Architecture

Four layers, one on-device closed loop

NeuroNAS is built on a "four chips, three zones" layered architecture. Capture, Transport, Decision and Application each stand alone yet work in seamless concert, closing the loop from EEG signal capture to whole-home intent control. The entire pipeline runs without ever touching the cloud.

① Application Layer · Application
Lights off by thought

intent → device

Focus training

theta/beta neurofeedback

AI sleep intervention

acoustic closed-loop control

Brain-controlled gaming

SDK ecosystem

Cognitive training

cognition

▼ Decision
② Decision Layer · Home AI NAS Hub
Neural LLM engine (on-device)

Lightweight Transformer + SNN hybrid inference (≤100M params/model)
Federated learning framework (local iteration + anonymous aggregation)
Neuromorphic chips: AKD1500 / Innatera Pulsar

Multimodal data processor (private cloud + edge AI)

Home neural-data private cloud (encrypted local storage + hot/cold tiering)
Home-grade on-device AI inference (NPU / Rockchip edge silicon)
Family member profiles + cross-generational neural-data linkage

▼ Transport
③ Transport Layer · Local Home Network
Bluetooth 5.4 / BLE

accessory connectivity

Wi-Fi 6 / Thread

home device connectivity

▼ Capture
④ Capture Layer · Wearable Accessories
Focus headband

8–16 channels · graphene dry electrodes

EEG sleep earphones

4–8 channels · behind-the-ear dry electrodes

Brain-controlled controller

2–4 channels · head EEG

Core Advantages

Why only on-device can deliver this

On-device inference

EEG data runs every AI inference locally on the NAS, keeping privacy 100% under your control. On-device processing cuts latency by over 60% versus the cloud.

Neuromorphic chips

The BrainChip AKD1500 uses an event-driven architecture, firing computation only on "spike" activity, so standby power is near zero and delivers 24/7 continuous sensing. The Innatera Pulsar cuts power by roughly 500×.

Matter support

No cloud required: thought-as-command runs through a local Matter gateway to control whole-home devices directly, and works offline. A Thread border router plus Zigbee bridging delivers seamless cross-brand orchestration.

Data flywheel: the focus headband captures theta/beta rhythms while the sleep headband captures Delta/SWM slow waves, two highly complementary neural signals converging on a single NAS. The cross-scenario dataset trains a home health prediction model all our own. Competitors would have to develop both hardware lines at once and amass millions of users' data, a 3–5 year effort at minimum.

Want the technical deep dive? See the specs

The "four chips, three zones" mainboard, the full chip list, channel counts and protocol stack, all out in the open.

View hardware specs →