Why mesh networks, why now
When Hurricane Helene hit western North Carolina in September 2024, it did not just knock out power. It took down cell towers, severed fiber lines, and made satellite uplinks unreliable for days. Communities that needed to coordinate mutual aid were cut off from the tools they had come to depend on.
This pattern repeats with every major disaster. The first infrastructure to fail is always communications, and the communities least served by it in normal times are the ones most exposed when it collapses.
Mesh networks are one answer. A mesh is a collection of radio nodes where every node can relay messages for every other node. There is no central tower, no ISP, no subscription. LoRa (Long Range) radio is the physical layer underlying all three systems described here. It operates in the unlicensed 915 MHz band in the US, requires no radio license, and can carry text and sensor data across 1 to 15 miles depending on terrain and antenna height.
Three protocols, one hardware family
All three systems run on the same LoRa hardware. A node can be reflashed from one protocol to another. The difference is entirely in software philosophy. Click a protocol below, then send a message to watch how it propagates through the network.
The plug-and-play mesh. Flash firmware, open the app, and you are on the network. The largest deployed LoRa mesh system in the world, with active community networks in dozens of US cities.
How it routes
Flood routing: every node that receives a message rebroadcasts it to all neighbors immediately. Simple, resilient, zero configuration. A new node joins and participates instantly.
Strengths
- Zero configuration to get started
- Enormous existing community coverage across the US
- AES-256 encrypted channels by default
- GPS tracking and sensor telemetry alongside text
- Strong mobile app with offline maps
Limitations
- Flood routing becomes inefficient at large scale
- No store-and-forward: messages are lost if the recipient is offline
- Limited sensor telemetry types supported
Infrastructure you build and own. Launched in early 2025, MeshCore uses structured routing through designated repeater nodes, and a Room Server that stores messages for recipients who are temporarily offline.
How it routes
Only designated repeater nodes forward messages. A Room Server holds messages until the recipient comes online. For rural communities where nodes are not always powered, this store-and-forward capability is essential rather than optional.
Strengths
- Store-and-forward via Room Server
- More efficient bandwidth use than flood routing
- Designed for permanent fixed solar-powered infrastructure
- Runs on the same hardware as Meshtastic
Limitations
- Smaller community than Meshtastic
- Requires deliberate planning and placement of repeaters
- Newer software, less battle-tested
Not firmware but a full network stack, analogous to TCP/IP but built from the ground up for off-grid, decentralized, surveillance-resistant communication across multiple transport types simultaneously.
How it routes
Reticulum spans LoRa, WiFi, Ethernet, and Bluetooth in the same network. It uses path-based routing: the network discovers routes and remembers them. A message can travel over LoRa, hand off to WiFi, then return to LoRa automatically. Identity is cryptographic, with no accounts or phone numbers required.
Strengths
- Multi-transport: LoRa, WiFi, Ethernet, Bluetooth in one network
- Scales to regional and inter-regional distances
- Store-and-forward at the protocol level
- Strong surveillance resistance by design
Limitations
- RNodes require a companion device currently and cannot act as standalone solar repeaters yet
- Steeper learning curve than Meshtastic or MeshCore
- Smaller user base
The standalone RNode limitation is an active development priority. A Raspberry Pi Zero 2W (~$15) on solar power is a workable bridge for now. We will not promise what we cannot guarantee, but we expect this changes soon.
Side-by-side comparison
| Feature | Meshtastic | MeshCore | Reticulum |
|---|---|---|---|
| Routing approach | Flood | Structured | Path-based |
| Store-and-forward | ✗ | ✓ Room Server | ✓ Deep support |
| Standalone solar node | ✓ | ✓ | Not yet |
| Multi-transport | ✗ | ✗ | ✓ |
| Existing community networks | Many US cities | Few | Very few |
| Setup complexity | Low | Medium | High |
| Scales to regional | Limited | Medium | High |
| Surveillance resistance | Moderate | Moderate | Strong |
| Hardware cost | $30–$80 | Same | Same + companion |
| Protocol maturity | High (2019+) | Low (2025) | Medium |
All three protocols run on the same hardware. A group can begin with Meshtastic and reflash nodes to MeshCore or Reticulum as capacity grows. A working Meshtastic node today is worth more than a perfect Reticulum network next year.
Soil sensor integration
Kiau's field sensing nodes are outdoor-capable, solar-powered, and transmit over LoRa into the same mesh networks described above. The same node carrying a text message can carry soil data. No separate network required.
How data moves through each protocol
In a Meshtastic network, sensor nodes transmit telemetry packets at set intervals, every 15 minutes for example. These packets hop across the mesh and are logged at a base station node connected to a local device. No internet required.
In a MeshCore network, sensor nodes transmit to the nearest repeater. The Room Server stores readings until the base station comes into contact. This is advantageous for remote fields where the sensor may be far from any base station and connectivity is intermittent.
In a Reticulum network, sensor payloads route across whatever transports are available: LoRa, then WiFi, then LoRa again if needed. For organizations with mixed infrastructure across large areas, this is the most flexible architecture for moving sensor data from field to farmer.
Use cases
Arid region farms
Water harvesting sites
Conservation grant work
Community gardens
Data sovereignty
For Indigenous land stewards and communities with histories of data extraction, the question of who owns and controls sensor data is not a technical afterthought. It is foundational. Kiau's architecture is designed with this in mind from the start.
Three horizons
The goal is a network of networks: regional clusters built on infrastructure each community owns, bridging to each other over time. No single protocol connects everywhere on day one, and that is fine.
- Urban groups join existing Meshtastic networks
- Rural groups stand up MeshCore infrastructure
- Technically advanced groups begin Reticulum experiments
- Sensor nodes deployed at priority farm sites
- Regional clusters emerge and connect
- Sensor data flows alongside communications
- Cross-protocol bridges tested between clusters
- Community-owned repeater infrastructure expands
- Reticulum transport nodes at each regional hub
- Cohort-wide backbone connects regional clusters
- Sensor, communications, and mutual aid on shared infrastructure
- Owned entirely by the communities it serves