Porting MySensors to work with the RadioHead library
-
@kolaf said:
I think we should expand the setupRadio function to configure some useful defaults for every radio
Can't you wrap the radio setup and some more stuff in a class for every radio type, derived from a single interface class?
I hope in the actual mesh layer etc. you don't have to distinguish between different radio types anymore.... -
@kolaf said:
I think we should expand the setupRadio function to configure some useful defaults for every radio
Can't you wrap the radio setup and some more stuff in a class for every radio type, derived from a single interface class?
I hope in the actual mesh layer etc. you don't have to distinguish between different radio types anymore....@Yveaux said:
@kolaf said:
I think we should expand the setupRadio function to configure some useful defaults for every radio
Can't you wrap the radio setup and some more stuff in a class for every radio type, derived from a single interface class?
I hope in the actual mesh layer etc. you don't have to distinguish between different radio types anymore....I guess I could do that, but it sounds like more work :-). For now I will stick to a define section for each radio that configures the defaults. After all, most of the radio settings for the original version of the library also relies on defines.
I think the RF 69 configuration of the library is ready for testing by others, and I have also verified that the NRF 24 version of the library compiles. I do not have any hardware for this one, so maybe someone else wants to test to see if it works? I have forked the original repository, and my current progress can be found in the following branch (Radiohead_port):
-
If you want to test, be aware that the default pins for the constructor set up to match the RF 69 radio. You have to specify the select and enable pins in the constructor explicitly.
-
I just had a quick look at the RadioHead implementation.
All drivers are derived from RHGenericDriver (from http://www.airspayce.com/mikem/arduino/RadioHead/classRHGenericDriver.html):
If you pass an instance of RHGenericDriver in e.g. the c'tor of MySensors you're done.... But maybe I had a too-quick look ;-)
-
It is not that simple because all the different specific drivers implement a bunch of their own functions which are not virtualised. This means that you need to static cast or dynamic cast the driver every time you want to use it which is a real pain.
On the other hand, each driver seems to come with a relatively intelligent set of defaults, so I think it is quite easy to cater for the generic case, and then we can expose the internals for those who really want to get into the specifics. Still, if you want to try to generalise it in a different way, feel free :-)
-
It is not that simple because all the different specific drivers implement a bunch of their own functions which are not virtualised. This means that you need to static cast or dynamic cast the driver every time you want to use it which is a real pain.
On the other hand, each driver seems to come with a relatively intelligent set of defaults, so I think it is quite easy to cater for the generic case, and then we can expose the internals for those who really want to get into the specifics. Still, if you want to try to generalise it in a different way, feel free :-)
@kolaf said:
It is not that simple because all the different specific drivers implement a bunch of their own functions which are not virtualised. This means that you need to static cast or dynamic cast the driver every time you want to use it which is a real pain.
Are those driver-specific functions needed during normal operation, or are they only for non-default setup? (eg: frequency, bandwidth, etc)
-
What about packet size? I'm guessing that the idea is to use the 32 byte limit of the nRF24L01+ for all radios (even if they have larger packet options) right?
(This approach is frequently called "least common denominator", tho the actual mathematical concept being misreferenced is actually the greatest common denominator)
-
I don't think the flash and RAM footprint is a deal breaker in this application, but it could be a concern for some multi-sensor nodes which include several sensor libraries. So I was wondering how much heavier it is, but not saying (yet) that it's too large.
Remember to save room for a possible crypto library, if we do a more secure version.
My own use case is unusual. I am using nRF24L01+ for christmas light control, which involves a high bandwidth transmission from hub to nodes (and uses a fair amount of RAM on the nodes for buffers). I've been thinking that I could have the Christmas light display nodes also transmit sensor data to a MySensors network periodically,since they use the same radio and processor. That isn't critical tho. I could just use separate nodes on different channels for the two applications.
-
What about packet size? I'm guessing that the idea is to use the 32 byte limit of the nRF24L01+ for all radios (even if they have larger packet options) right?
(This approach is frequently called "least common denominator", tho the actual mathematical concept being misreferenced is actually the greatest common denominator)
@Zeph said:
I'm guessing that the idea is to use the 32 byte limit of the nRF24L01+ for all radios (even if they have larger packet options)
Why? MySensors can easily handle payloads with different sizes. Only when you start mixing different radios in one network ( hek apparently thinks in this direction) you have to handle fragmentation.
-
@hek Maybe, but I have yet to see a valid usecase for this. You can always use 2 Arduino's running the default serial gateway and tie them together ;-)
I haven't looked into the radiohead interface yet (have to admit you guys got me curious, though) but isn't there simply one base class defining the interface and every radio derives from it? That would definately make things simple!@hek said:
Yes, defines certainly make sense now! But it would be cool to have gateway-nodes between different radio models in the future@Yveaux said:
Maybe, but I have yet to see a valid usecase for this. You can always use 2 Arduino's running the default serial gateway and tie them together ;-)
The RF69 has some nice possibilities for range using 433 or 868/915 MHz, albeit with less bandwidth than the nRF24L01+ at 2.4 GHz. That makes the former attractive for low bandwidth applications like a sensor network. I like the latter for some other projects that make use of the higher bandwidth, but it could be nice to include MySensor network as well using the same radio. That's what brought me here. So I could envision using both radios - RF69 for pure low bandwidth sensors and nRF for combined nodes.
