Microcontroller Selection for My loT edge device
What selection criteria should I use to cost-optimize microcontroller selection for my IoT edge device?
JIt depends on the application you want. It will determine the route to take
JCan you provide any details on it
SI'm exploring options for a smart agriculture IoT application where we need to monitor soil moisture levels and control irrigation remotely. The devices should be energy-efficient and support wireless communication.
JIf the land is vast enough i think using lora connectivity will be a good choice, since is low power and has a large coverage. You can keep a central node for that for internet access.
JAlso check out STM32 Lora based SoC, if its in stock. It might be a god option. Since it has both MCU and Lora capabilities
I"cost optimization" is tricky - needs context. Is it "lowest possible COGs" or, "least likely to require redesign NRE due to component EOL" or "most robust supply chain for an X year product life cycle?" Many of us have gone through cycles of COGs/Labor reductions that made our supply chains increasingly fragile - without recognizing that risk until it came back to bite us. In any case, sticking with standards based schemas and protocols will give you a good combination of cost and flexibility. LoRaWAN is certainly a good wireless choice for long range/low power sensor data and gives you an ecosystem of suppliers to work with.
SYour insights on the multifaceted nature of "cost optimization" in the IoT context are spot on. It's a delicate balance between minimizing COGs, ensuring long-term viability, and maintaining a robust supply chain.
Given your experience with supply chain fragility, have you found specific strategies or practices that effectively mitigate risks associated with component EOL and ensure sustained product success over an extended product life cycle?
Additionally, considering your positive stance on LoRaWAN for long-range/low-power sensor data, are there specific use cases or challenges where you've seen particular success with this wireless protocol?
Given your experience with supply chain fragility, have you found specific strategies or practices that effectively mitigate risks associated with component EOL and ensure sustained product success over an extended product life cycle?
Additionally, considering your positive stance on LoRaWAN for long-range/low-power sensor data, are there specific use cases or challenges where you've seen particular success with this wireless protocol?
IA few tips on the selection process; a) Look at the vendors track record for EOL processes, b) look at parts that are in the "young" end of their lifecycle yet mature enough to have gained popularity, c) look at parts with broad adoption, talk to your disti's to get info on which parts have lots of customer/design wins, d) identify and design in/qualify alternative parts wherever possible on the front end, often simple design tweaks will allow your production pcb to support alternative parts, and where possible, get those alternative parts listed in your compliance reports, e) learn what the suppliers supply chain looks like, are they fabless, are they on "mainstream" fab processes or on legacy (for example, 8" wafers). f) for high risk parts, place long term orders to secure the supply chain (with a trusted disti, this could be a chapter in itself). Finally (actually, firstly), do a bit of risk analysis. How long is the expected lifecycle, what is the engineering cost of redesigns, the compliance cost, revenue risk.
IRegarding LoRaWAN - seeing lots of adoption in enterprise environments, commercial space, retail, warehouse. It also shines in outdoor, environmental, ag applications. The ecosystem is strong and growing, a number of good choices for gateway providers, and good architecture flexibility in public/private/local network server choices. The one point I don't like is that LoRaWAN is dependent on LoRa and LoRa (the physical layer) is proprietary to Semtech. So unlike wifi, BT, it's not a licensed technology adopted across many vendors. That said, Semtech did perform well through the supply chain crisis over the last few years and LoRa is clearly a strategic growth segment for them.
SYou've provided a comprehensive overview of LoRaWAN's strengths and adoption across various sectors. The proprietary nature of the LoRa physical layer does present a challenge in terms of vendor diversity. However, the robust ecosystem and strategic focus by Semtech showcase its resilience. As technology evolves, it will be interesting to see how LoRaWAN addresses these challenges and maintains its position in diverse IoT applications.
