wireless network using Zigbee

[Kabanga]

Hello PICLIST familly

I'd like to design a wireless sensor network (20 sensors) using PIC microcontrollers.

Which PICs do you think is better for that purpose and which wireless technology (Zigbee. WiMi ?)

Is there any link or examples? Thanks for your help!

Best regards
Kabanga


       
Jetzt neu! Schützen Sie Ihren PC mit McAfee und WEB.DE. 30 Tage
kostenlos testen. *http://www.pc-sicherheit.web.de/startseite/?mc=022220* [http://www.pc-sicherheit.web.de/startseite/?mc=022220]
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Timothy J. Weber-2]

mamouda@... wrote:
>
> I'd like to design a wireless sensor network (20 sensors) using PIC microcontrollers.
>
> Which PICs do you think is better for that purpose and which wireless technology (Zigbee. WiMi ?)

I have used XBees for several devices, but would like to move over to
PICs with a Microchip stack at some point to reduce per-unit cost.

My feeling at this point is that ZigBee has a lot of nice features, but
if you don't need them, the fact that it's a standard is NOT important,
because no two ZigBee stacks interoperate yet (as I understand it).  So
it's really no more "standard," at this point, than Microchip's MiWi.

So - my decision tree goes like this:

A. Is short development time far more important than per-unit cost?  Or,
are you looking to sell the product commercially?  Go with XBee.  (FCC
approval is a major cost otherwise.)  There are some other similar
modules that are no worse, but as far as I can tell, not particularly
better either.

B. If per-unit cost is important, and FCC approval isn't (or you can
afford it), try Microchip's ZigBee solution.  Their stack is free, but
you still have to join the "ZigBee Alliance" ($K).

C. Or if you are OK with the subset included in MiWi (<= 1024 nodes, <=
8 coordinators), use that.

So far, for home use, I have done (A) but lean toward (C) next.

For small commercial products, I would go with (A) since the FCC
certification process scares me.

Hope that's of some use.
--
Timothy J. Weber
http://timothyweber.org
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Marc Nicholas-2]

Can you build a Zigbee stack, microcontroller, RFN PCB for less than
$21? The new XBees cost that much in qty 1. Quite compelling!

-marc



On 5/7/08, Timothy J. Weber <tw@...> wrote:

> mamouda@... wrote:
> >
> > I'd like to design a wireless sensor network (20 sensors) using PIC
> microcontrollers.
> >
> > Which PICs do you think is better for that purpose and which wireless
> technology (Zigbee. WiMi ?)
>
> I have used XBees for several devices, but would like to move over to
> PICs with a Microchip stack at some point to reduce per-unit cost.
>
> My feeling at this point is that ZigBee has a lot of nice features, but
> if you don't need them, the fact that it's a standard is NOT important,
> because no two ZigBee stacks interoperate yet (as I understand it).  So
> it's really no more "standard," at this point, than Microchip's MiWi.
>
> So - my decision tree goes like this:
>
> A. Is short development time far more important than per-unit cost?  Or,
> are you looking to sell the product commercially?  Go with XBee.  (FCC
> approval is a major cost otherwise.)  There are some other similar
> modules that are no worse, but as far as I can tell, not particularly
> better either.
>
> B. If per-unit cost is important, and FCC approval isn't (or you can
> afford it), try Microchip's ZigBee solution.  Their stack is free, but
> you still have to join the "ZigBee Alliance" ($K).
>
> C. Or if you are OK with the subset included in MiWi (<= 1024 nodes, <=
> 8 coordinators), use that.
>
> So far, for home use, I have done (A) but lean toward (C) next.
>
> For small commercial products, I would go with (A) since the FCC
> certification process scares me.
>
> Hope that's of some use.
> --
> Timothy J. Weber
> http://timothyweber.org
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>

--
Sent from Gmail for mobile | mobile.google.com
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Ruben Jönsson]

My understanding is also that ZigBee is not quite ready to go industrial yet
and it has quite a lot of overhead for just a simple wireless network. But you
may also take a look at IEEE 802.15.4 radios. This has a lot of the features
provided by ZigBee (since ZigBee is built upon IEEE 802.15.4) and looks like it
could be very usefull for a small wireless sensor network.

There are also other stacks than ZigBee built on the PHY and MAC IEEE 802.15.4
layers (WiMi?) that could be worth checking out.

Things to consider when deciding a technology and standard for wireless
devices:

Range
Power requirement (battery operated or mains power?)
Data rate
Point to point, point to multi point or multi point to multi point
Number of nodes in one network
Single or several networks (of same type) in the same area
World wide or local
Certification
Interoperability with other networks on same frequency
Security
Price
Second source

