English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

There is much interest in incorporating inference capabilities intosensor-rich embedded platforms such as autonomous vehicles, wearables, andothers. A central problem in the design of such systems is the need to extractinformation locally from sensed data on a severely limited energy budget. Thisnecessitates the design of energy-efficient sensory embedded system. A typicalsensory embedded system enforces a physical separation between sensing andcomputational subsystems - a separation mandated by the differing requirementsof the sensing and computational functions. As a consequence, the energyconsumption in such systems tends to be dominated by the energy consumed intransferring data over the sensor-processor interface (communication energy)and the energy consumed in processing the data in digital processor(computational energy). In this article, we propose an in-sensor computingarchitecture which (mostly) eliminates the sensor-processor interface byembedding inference computations in the noisy sensor fabric in analog andretraining the hyperparameters in order to compensate for non-idealcomputations. The resulting architecture referred to as the Compute Sensor - asensor that computes in addition to sensing - represents a radical departurefrom the conventional. We show that a Compute Sensor for image data can bedesigned by embedding both feature extraction and classification functions inthe analog domain in close proximity to the CMOS active pixel sensor (APS)array. Significant gains in energy-efficiency are demonstrated using behavioraland energy models in a commercial semiconductor process technology. In theprocess, the Compute Sensor creates a unique opportunity to develop machinelearning algorithms for information extraction from data on a noisy underlyingcomputational fabric.

Reducing the Energy Cost of Inference via In-sensor Information Processing