Thread protocol for outdoor sports accessories

[Mxoh]

Hi there,

We are a cycling accessories company (mainly light accessories) and have been using Nordic chips for BLE 1 to 1 connection of our helmet to phone (for configurations) and ESB/2.4GHz of our helmet to remote (for turn signals). As our company's direction is to produce more accessories and for them to be in a network, we have been researching and experimenting on this.

Frankly we are no experts and initially tried BLE mesh from our experience with BLE but found that the power consumption might be too high due to need for constant advertising and scanning. We came across Nordic BLE multi-link and planning to start some experiments with it soon but I am apprehensive as it is requires to define central and peripheral. My questions are:

1. Due to time constraint, we are likely still using BLE to create the network but I am wondering if there is possibly that in the future we would be able to support a supposedly better network protocol like Thread?

2. As our products are mainly outdoor accessories, we do not have a home network to tap into and rely on battery. I know that Zigbee requires a hub to be connected to a router but is there still a chance we can use Thread?

FYI we are currently using Nordic IC nRF52832 but have purchased Nordic nRF52840 with a development board router to experiment with Thread.

[Simonr]

Hi

Thank you for the explanation.

Should all these devices be able to send messages to all other devices? If so, it seems like they all have to be "Routers" which will require them to have their transceiver enabled at all times, which will impact current consumption significantly. As you can see here, only the end nodes will at times disable their transceiver to reduce current consumption, but in turn, these won't be able to send messages to all other nodes in the network, and only to one router. I'm afraid a Thread network won't be what you're looking for, as you won't be able to have your cake and eat it too in this scenario. It is either low current consumption as an end node or "able to forward packets and listening/transmitting to all other devices".

I still think the multi-link would be the go-to. Yes, you'll need a central, and multiple peripherals, but a peripheral can send a message to "change light pattern" to the central, which then can tell the other peripherals to also change the light patterns. You could either have the phone as the central (although it would require you to bring the phone on each trip), or you could have any of the devices configurable as either central or peripheral (depending on what devices the customer has). So you could have a glove be the central, the helmet, or either light, and all the others be the peripheral.

Yes, if the central goes down for some reason, the other devices will eventually time out and disconnect. If all devices then are configurable as centrals, you could choose a new central and set up a new network with the remaining devices.

Best regards,

Simon

[Mxoh]

Thanks for the super clear answer Simon! 

It seems more clear to me now that multi-link is our way forward then. I guess we were too heavily influenced by the hot and trendy networking technologies and multi-link is somehow not as well-known. I am curious though as whether there is more information or numbers on power usage of the various technologies as it seems that network technologies are mostly marketed as low power and we weren't sure until we actually tried them.

For multi-link would it be possible if 1 central goes down and the remaining devices are able to automatically pick the new central without phone intervention? 

Also would you recommend other technologies such as ANT+ for our outdoor accessories use-case? I do think my company is at a stage now where we are open to experimenting but I believe that once we pick one technology we would be stuck with it for a while.