@Yveaux, is there really a way to tie two hubs together for MySensors.
-
@Zeph said:
I'm guessing that the idea is to use the 32 byte limit of the nRF24L01+ for all radios (even if they have larger packet options)
Why? MySensors can easily handle payloads with different sizes. Only when you start mixing different radios in one network ( hek apparently thinks in this direction) you have to handle fragmentation.
@Yveaux said:
@Zeph said:
I'm guessing that the idea is to use the 32 byte limit of the nRF24L01+ for all radios (even if they have larger packet options)
Why? MySensors can easily handle payloads with different sizes. Only when you start mixing different radios in one network ( hek apparently thinks in this direction) you have to handle fragmentation.
Yeah, but handling fragmentation might add a lot of complication and code. If the network was designed around "expect all radios to handle at least 32 byte packets" it could be lighter and simpler.
And this is not just about mixing radios in one system; if one writes a sensor which makes use of more than 32 byte packets because it's initially using the RF69, then it would not port well to another system using a non-fragmentation nRF implementation.
-
@hek said:
Yes, defines certainly make sense now! But it would be cool to have gateway-nodes between different radio models in the future@Yveaux said:
Maybe, but I have yet to see a valid usecase for this. You can always use 2 Arduino's running the default serial gateway and tie them together ;-)
The RF69 has some nice possibilities for range using 433 or 868/915 MHz, albeit with less bandwidth than the nRF24L01+ at 2.4 GHz. That makes the former attractive for low bandwidth applications like a sensor network. I like the latter for some other projects that make use of the higher bandwidth, but it could be nice to include MySensor network as well using the same radio. That's what brought me here. So I could envision using both radios - RF69 for pure low bandwidth sensors and nRF for combined nodes.
@Yveaux, is there really a way to tie two hubs together for MySensors.
@Zeph said:
is there really a way to tie two hubs together for MySensors.
Maybe it won't work out of the box, but in principe the serial format for send/receive is identical. I think that just cross-connecting the serial ports of two gateways will get you close to a working solution. When different speeds have to bridged, of (de)fragmentation is required the whole story changes...
Btw. if you want different networks, why not just use a different gateway for each network? Or is Vera not able to handle multiple gateways?
-
One other consideration - there is a thread on Security. if MySensors were to migrate towards multi-radio support (the nRF and the RF69 being the most attractive options), then it might be tempting to design any security layer to make use of the latter's built in AES encryption, with its strengths (fast with little code) and weakneses (less flexible than a software implementation). That is, use AES in software on the nRF to do the same approach as is done in hardware with the RF69.
This would not be required (each radio could use it's own security approach, or the same software security could be used for all of them), just somewhat attractive to keep down the number of separate implementation while also making use of the hardware.
-
@Yveaux said:
@Zeph said:
I'm guessing that the idea is to use the 32 byte limit of the nRF24L01+ for all radios (even if they have larger packet options)
Why? MySensors can easily handle payloads with different sizes. Only when you start mixing different radios in one network ( hek apparently thinks in this direction) you have to handle fragmentation.
Yeah, but handling fragmentation might add a lot of complication and code. If the network was designed around "expect all radios to handle at least 32 byte packets" it could be lighter and simpler.
And this is not just about mixing radios in one system; if one writes a sensor which makes use of more than 32 byte packets because it's initially using the RF69, then it would not port well to another system using a non-fragmentation nRF implementation.
@Zeph said:
Yeah, but handling fragmentation might add a lot of complication and code. If the network was designed around "expect all radios to handle at least 32 byte packets" it could be lighter and simpler.
I'm certainly not in favour of implementing fragmentation in MySensors, and if it would be added it should be handled at a lower level anyway.
And this is not just about mixing radios in one system; if one writes a sensor which makes use of more than 32 byte packets because it's initially using the RF69, then it would not port well to another system using a non-fragmentation nRF implementation.
All the examples that come with the MySensors library use a lot less than the maximum MySensors payload (which is less than 32 bytes due to the header btw). I don't have the actual figures, but I think the largest payload is sent in the sensor presentation (the version string).
For radio-portability it would indeed be benificial to have a minimum payload size defined. -
@Zeph said:
is there really a way to tie two hubs together for MySensors.
Maybe it won't work out of the box, but in principe the serial format for send/receive is identical. I think that just cross-connecting the serial ports of two gateways will get you close to a working solution. When different speeds have to bridged, of (de)fragmentation is required the whole story changes...
Btw. if you want different networks, why not just use a different gateway for each network? Or is Vera not able to handle multiple gateways?
@Yveaux said:
Maybe it won't work out of the box, but in principe the serial format for send/receive is identical. I think that just cross-connecting the serial ports of two gateways will get you close to a working solution. When different speeds have to bridged, of (de)fragmentation is required the whole story changes...
