> > 1. The idea of individual switches with some kind of autonomy is
> worth
> > to examine it's feasibility. The switch could talk directly to the
> lamp
> > (group of lamps) and vice versa.
> That seems like a good idea to me. I think the switch has to tell the
> lamps individually. It can't rely on the lamps receiving a broadcast
> because we have to assume the communications medium is unreliable.
> Possibly it could send a broadcast and await acknowledgement(s), but
> how does it know what should have replied, or if anything should reply
> at all? If it knows who should reply, it may as well address then in
> the first place. No, I think the sensors have to be responsible for
> reliably transmitting their outputs to the module(s) registered with
> them.
>
Maybe you can do that with groups.
> > -the µ-controllers should have some EEPROM on board (to store the
> > address & parameters)
> Or be battery-backed. Assuming the switch can get power from the
> mains
> or dedicated wiring, the backup only has to tide it over short
> outages.
> Anyway, most microcontrollerss have at least some protected memory
> these
> days, and an address isn't going to be many bytes even in the most
> grandiose scheme.
>
Albeit a battery takes up a lot of space which I don't have in
my wallplates.
> > -is it necessary to have a multi master protocol? I would say yes,
> > although this complicates the issue a bit.
> I would say yes too. I don't think it can be avoided, given the goal
> of decentralized control. Every module has to be able to do its bit,
> regardless of the state of the rest of the system, so everything
> capable of initiating an action (which could be any sensor) has to
> be master-capable.
>
Therefore the protocol should be 'scalable' (if this is the
correct term?). Only implement the things you really need.
It is possible to use RS485 as the physical layer for CAN. But
CAN resolves conflicts by collision resolution (bit dominance). If 2
RS485 devices on the a bus transmit different data, the RS485 output
stage goes in thermal shutdown? Can someone explain me how this problem
is solved, or am I missing something?
Take a look at http://www.linet.fi and
http://www.ncl.ac.uk/~n07tv/index.html. The site from Warriner has an
intresting paper, but his hardware is too simple. I cannot believe that
has achieved a high noise immunity level with this. I cannot judge the
software. However, he explains some intresting topics.
Have a nice weekend
Zeb