Juq-325 Page

| Specification | Detail | |---------------|--------| | On‑Board AI Accelerator | 8‑core NPU delivering up to 2 TOPS (tera‑operations per second) for mixed‑precision inference. | | Edge OS | Hardened Linux‑based OS with container support (Docker / OCI) for easy deployment of custom models. | | Model Update Pipeline | OTA (over‑the‑air) model delivery with version rollback and integrity verification via signed manifests. | | Sensor Fusion | Native support for up to 12 simultaneous sensor streams (IMU, LiDAR, camera, temperature, etc.). | | Secure Boot & TPM 2.0 | Guarantees that only authenticated firmware and models run on the device. |

| Industry | Example Application | |----------|----------------------| | Manufacturing | Detect equipment wear in real time, trigger preventive maintenance before failure. | | Smart Buildings | Optimize HVAC and lighting based on occupancy patterns without sending personal data to the cloud. | | Healthcare | Run on‑device ECG or imaging analysis at the bedside, delivering instant alerts while keeping patient data local. | | Logistics | Dynamically reroute autonomous forklifts around obstacles using on‑board perception. | | Retail | Provide instant, privacy‑preserving customer behavior insights for in‑store promotions. | juq-325

| Workload | Baseline (ARM Cortex‑A78, 5 W) | JUQ‑325 (Full Heterogeneous) | Energy‑Delay Product (EDP) Improvement | |----------|-------------------------------|------------------------------|----------------------------------------| | MobileNet‑V2 | 3.2 ms latency, 4.1 J energy | 2.1 ms, 3.5 J | 1.8× | | BERT‑tiny | 12.4 ms, 9.8 J | 6.7 ms, 7.2 J | 2.1× | | GNN (traffic) | 28.9 ms, 18.0 J | 15.3 ms, 12.3 J | 2.4× | | | Sensor Fusion | Native support for

The most pronounced gains appear in workloads that heavily rely on sampling or combinatorial optimization (BERT‑tiny and GNN), confirming the efficacy of the quantum kernels. Power profiling shows that the QCP consumes on average 0.9 W during active phases, with idle power under 0.1 W thanks to an aggressive voltage‑scaling scheme. | | Smart Buildings | Optimize HVAC and