/Ruben
<http://www.rjjournal.net>

> mamouda@... wrote:
> >
> > I'd like to design a wireless sensor network (20 sensors) using PIC
> > microcontrollers.
> >
> > Which PICs do you think is better for that purpose and which wireless
> > technology (Zigbee. WiMi ?)
>
> I have used XBees for several devices, but would like to move over to
> PICs with a Microchip stack at some point to reduce per-unit cost.
>
> My feeling at this point is that ZigBee has a lot of nice features, but
> if you don't need them, the fact that it's a standard is NOT important,
> because no two ZigBee stacks interoperate yet (as I understand it).  So
> it's really no more "standard," at this point, than Microchip's MiWi.
>
> So - my decision tree goes like this:
>
> A. Is short development time far more important than per-unit cost?  Or,
> are you looking to sell the product commercially?  Go with XBee.  (FCC
> approval is a major cost otherwise.)  There are some other similar
> modules that are no worse, but as far as I can tell, not particularly
> better either.
>
> B. If per-unit cost is important, and FCC approval isn't (or you can
> afford it), try Microchip's ZigBee solution.  Their stack is free, but
> you still have to join the "ZigBee Alliance" ($K).
>
> C. Or if you are OK with the subset included in MiWi (<= 1024 nodes, <=
> 8 coordinators), use that.
>
> So far, for home use, I have done (A) but lean toward (C) next.
>
> For small commercial products, I would go with (A) since the FCC
> certification process scares me.
>
> Hope that's of some use.
> --
> Timothy J. Weber
> http://timothyweber.org
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>

==============================
Ruben Jönsson
AB Liros Electronic
Box 9124, 200 39 Malmö, Sweden
TEL INT [phone]
FAX INT [phone]
ruben@...
==============================

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Timothy J. Weber-2]

Marc Nicholas wrote:
> Can you build a Zigbee stack, microcontroller, RFN PCB for less than
> $21? The new XBees cost that much in qty 1. Quite compelling!

True, they have come down in price - and it's been compelling enough so
far, so that's what I've used.

But it seems to me that using a free stack (Microchip's ZigBee or MiWi)
and a transceiver that costs <$2... yes, you'll have a micro but you
probably have one in there anyway for other functionality... and you
presumably have a PCB as well.  I don't know enough to estimate how much
more expensive the PCB would have to be.

I think ZigBee would *really* take off if it were economically feasible
to include in products that retailed for $50 or less.  $21 is pretty
good, but it's not quite there yet (especially if your product requires
a sender and receiver).
--
Timothy J. Weber
http://timothyweber.org
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Sean Breheny]

One more item to note: Maxstream now offers a development kit (free
download) which allows you to put your own code in the micro which is
on the XBee. I don't know the details but I imagine that if your code
is relatively simple, you may not need a second micro.

Sean


On Thu, May 8, 2008 at 10:31 AM, Timothy Weber <tw@...> wrote:

> Marc Nicholas wrote:
>> Can you build a Zigbee stack, microcontroller, RFN PCB for less than
>> $21? The new XBees cost that much in qty 1. Quite compelling!
>
> True, they have come down in price - and it's been compelling enough so
> far, so that's what I've used.
>
> But it seems to me that using a free stack (Microchip's ZigBee or MiWi)
> and a transceiver that costs <$2... yes, you'll have a micro but you
> probably have one in there anyway for other functionality... and you
> presumably have a PCB as well.  I don't know enough to estimate how much
> more expensive the PCB would have to be.
>
> I think ZigBee would *really* take off if it were economically feasible
> to include in products that retailed for $50 or less.  $21 is pretty
> good, but it's not quite there yet (especially if your product requires
> a sender and receiver).
> --
> Timothy J. Weber
> http://timothyweber.org
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Timothy J. Weber-2]

Sean Breheny wrote:
> One more item to note: Maxstream now offers a development kit (free
> download) which allows you to put your own code in the micro which is
> on the XBee. I don't know the details but I imagine that if your code
> is relatively simple, you may not need a second micro.

Yes, that's on my list to check out too.  I glanced at the code and felt
more of a "yuck" than a "yum," but maybe I just need to spend more time
with it and learn their toolchain (and possibly buy some of it - the
docs only say the free version of CodeWarrior version blah "may
suffice", but in other cases you need to use version blah blah).

The other issue with that approach is that you're limited to the XBee
module's pinout, which means what, something like 8 GPIOs and a few pins
you can cannibalize from other functions.  So it's a great approach for
the typical single-purpose sensor node, but less great for something
more complex.
--
Timothy J. Weber
http://timothyweber.org

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[dpharris]

Timothy J. Weber wrote:

> Sean Breheny wrote:
>  
>> One more item to note: Maxstream now offers a development kit (free
>> download) which allows you to put your own code in the micro which is
>> on the XBee. I don't know the details but I imagine that if your code
>> is relatively simple, you may not need a second micro.
>>    
>
> Yes, that's on my list to check out too.  I glanced at the code and felt
> more of a "yuck" than a "yum," but maybe I just need to spend more time
> with it and learn their toolchain (and possibly buy some of it - the
> docs only say the free version of CodeWarrior version blah "may
> suffice", but in other cases you need to use version blah blah).
>
> The other issue with that approach is that you're limited to the XBee
> module's pinout, which means what, something like 8 GPIOs and a few pins
> you can cannibalize from other functions.  So it's a great approach for
> the typical single-purpose sensor node, but less great for something
> more complex.
>  

Yes, this was my impression, too.  Although they give you options, all
the options looked either too limiting, or you needed compilers I did
not have.  However, I am hoping that has changed.

David



--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist