Multhops using BH modules

We have to extend data connection upto 120 KM distance using BH modules. This means we would have to implement two hops at approximate distances of 40 KM. We dont want to use CMM.
How should we design our network? What are possible options? How many slave,master modules would be used? Which master module should be configured to provide sync signal??Is it possible for slave BH module to get sync signal from Master modules and pass on to other modules in network??
Please advice…

Junaid

A 900MHz AP & SM could make this in a single hop, but you’d need your antennas on opposite sides on a shallow valley. Earth curvature and the Fresnel zone conspire to make this a difficult hop.

Two hops may be possible using the 45Mbps back hauls, but this pushes their limits. Antenna height may again be an issue; you’ll also likely need a CMM in the center to synchronize the two links, unless the unit’s DFS eliminates the need for sync.

2.4 and 5.7GHz back hauls are limited to about 56km so you’ll likely need three hops. You may be able to push the 2.4 further using the connectorized version and high-gain antennas. Three hops, using the same frequency band, is problematic – even with CMMs.

Your best option may be two or three hops using alternating frequency bands; e.g., 5.7–2.4–5.7. If, however, any segment needs to coexist with other Canopy equipment, you may still need the CMMs.

Muliple-hop links, using the same frequency band, might be possible using CMMs if the back hauls had a synchronization option specifying transmit or receive on the sync pulse. Hopefully, Canopy Support could comment on this possiblity.

We have one customer that required connectivity over only 32 km - very mountainous area. We ened up using 5.7 BH’s making up 6 links. We put a CMM in between the first and second link and just daisy chained the timing all along the network. With the non-overlapping frequencies there are no problems creating a frequency plan to accomodate this - only two links into each site is easy.

Aaron

My understanding is that daisy-chained timing – connecting the timing port output of a BH slave or SM to the input of a BH master or AP – does not accurately synchronize the two units. I haven’t personally confirmed this with a dual-trace oscilloscope but, apparently, there’s a small delay between the start of the transmit sequences.

I don’t know if eliminating the timing completely – at least the daisy-chained timing – would make the units less reliable or the same. I also don’t understand why Motorola suggests daisy-chained timing for remote APs when the AP is in a different frequency band than the associated SM.

Acherman,

Is your aggregate link 32 or 320 km?

There is a minor delay in the timing from link to link. This is why Motorola doesn’t “support” daisy chaining over more than one link, because the timing can be off.

I see it like this - if the timing is off 10 us (just a number) on one link, then the next one will be off 10 us. Each subsequent link will be off 10 us from the previous. I’s rather have two collocated radios off by 10 us every cycle than increase the chances of them desensing each other all the time. Worst case - jitter climbs a little. :?: Does that make sense? :lol:

Motorola suggests the daisy chained timing even if the Sm and AP are in diferent bands because they can still desense each other if they are close enough. A stronge enough signal will desense another receiver in a different band.

On reviewing my documentation, our customer’s total link is only about 25 km, with longest shot at just under 11 km.

Aaron