LoRa Network Coverage Planner

Deploying a LoRa sensor network without planning usually ends in dead zones: gateways get mounted wherever a pole is handy, and nodes at the far end silently drop packets. Professional RF planning software is expensive and assumes a radio-engineering background, so student and campus-scale LoRa projects are deployed by guesswork. LoRaPlan fills that gap with an honest, explainable planner. You describe the site (rural, suburban or urban-campus), pick the transmit power and the slowest acceptable spreading factor, and place gateways on the map. The tool applies a log-distance path-loss model with environment-specific exponents, computes the expected RSSI at every node, and colours each node by link margin against real Semtech SX1276 receiver sensitivities. The result is a coverage map you can defend in a viva, a link-budget table for the report, and a deployment plan that says how many gateways the site actually needs — before any hardware is ordered.

1. The student selects the environment preset, which sets the path-loss exponent n in PL(d) = 80.5 + 10·n·log10(d/10 m) at 868 MHz.
2. Gateways are placed on the site map with a chosen TX power (10, 14 or 20 dBm) and mast height note.
3. For every planned end node, the tool computes path loss to the nearest gateway and the expected RSSI from TX power plus antenna gains minus path loss.
4. Each node's RSSI is compared against the SX1276 sensitivity for the selected spreading factor and coloured strong (≥ −90 dBm), marginal (−110 to −90 dBm) or weak (below −110 dBm).
5. The link-budget table derives maximum coupling loss and range per spreading factor with an 8 dB fade margin, giving the report its range figures.
6. The duty-cycle checker converts payload bytes, spreading factor and interval into airtime and flags any node over the 1% band rule.
7. The deployment report summarises gateway count, covered nodes, ADR plan and validation walk-test steps for the final report.

Tech stack:

• HTML5 canvas + JavaScript (interactive map, single-file app)
• Log-distance path-loss propagation model
• Semtech SX1276 datasheet sensitivity values
• CSS dashboard UI (dark theme)
• Report templates (deployment plan, walk-test sheet)

• [Interactive coverage map] Click to place gateways on a stylised site map; each node is coloured strong / marginal / weak by its predicted RSSI, recomputed live as you move gateways.
• [Link-budget calculator] Per-spreading-factor table (SF7–SF12) with SX1276 datasheet sensitivities, maximum coupling loss and range estimates for rural and urban environments at 14 dBm TX.
• [Environment presets] Path-loss exponents for rural (n=2.0), suburban (n=2.4) and urban campus (n=2.9) — documented so the student can explain the model.
• [Spreading-factor trade-off view] Shows why SF12 reaches farthest but costs airtime, and how ADR lets near nodes drop to faster spreading factors.
• [Duty-cycle checker] Flags the worst-case node against the 1% EU 868 MHz band rule from its payload size, spreading factor and reporting interval.
• [Deployment report] Gateway rollout list, coverage percentage, backhaul notes and a cost-vs-gateway diminishing-returns note for the project report.
• [Validation procedure] A walk-test method with a GPS-tagged node to compare the predicted RSSI map against measured values on the student's own site.

• LoRaPlan single-file web app (coverage map, link budget, deployment report)
• Propagation model documentation with worked numerical examples
• Three sample site plans (campus, farm, industrial estate)
• Duty-cycle and ADR planning worksheets
• Walk-test validation procedure template
• Project report PDF (background, LoRa PHY, link-budget theory, methodology)
• PPT presentation for final review
• Viva Q&A preparation document (spreading factors, link budget, ADR, duty cycle)

• The planner uses a statistical path-loss model, not ray tracing — buildings, foliage and terrain are approximated by the exponent, so predictions are estimates to be validated by survey.
• Range figures are model outputs with an 8 dB fade margin assumption, not measured claims; the report states this explicitly.
• Only the 868 MHz band model is preconfigured; 433 MHz or 915 MHz need re-parameterisation (procedure included).
• The tool plans coverage, not capacity — dense networks need a separate collision/airtime study, noted as future scope.
• No live hardware integration; it is a planning and teaching tool, not a network server.