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

The goal of this work is the accurate prediction of millimeter-wave receivedpower leveraging both radio frequency (RF) signals and heterogeneous visualdata from multiple distributed cameras, in a communication and energy-efficientmanner while preserving data privacy. To this end, firstly focusing on dataprivacy, we propose heteromodal split learning with feature aggregation(HetSLAgg) that splits neural network (NN) models into camera-side and basestation (BS)-side segments. The BS-side NN segment fuses RF signals anduploaded image features without collecting raw images. However, the usage ofmultiple visual data leads to an increase in NN input dimensions, which givesrise to additional communication and energy costs. To overcome additionalcommunication and energy costs due to image interpolation to blend differentframe rates, we propose a novel BS-side manifold mixup technique that offloadsthe interpolation operations from cameras to a BS. Subsequently, we confrontenergy costs for operating a larger size of the BS- side NN segment due toconcatenating image features across cameras and propose an energy-efficientaggregation method. This is done via a linear combination of image featuresinstead of concatenating them, where the NN size is independent of the numberof cameras. Comprehensive test-bed experiments with measured channelsdemonstrate that HetSLAgg reduces the prediction error by 44% compared to abaseline leveraging only RF received power. Moreover, the experiments show thatthe designed HetSLAgg achieves over 20% gains in terms of communication andenergy cost reduction compared to several baseline designs within at most 1% ofaccuracy loss.

Distributed Heteromodal Split Learning for Vision Aided mmWave Received Power Prediction