I've been thinking of using a APM as a "smart peripheral" to a node (for Christmas lighting, because driving some pixel light strings requires turning off interrupts for long periods). A 5V APM is as little as $2.30 shipped, so I could make a light controller intermediary with a more asynchronous interface (so the main node processor can be interrupted).
Something similar might also work with two MySensor radio hubs, especially given the relatively low bandwidth. One radio hub might be an APM connected to the main hub.
If the home automation controller will accept multiple gateways, tho, you may be right that it would be the simpler solution.
-
@Yveaux said:
Maybe it won't work out of the box, but in principe the serial format for send/receive is identical. I think that just cross-connecting the serial ports of two gateways will get you close to a working solution. When different speeds have to bridged, of (de)fragmentation is required the whole story changes...
I've been thinking of using a APM as a "smart peripheral" to a node (for Christmas lighting, because driving some pixel light strings requires turning off interrupts for long periods). A 5V APM is as little as $2.30 shipped, so I could make a light controller intermediary with a more asynchronous interface (so the main node processor can be interrupted).
Something similar might also work with two MySensor radio hubs, especially given the relatively low bandwidth. One radio hub might be an APM connected to the main hub.
If the home automation controller will accept multiple gateways, tho, you may be right that it would be the simpler solution.
-
I think we need to think about choosing a bitter atmega for this.. Its a bit too bad that atmega havn't made a atmega628p which is pin compatible with 328 but with 64KB of program storage instead.. :) so the next up is 644 which arduino has no boards for. but there is sanguino.cc - https://code.google.com/p/sanguino/ though.
Next up would be Arduino Mega 2560 with what it feels like 'Unlimited ram!'
They are not That expensive today since china made its own ch340 usb - serial chip (Why this was the expensive component blows my mind :P) ATmega2560-16AU CH340G MEGA 256 you can find for ~10EUR / 13.50USD..@Damme said:
I think we need to think about choosing a bitter atmega for this..
Next up would be Arduino Mega 2560 with what it feels like 'Unlimited ram!'
They are not That expensive today since china made its own ch340 usb - serial chip (Why this was the expensive component blows my mind :P) ATmega2560-16AU CH340G MEGA 256 you can find for ~10EUR / 13.50USD..Also note the ATMega2560 small form factor offering: http://www.ebay.com/itm/261541870545 which is similar to an Arduino Pro Mini but with the larger processor. And it has even more pins broken out than the Arduino Mega2560.
-
@kolaf said:
It is not that simple because all the different specific drivers implement a bunch of their own functions which are not virtualised. This means that you need to static cast or dynamic cast the driver every time you want to use it which is a real pain.
Are those driver-specific functions needed during normal operation, or are they only for non-default setup? (eg: frequency, bandwidth, etc)
@Zeph said:
@kolaf said:
It is not that simple because all the different specific drivers implement a bunch of their own functions which are not virtualised. This means that you need to static cast or dynamic cast the driver every time you want to use it which is a real pain.
Are those driver-specific functions needed during normal operation, or are they only for non-default setup? (eg: frequency, bandwidth, etc)
Just for setup. Runtime operation is handled by the manager and is independent of the radio driver.
-
@Damme said:
I think we need to think about choosing a bitter atmega for this..
Next up would be Arduino Mega 2560 with what it feels like 'Unlimited ram!'
They are not That expensive today since china made its own ch340 usb - serial chip (Why this was the expensive component blows my mind :P) ATmega2560-16AU CH340G MEGA 256 you can find for ~10EUR / 13.50USD..Also note the ATMega2560 small form factor offering: http://www.ebay.com/itm/261541870545 which is similar to an Arduino Pro Mini but with the larger processor. And it has even more pins broken out than the Arduino Mega2560.
@Zeph And you even get the "mini mega" presoldered with the radio chip of your choice for $20 :-)
-
@Zeph And you even get the "mini mega" presoldered with the radio chip of your choice for $20 :-)
@kolaf said:
And you even get the "mini mega" presoldered with the radio chip of your choice for $20 :-)
Well, it's actually another $6-7 for the radio, $1 for pins, and optionally $3 for extra flash. But it's still a good deal.
It's nice to have more ATMega1284 based Arduino styles options available. The official Arduino folks went for the ATMega1280/2560 instead for their high end AVR based system, and they have their advantages, but the extra RAM on the 1284 is probably more useful than the extra Flash memory.
For our nodes, the Moteino Mega probably has more than enough IO. Compared to the Arduino Pro Micor, it basically has 10 additional digital pins (plus all 8 analog pins can be used as digital, not just 6 of them). I2c is done with digital IO pins rather than analog pins. And it has another serial port.
The ATMega2560-core (I posted an ebay link a couple msgs back) has a huge number of IO pins, including many that were not available in the Arduino Mega board - eg: the pins needed for doing SPI master mode with the additional USARTS. There are applications that might be useful for, but for sensor network nodes we would rarely need so many I/Os.
(I'll stop there - these messages on other boards and chip might be a good candidate for a moderator to move to another thread)
