Distributed On-Sensor Compute System for AR/VR Devices: A Semi-Analytical Simulation Framework for Power Estimation

TL;DR

This paper presents a semi-analytical simulation framework for power estimation of a novel distributed on-sensor compute architecture in AR/VR glasses, enabling optimization of system modules and algorithms to reduce power, latency, and enhance privacy.

Contribution

It introduces a new simulation framework for power estimation and optimization of distributed on-sensor architectures in AR/VR devices, integrating hardware-software co-design.

Findings

Distributed architecture reduces power consumption for machine learning hand tracking.

The framework enables optimization of system modules and algorithm partitioning.

Distributed system improves latency and privacy over centralized systems.

Abstract

Augmented Reality/Virtual Reality (AR/VR) glasses are widely foreseen as the next generation computing platform. AR/VR glasses are a complex "system of systems" which must satisfy stringent form factor, computing-, power- and thermal- requirements. In this paper, we will show that a novel distributed on-sensor compute architecture, coupled with new semiconductor technologies (such as dense 3D-IC interconnects and Spin-Transfer Torque Magneto Random Access Memory, STT-MRAM) and, most importantly, a full hardware-software co-optimization are the solutions to achieve attractive and socially acceptable AR/VR glasses. To this end, we developed a semi-analytical simulation framework to estimate the power consumption of novel AR/VR distributed on-sensor computing architectures. The model allows the optimization of the main technological features of the system modules, as well as the computer-vision algorithm partition strategy across the distributed compute architecture. We show that, in the case of the compute-intensive machine learning based Hand Tracking algorithm, the distributed on-sensor compute architecture can reduce the system power consumption compared to a centralized system, with the additional benefits in terms of latency and privacy.