LoRaWAN vs Wi‑Fi HaLow vs Meshtastic — Demory farm sensor layer¶
Question¶
For SITE_FARM (Demory), ~120 acres, off-grid-first field sensing, how do LoRaWAN, 802.11ah / Wi‑Fi HaLow, and Meshtastic (LoRa mesh) compare on coverage, power, ops, survivability, and IP vs sparse telemetry?
Scope: Field RF for sensors—not Starlink as a radio (Starlink is WAN only—Farm sensor architecture — Demory).
Wider four-way (incl. conventional Wi‑Fi): Wi‑Fi HaLow vs LoRaWAN vs Meshtastic vs conventional Wi‑Fi. LoRaWAN vs Meshtastic only: LoRaWAN vs Meshtastic for fixed farm telemetry.
Sources: Demory farm sensor layer — official and operator-grade source cluster.
~120-acre coverage (assumptions)¶
| Stack | Coverage mental model | Planning notes |
|---|---|---|
| LoRaWAN | Star to one or more gateways; km-class links possible with height and clearing | Terrain may require multiple G or tall mount — RF survey . |
| Meshtastic | Multi-hop mesh fills holes where nodes exist | Density and map ≠ RF ; partitions possible . |
| Wi‑Fi HaLow | Sub-GHz IP links ; range depends on AP/CPE and rate | Higher throughput than LoRa but not automatic farm-wide mesh . |
No single answer without terrain map + gateway locations (not asserted here ) .**
Criteria (sensor-layer focus)¶
| Criterion | LoRaWAN | Wi‑Fi HaLow | Meshtastic |
|---|---|---|---|
| Bandwidth | Very low (application payloads )** | Mbps-class IP where deployed | Low (text / telemetry class )** |
| Battery life | Excellent for sleeping Class A -style devices | Worse for always-on CPE vs deep sleep LoRa | Good leaf ; higher when routing |
| Solar friendliness | Strong for sparse uplinks | Plan for AP + CPE load | Strong for small nodes ; budget repeaters |
| Deployment complexity | Gateway + NS (e.g. ChirpStack ) + join model** | 802.11 ops (patching , segmentation )** | Mesh map + firmware cadence |
| Gateway architecture | Central GW → NS | AP / bridge → LAN | Bridge → MQTT at G |
| Local survivability | Works without WAN ; GW SPOF risk | LAN segment without WAN | Mesh local ; partition risk |
| Maintenance burden | NS + GW health | Firmware / Wi‑Fi security | Mesh health + releases |
| IP-native | No (app layer on NS )** | Yes | No (bridge for IP )** |
| Sparse telemetry | Excellent | Overkill unless you need IP | Excellent |
| Richer edge / actuation | Limited by duty-cycle / rate | Better when IP needed | Limited ; not PLC replacement |
| Observational sensors | Strong fit | Fit if IP sensors / throughput | Strong fit |
| Control / actuation | Usually downlink constrained ; design carefully | Possible on IP path | Do not assume without engineering review |
Recommendation heuristic (wiki ops)¶
- Default sparse soil / weather / tank : LoRaWAN or Meshtastic leaf — see
Sensor checklist matrix — Demory. - Need IP throughput or MQTT on MCU at distance : HaLow segment where hardware ROI clears .
- Never treat Starlink as field sensor link .