GPS clock (synchronization)

we going to install two point to point radio links (cnReach N500 900Mhz unlicensed) without external GPS clock,
topology:
PTP Link#1 SiteA<>SiteB (cnReach N500 900Mhz unlicensed)
PTP Link#2 SiteA<>SiteC (cnReach N500 900Mhz unlicensed)

do we have to install GPS clock timing equipment to provide an external synchronization to the radio? or those radios have an internal clock

Another questions: is the external clock will help for the frequency scan? and what the point of using the external clock if the radio work as a point-to-point link?

Hi Edwar, for this scenario GPS sync isn’t really necessary. Just make sure each link has a unique network ID and set the hop pattern to 0 (for pseudo-random). The advantage of sync is primarily in licensed band operation where you want to have adjacent masters transmitting at the same time and receiving at the same time.

In theory you can sync an ISM network so that each hop sequence starts at the same time. In that case you would set the hop pattern to say 4 (radio hops to every 4th channel) and then set the hop offset to 2 on the second master so the radios never step on each other. You can set cnReach radios to generate the sync pulse and cable it to the 2nd master. Or you can sync an entire network using cnPulse as a sync source.

But in the end i wouldn’t recommend sync on this simple setup as it will keep things simple and straight-forward.

Regards, Bruce

The internal clock is a free-run cycle. It wll work but has limitations.

You have 28mhz in the 900mhz band. If you need to have more than 6mhz channel then use gps timing. There is a lot of things to consider with these systems.

Word of caution, put both APs at site A with gps. This will save you a ton of headaches. If the look angle is small enough, use a sector antenna and only one AP then you wont need gps until you add another AP.

The last question is simple, the point of a timed network is to ensure all APs in a network transmit at the same time. Thus reducing self-interference between towers and client side. With this, you could time the middle leg of a two hop link and reuse the same frequency and only suffer minor losses from this. If this was un-timed then you would need two clear channels with enough guard band in between.

Looks like we posted together.

Thank you for answer my questions,

I was plan to limiting the frequency band range or split the frequency band between the two PTP radios
what do you think about splitting the frequency band?

PTP Link#1 SiteA<>SiteB (cnReach N500 900Mhz unlicensed) Band Start/End (902/914 Mhz)
PTP Link#2 SiteA<>SiteC (cnReach N500 900Mhz unlicensed) Band Start/End (916/928 Mhz)

Band splitting is fine but you loose just over half the throughput due to how these radios do encoding. Again like always your mileage will vary, but from having a couple hundred canopy 900 radios, I can tell you that I wished for a 12 or even a 20Mhz channel to use over band splitting.

Another consideration is you must ensure that you have a map that you accurately trace the radio paths on to ensure future installations (think separate systems of the same equipment not the same client) have a chance to not interfere with this system. The cnreach guide is a good reference for this and should be followed.

Thank you and appreciated for your replay

With these frequency hopping radios you won’t lose bandwidth. Instead you have fewer channels for each radio to hop on and occasionally they’ll hop on the same channel at the same time but statistically this is pretty rare with the pseudo-random hopping pattern.

So splitting can help but my guess is you won’t see much difference in performance.

The main reason these radios have a way to split or block certain channels is in case you are co-locating with a 900 MHz radio that sits on a single channel (like a PMP 450 or old Canopy FSK). Then you can block those channels and the cnReach radio will never hop on the channel in fixed use.