Fri Aug 07, 2009 10:36 pm
Location: Limerick,Ireland

Re: 1km link no Line Of Sight

Mon Apr 14, 2014 11:52 am

Sorry; http://en.jirous.com/passive-relay/

Very interesting. Has anybody ever tried something like it?
Not sure about the use as PTP but maybe for signal to a few customers
 
WirelessRudy
Forum Guru
Forum Guru
Posts: 3119
Joined: Tue Aug 08, 2006 5:54 pm
Location: Spain

Re: 1km link no Line Of Sight

Tue Apr 15, 2014 2:09 am

Sorry; http://en.jirous.com/passive-relay/

Very interesting. Has anybody ever tried something like it?
Not sure about the use as PTP but maybe for signal to a few customers
Well, why not? It just depends on the signal strengths at both end points. If the signals are good enough, conn. rates are high and you can pump data over it. Since in the middle the 2 antennas are just back to back inter connected with coax so theoretically you have no latency added.

I have been thinking of it to reach a remote hotel in a furthermore inhabited area in a deep valley surrounded by mountain ridges. From my central I can see one of the ridges (that could be reached with a 4-wheel drive through a long journey) and from that hotel you'll see the same ridge. Nowhere near electricity to be found and solar panels with batteries have to be guarded against theft and guard against very high winds. For just a hotel it would make it too expensive for me.... But two antenna's on a big pole cemented into the rock would not make an interesting product to steel and can be made wind resisting while once it works no further maintenance would be needed. I made a link budget calculation some 2 years ago and with the 27dBi radio we had I could achieve a 60dBm signal level at both ends. I presume now with the high power cards that can be done better. It would be enough to put a 40-50Mb link running over it. I could than even serve some weekend bungalows in the same area through a small AP.

It never happened though, it was merely to give an alternative option to the hotel with higher bandwidth than what copper lines brought them. Due the crisis and the financial situation due its remote location the hotel changed hands already 4 to 5 times anyway and recently I even heard its completely abandoned.... So I'm glad I never invested any in that project.....
 
n21roadie
Forum Guru
Forum Guru
Posts: 1949
Joined: Fri Aug 07, 2009 10:36 pm
Location: Limerick,Ireland

Re: 1km link no Line Of Sight

Tue Apr 15, 2014 11:37 am

Well, why not? It just depends on the signal strengths at both end points. If the signals are good enough, conn. rates are high and you can pump data over it. Since in the middle the 2 antennas are just back to back inter connected with coax so theoretically you have no latency added.
...............................
Must add this to my list of projects ?

Who is online

Users browsing this forum: lmeira, popecix and 29 guests