Approaches to human activity recognition via passive radar

TL;DR

This paper presents a novel, privacy-preserving approach to human activity recognition using passive radar and Wi-Fi CSI data, employing spiking neural networks and symbolic reasoning for improved accuracy and interpretability.

Contribution

It introduces the use of SNNs combined with DeepProbLog for passive radar-based HAR, enhancing adaptability and reducing power consumption.

Findings

High accuracy achieved with SNN-based models

Effective non-intrusive HAR using Wi-Fi CSI data

Enhanced interpretability through neurosymbolic integration

Abstract

The thesis explores novel methods for Human Activity Recognition (HAR) using passive radar with a focus on non-intrusive Wi-Fi Channel State Information (CSI) data. Traditional HAR approaches often use invasive sensors like cameras or wearables, raising privacy issues. This study leverages the non-intrusive nature of CSI, using Spiking Neural Networks (SNN) to interpret signal variations caused by human movements. These networks, integrated with symbolic reasoning frameworks such as DeepProbLog, enhance the adaptability and interpretability of HAR systems. SNNs offer reduced power consumption, ideal for privacy-sensitive applications. Experimental results demonstrate SNN-based neurosymbolic models achieve high accuracy making them a promising alternative for HAR across various domains.