How Canopy NOC Works
A continuous operational loop — from raw LoRaWAN data to prioritized action — powered by AI at every step.
The Operational Loop
A continuous 24/7 cycle — from raw network data to prioritized action.
Collect
Ingest raw network data close to the gateways and LNS through direct integrations, HTTP endpoints, or the public API.
Normalize
Transform data into a shared model of networks, gateways, devices, packets, and statistics — regardless of source.
Compute
Continuously compute KPIs that describe service quality, RF coverage, capacity, and infrastructure health.
Detect
Evaluate KPIs against alarm thresholds with severity levels and flapping suppression to surface what matters.
Investigate
Drill into maps, dashboards, and AI-generated context to understand every detected condition in full detail.
Act
Follow AI recommendations with full operational evidence — resolve issues with confidence and close the loop.
Data Ingestion
LNS Integrations
Direct connections to LoRaWAN Network Servers through implementation-specific channels. This is the preferred path — providing the richest view of uplinks, downlinks, gateway metadata, gateway statistics, device identity, and LNS-side context.
HTTP Ingress
Network-specific HTTP endpoints that receive pushed payloads from an LNS or routing layer. Configurable decoder scripts interpret inbound payloads into the platform’s normalized format — useful when a full LNS integration isn’t available.
API Ingress
API-based ingestion for external collectors or custom middleware. The public Kudzu analytics repository contains protobuf definitions and examples for collecting packet-forwarder traffic, including Semtech UDP and LoRa Basics Station deployments.
Manual & Imported Data
Complement live telemetry with managed gateway and device inventory, network zones, frequency plans, deployment simulations, survey files, and GeoTIFF layers — so the platform reasons about both the live network and the intended design.
Problem Taxonomy
Organized from Broad Concern to Specific Evidence
Canopy NOC organizes analysis algorithms, metrics, alarms, and AI insights into an operational tree. Users can browse from a broad operational concern to specific evidence — opening a sub-problem, inspecting active alarms, drilling into a KPI chart, viewing affected devices or map cells, and reading AI-generated recommendations.
Quality of Service
Uplink loss, silent devices, join behavior, downlink delivery, and service-level adherence across the network.
Coverage & RF Quality
Weak coverage zones, low-redundancy areas, RF degradation, signal quality distribution, and spreading factor trends.
Capacity & Performance
Airtime saturation, duty-cycle utilization, channel distribution, and capacity bottlenecks that could affect service delivery.
Infrastructure & Connectivity
Gateway availability, connectivity drops, unusual infrastructure behavior, and integration health across all data sources.
Compliance, Security & Operational Integrity
Automated compliance checks, Certificate of Nominal Operation, regulatory reporting, ADR compliance, and SLA adherence documentation.
AI-Driven Operations
Action Center
The first screen users see after login. It combines a color-coded network status banner, device/gateway/alarm overview widgets, and AI-generated optimization suggestions. Each suggestion includes a problem summary, proposed solutions with feasibility scores, implementation guidance, and references to affected entities.
Network Insights
AI-generated structured findings about network conditions. Each insight includes the problem, proposed solutions with summaries, supporting evidence references, and actionable recommendations. Users can inspect, validate, dismiss, or act on every insight.
Nomi — AI Assistant
The built-in AI chatbot accessible from anywhere in the interface. Ask questions about any metric, alarm, gateway, device, or network condition. Nomi answers with full operational context — evidence, map references, metric history, and recommendations.
Automated Reports
Generated network documents that summarize a reporting period. Reports can include AI-written analysis, supporting charts, and evidence datasets. Use them for compliance, stakeholder communication, or internal operational reviews.
Deployment Options
Cloud (Default)
Fully hosted and managed by Kudzu. No infrastructure to maintain — sign up and connect your network. Automatic updates, scaling, and redundancy included.
Dedicated
A dedicated Canopy NOC instance running in an isolated cloud environment. For customers who need separation from the shared platform but don’t require on-premises hardware.
On-Premises
Full Canopy NOC deployment on your own infrastructure. For organizations with strict data residency, air-gap, or regulatory requirements. Includes engineering consultation for deployment planning and integration setup.
Ready to get started?
Every plan includes the full Canopy NOC platform. Dedicated and on-premises deployments are available with the Enterprise plan.