UoE-Huawei Tech Talk Series: Reality or Sci-Fi: Zero-Calibration Indoor Positioning, by Dr. Firas Alsehly

Abstract

The recent development of IoT and smart spaces demand the availability of location context in every device and everywhere. However, improving the user’s experience under the “smart” schema contradicts the tedious calibration commonly operated to enable indoor positioning. Until now, modelling signal propagation, such as Bluetooth and WiFi, set the pillars for indoor positioning while visual features mapping continues to grow as rival technology. Nevertheless, power consumption, calibration and costly site survey create barriers for any algorithm targeting smartphone-centric deployment. While prescribed data collection or beacon installation appeared simple and feasible solutions, operational overheads prevented its globalisation. Hence, investing in research that improves operational resiliency, automate data pipelines, reduce maintenance overheads and scale to global deployment became the trend in this era.

In this talk, we discuss IEEE 802.11 wireless networking signal propagation under the lenses of data science. We aim to entertain the concept of zero-calibration indoor positioning by describing a hybrid cooperative framework with a spectrum of technologies. The talk will demonstrate the impact of proximity estimation in geospatial communities and the influence of signal attenuation. We will try to simplify the use of some well-known techniques such as signal propagation modelling, regression, data cleansing, clustering & classification feeding into self-governing survey automation to extract radio-maps. Finally, we will shed some light on the gap between reality and academic research in this area.

About the speaker

Dr. Firas Alsehly is the leader of the Indoor Positioning and Navigation Lab in Huawei Edinburgh Research Center. He sums over 10 years of R&D leadership & experience in geospatial data analysis delivering global scale positioning algorithms and data pipeline for unsupervised learning. In addition to his PhD in signal processing and adaptive systems from the University of Edinburgh, his previous experience extend to Fin-Tech, high availability systems, micro-services and performance optimisation.

Date & Time: Friday, 5th Nov 2021, 12:45pm (UK)

Zoom Meeting Link: https://ed-ac-uk.zoom.us/j/81574574945

Meeting ID: 815 7457 4945

Passcode: Dtd8hmq0