Questions on project

New to group and have several questions on project I’m working on. Basically am building out a device used by scuba divers as a gps tracker. Diver carries a touchscreen device underwater which is tied via tether coax cable (with wifi) to a buoy which uses xbee mesh network to communicate over long distance. Looking for an open source dashboard control which would allow for google maps usage along with control-ability of UX. HA seems like a great option for this.

Looking through documentation of HA and have several questions in regards to if this design should work.
Am wanting network architecture to be versatile and portable so user can setup remotely. Diver Unit would be running Pi CM4 and buoy with esp32. Am considering having HA setup on Pi as OS allowing user to view sensor data from multiple devices including buoy. Should it be possible to do so with HA setup on multiple units without conflict? Am wanting to setup so devices would not necessarily be internet connected yet be on same network. Should internet become available would want to enable automation where data could be streamed to cloud if need be.

How easily could this be setup with Xbee network? Xbee attached within buoy with esp32. Buoys would be setup in mesh configuration to allow users from shore to access. Not expecting to share anything more than sensor data between units (ex. IMU, GPS, battery status).

A good question is how deep will the divers go? Water temps? How far away from shore?

I do not know if HA will be appropriate on the diver level just from a complexity standpoint, though using it to build a proof of concept interface is just fine. ESP might be ok on both the diver and the diver buoy, and be far superior from a power usage standpoint.

If you do want internet connectivity for sensor readings, you can use LoRa on the buoys instead, and have an aggregate “local” buoy that gets all the readings from the divers (if they are in a group), then transmits sensor/comm data back to shore over the buoy network. Lower frequencies like LoRa provide a much more reliable network at the same range as xbee, or allow the same reliability at longer range, meaning less buoys needed. You can expect 5kms from LoRa on open water, even more with better antennas and full power transmitters (a 1.2 meter antenna has gotten more than 20km), though maybe only 1 to 2 Kb/s of bandwidth, but that is more than enough for periodic location/sensor measurements and 2-way text communication even with multiple divers.

The local buoy would comm with the divers over the 2.4GHz network, and any messages between them would stay local, allowing things like push to talk audio. High frequency sensor readings can be averaged on the buoy then transmitted at less frequent intervals, like every 15s. A message queuing system there would keep the bandwidth consistent, and provide a means of retransmitting messages if you do not get an ack.

Am expecting divers to go to 100ft with buoys having max range of 1000ft to 10 miles. Typical distance would be 100-500 ft. Am also considering using Xbee rather than Lora or wifi. Ideally hoping to use Meshtastic as this would allow an ad-hoc mesh wireless network independent of web or cloud use. However finding that meshtastic is currently limited to preexisting hardware. Looking for easy to setup and quick deployment solution. Meshtastic uses bluetooth from android app to sync to the device for setting updates. Firmware would still require serial usb connection. Not looking to transmitting large files (such as voice, audio or video), just streamed sensor data and txt messages. Does there exist any code examples involving xbee with HA? Looking in particular for mapping of gps data.