Hello Everyone!

Hi @ao,

   Just a first impression thought/query, based on the diagram.

If I have understood the diagram correctly, this appears to be a system that (hopefully) will rarely need to be activated, but when activated, any failure to operate could be life-threatening, assuming a typical alert might be a fire, as suggested in the diagram.

Furthermore, the diagram suggests that an alert must pass through several relay stations before reaching the team responsible for dealing with the problem. This implies that any failure in the message passing chain is likely to prevent an alert arriving, and having several potential failure points increases the chance that an alert will fail to arrive. 

There are sometimes ways of reducing or overcoming such concerns, but these need to be identified and evaluated, to ensure the system will meet the reliability requirements of such a system. I have a suspicion this may require changing the system design.

Apologies if I am raising a question that you have already addressed.

Good luck with your project.

.

Hi @ao,

  If I understand your diagrams correctly, then it looks like you are hoping to have at least two different routes that a message could take. This obviously significantly improves the chance that a message will arrive, which I think is an important part of such a project. Of course, you will be constrained by the physical positioning of nodes, etc., so no real system can be guaranteed to function when it is most needed, but hopefully you can analyse the design, and possibly take measures to improve the probability of its success.

As a general suggestion, try looking into how meshes are constructed, which aim to address this sort of problem. 

Also, consider how your system monitors and checks it own performance, including alerting users to failing links on a timely basis, so that the chance that a vital link is faulty, at the time of an emergency, is minimised.

My understanding is that the origins of the internet were based on US initiative called Arpanet, with the aim of producing a network with many redundant paths, so that the total effect of the loss of specific nodes was mitigated by the network routing packets of data by alternate paths. In addition, protocols were defined that automatically retried or rerouted data packets that failed to arrive. I am not an expert in this field, but my understanding is that all the fundamental protocols etc. are openly published, and hence can be consulted for guidance. Of course, this is dependent upon the availability of hardware to support redundant paths, which may be challenging in your situation. 

I note your diagram 'suggests' you may be thinking about WiFi and Bluetooth, although that maybe a mistaken interpretation on my part. In general, these are usually considered as short range, although the range can sometimes be extended by using higher gain antenna, etc. 

I presume you are aware of different radio networks, one or more of which may be more suitable for your requirements. Again my personal experience is limited, but as an example, Bill (@dronebot-workshop), this forum's kind host, has an excellent introduction to LoRa 

https://dronebotworkshop.com/lora/

which includes sending the temperature of his freezer, from a sensor and transmitter, entirely enclosed within the freezer! The report includes "Try and see how far you can get for range. I was able to get two blocks with one of my remote sensors in the freezer in my basement! All this with a 3-inch wire antenna!"

So with a more 'sensible' antenna position and type, it seems reasonable to expect a considerable range.

I also note that there is a related LoRaWAN initiative, that uses LoRa technology to pass small messages across an arbitrary network. You may like to explore this development as well.

Of course, this suggestion might be extended to a hybrid approach, with say WiFi or Bluetooth messaging a local node, which then communicates with other nodes using LoRa or some other system, for a longer distance hop.

------------------

I am sorry, this is not my general area of experience, and I cannot provide any guarantee of personal support for your project, but I suggest you continue to post specific questions, to this forum, and possibly elsewhere. Hopefully, someone will be able to provide a helpful response. Of course, you must be careful to judge the quality of each response, and only follow those that you think are factually correct and helpful. That is true of research in all fields and levels.

I wish you good luck, and best wishes, Dave