Wi-Fi Mesh Sensing
Whole-Home Spatial Awareness — Natively Private, Inherently Scalable
nami’s Wi-Fi Mesh Sensing technology transforms standard Wi-Fi infrastructure into a
whole-home spatial awareness system.
By understanding how radio signals interact with people and spaces, nami delivers reliable motion, presence, and activity insights across large areas, especially across rooms not always fit for cameras, radars or invasive sensors.
Built on Physical Principles
Not Fragile Assumptions
Our sensing engines are grounded in physical reality, not brittle heuristics or one-off lab conditions.
- Wi-Fi signals naturally interact with the human body and the environment.
As people move, breathe, or change posture, they introduce subtle but measurable disturbances in radio propagation. - nami extracts structure from these fluctuations.
Our engines analyze fine-grained signal variations to identify human motion, true presence, and rest/activity patterns. - Sensing through walls, in darkness, by design.
Using standard Wi-Fi signals and existing infrastructure, nami senses across rooms and floors — day or night. - Privacy-first by nature.
Unlike cameras or microphones, Wi-Fi sensing captures only anonymous signal dynamics — never images, sound, or personally identifiable data.
Designed for Real Homes and Commercial Spaces
Not Static Test Environments
Homes evolve. Signals drift. Furniture moves. People behave unpredictably.
nami systems are designed to improve over time, not degrade.
- Adaptive Learning
Our data-centric AI continuously learns from new real-world examples, adjusting to environmental changes in real time. - Long-Term Robustness
Past errors are not discarded — they are reintegrated into the learning loop. This prioritizes durability and consistency over short-term benchmark accuracy.
How Wi-Fi Sensing Works
In a Nutshell
Channel State Information: Turning Radio Waves into Spatial Intelligence
Wi-Fi sensing is based on Channel State Information (CSI) — a low-level view of how radio signals propagate between a transmitter and a receiver.
1. Transmission
A Wi-Fi transmitter emits radio signals across multiple sub-carriers.
2. Interaction with Space & Humans
Walls, furniture, and the human body reflect, absorb, and diffract these signals.
Even minimal movement causes measurable changes in amplitude and phase.
3. Reception & CSI Extraction
The receiver captures these variations at a very fine temporal and frequency resolution.
4. Pattern Analysis
nami’s sensing engines extract structured patterns from CSI streams to infer motion, presence, and activity — without knowing who the person is.
Breaking It Down
From Radio Waves to Spatial Awareness
1
Wi-Fi Signals Fill the Space
Every Wi-Fi device continuously emits radio waves that travel through walls, furniture, and open spaces. These signals form a stable radio environment inside the home.
Unlike cameras or wearables, nothing needs to be worn, touched, or activated by the user.
2
Humans Naturally Disturb Radio Waves
The human body absorbs and reflects radio waves.
When a person moves, changes posture, or even breathes, they introduce tiny but measurable distortions in the Wi-Fi signal.
These disturbances are invisible to humans — but highly visible to radios.
3
Transmitter & Receiver Observe the Changes
- amplitude (strength)
- phase (timing)
- frequency response across subcarriers
4
CSI Is Converted into Motion States
- stable environments produce stable patterns
- human motion creates characteristic temporal changes
5
AI Interprets Human Activity
- presence vs absence
- short activity vs long-term behavior
- fast reaction for security
- long-term understanding for care and automation
Privacy by Design
Wi-Fi sensing never captures images, audio and personal identifiers.
This makes Wi-Fi sensing inherently privacy-preserving, regulation-friendly and socially acceptable in sensitive environments (bedrooms, bathrooms, elderly care)
Wi-Fi Sensing Engines integrated by nami
Aerial and Origin CSI Processing Engines in the nami Platform
Today, Aerial and Origin CSI processing engines coexist within the nami platform. Aerial engines unlock more specialized datasets, particularly for CareTech use cases. Whereas Origin technology serves existing and future customers partnering with Origin Wireless.
This dual-engine approach enables nami to deliver AIoT devices and fusion sensing systems while offering customers a choice between two engine families, each backed by strong IP.
Aerial PulseCSI
Security-Grade Motion Intelligence
PulseCSI is nami’s security-focused Wi-Fi sensing engine, designed for fast, reliable detection in real-world environments.
- Optimized for rapid human motion detection (typically within 6–8 seconds)
- Tuned for mesh-based sensing topologies, not single-link assumptions
- Resilience to non-human activity, environmental noise, and RF clutter
- Ideal for alarm enhancement, auto-arming logic, and intrusion detection
PulseCSI prioritizes speed, precision, and confidence, making it a natural complement to professional security systems.
Aerial ActiveCSI
Continuous Awareness for Care & Automation
ActiveCSI is optimized for longer-term behavioral insights, powering care and automation use cases.
- Tracks activity trends and presence over extended periods
- Filters short-term noise in favor of meaningful behavioral signals
- Feeds higher-level engines such as true occupancy / vacancy detection
- Enables proactive scenarios: care monitoring, energy optimization, and smart automation
ActiveCSI focuses on context, continuity, and understanding, rather than instant alerts.
CSI Processing Engines Powered by Aerial.ai
Aerial.ai is nami’s in-house Wi-Fi sensing research and engineering lab, dedicated to advancing the core CSI data processing and RF intelligence that power nami’s AI sensing engines.
In 2025, nami took a significant strategic shareholding in Aerial.ai, bringing its deep signal-processing expertise, patented CSI technologies, and advanced RF research fully into the nami technology stack. This close integration allows nami to control the sensing pipeline end-to-end — from raw Wi-Fi signals’s extraction, to production-grade AI engines deployed at scale in real homes and buildings.
Through Aerial.ai, nami continuously evolves its CSI processing engines (PulseCSI and ActiveCSI) to meet the stringent requirements of security, care, and automation use cases — ensuring accuracy, resilience, privacy, and long-term scalability.
Together, nami and Aerial.ai form a single, vertically integrated sensing platform, backed by strong IP — where fundamental RF science meets real-world deployment.
Origin CSI Engine
Proven RTOS-Grade Foundation
nami also integrates and deploys Origin CSI engines directly on its RTOS-based devices.
- Production-proven, with multi-year deployments of nami devices powered by Origin Engine
- Optimized for low power consumption, embedded constraints, and long-term stability
- Origin-powered Wi-Fi sensors are also enhanced by nami’s mesh, multi-node sensing architecture.

Why Mesh Topology Changes Everything
Spatial Awareness Requires Spatial Infrastructure
Most Wi-Fi sensing solutions rely on star or point-to-point topologies — a single transmitter-receiver pair attempting to infer motion across an entire home.
nami takes a fundamentally different approach.
With nami’s multi-node mesh topology, each device acts as both a sensing node and a network participant. Such topology has been patented by nami, independently of which CSI processing engine vendor is used.
Multi-Node Mesh Sensing:
A Native Advantage
Extended Coverage
Multiple sensing paths cover large homes, corridors, and multi-floor layouts where single-link systems fail.
True Spatial Diversity
Motion is observed from multiple angles, improving reliability and reducing blind spots.
Graceful Degradation
If one link becomes noisy or obstructed, others continue to provide signal continuity.
Environment Agnostic
Works reliably in long apartments, complex layouts, dense urban housing, and RF-challenged environments.
Scalable by Design
Coverage and accuracy improve simply by adding nodes — no recalibration required.



















