ANT Forum

I'd like to know if there is possibility to create such a network:
I have 80 stations on each 3 sensors and one Master which collect the data from the stations.

If I configure each staion with different Network Number(I'll have 80 Network Numbers), can I collect thier data by the singel Master?

In other words is one master can establish connection with 80 different networks?

For mor info pls see the attached pic. [img size=616]http://www.thisisant.com/images/fbfiles/images/struct-d5527e06095291edae468e373226bb25.PNG[/img]      

The topology you speak of here is quite doable with ANT, although depending on your use case this could be quite a bit simpler.

I would recommend reading through at least the opening sections of the "ANT Message Protocol and Usage" document. As in ANT each sensor is a "master" it means that any number of listeners or "slaves" can be listening to the same sensor at the same time.

For example, to enable something close to this use case, AP2 can store up to 3 unique "network keys", but it would be simpler to assign each sensor a different device ID. Each "unit" could then listen to 3 sensors each, and then you could have every station broadcast where the "master" could then open continuous scanning to accept all of the messages from all of the "units". The advantage of this topology I can see is extending wireless range.

But, if each of your sensors did not exceed say 0.5 Hz, a single AP2 is capable of 190Hz of total messaging rate, and 300Hz of over the air coexistence, meaning you would only need one "master" to listen to all 240 sensors simultaneously on the same RF channel.

To tailor a network topology best for your use case it would be very useful to know your sensor's required channel period (ANT sends 8 bytes per message period), distance between "master" to "sensor" and power requirements: do your sensors, units, and "master" run off battery or mains respectively?      

I think I didn't ask the question right.

The sensors are ant sensors but I cann't change it's code because I bought something ready like cadence sensor.

The distance between the master and the units is up to 20m.But it's not the problem.

My problem is the distance between one unit to other is very close - 0.5-1 meter.

I'm afraid of reading my neighboring station's sensor (instead of reading the sensor located at our station).

What guarantee that won't be misreading of the sensors' data.How you can I avoid reading the neighboring station's sensor (instead of reading the sensor located at my station) ?



Yaniv      

There are at least a couple of ways you can ensure this is not a problem.

In this case I would recommend implementing some form of proximity pairing mode on each station, so that at each station only sensors which were very close to it are paired to it before use. Both AP2 and C7 offer proximity search which gives "search bins" so that the channel will only pair to a nearby device.

However, at 0.5m even the closest proximity bin may be too large depending on the RF performance of your station (bins are based on relative signal strength). One solution would be to use C7's or multi-mode comm chips which offer high resolution RSSI. This would allow your station to do a background scan, and then open channels only with sensors that were within your own custom tuned RSSI distance threshold.

Once you know the device ID of the sensor, you can use inclusion lists to ensure the station only pairs with those specific sensors if the channels need to be re-established without going through proximity pairing again.

We offer some additional information on device pairing in the AN02 App Note found here.

Cheers      

hfchin wrote:

There are at least a couple of ways you can ensure this is not a problem.

In this case I would recommend implementing some form of proximity pairing mode on each station...

However, at 0.5m even the closest proximity bin may be too large depending on the RF performance of your station (bins are based on relative signal strength).

Could you tell me if there is any estimation for the distance of the closest proximity bin or it's mather of a trial and error deppend of RSSI?      

From my experience an AP2 module with direct line of sight to another AP2 module on USB interface boards using Bin 1 (the smallest radius bin) will pair at ~1m give or take ~10%. This is probably pretty close to a best case scenario for RF conditions. Any changes to RF design (enclosures, antenna design, RF interference, etc) will affect what your end product will see.      

Good, now I can determine the distance between stations.

Another question:
I'd like to know if there is any problem to use proximity search
in order to connect 3 sensors in the threshold area.

The aim of this is a connection btween 3 sensors which are in the threshold area to master.

Is it possiable ?

Thanks      

This depends on the power requirements of your station and/or cost considerations.

The most straightforward and least power intensive method would be to use an 8 channel ANT radio and separately pair each sensor to a single channel but other methods would also work (such as continuous scanning + RSSI).

Cheers