EventsThe 1st International Online Conference on Technologies
Published
This submission belongs to the session S1. Information and Communication Technologies of the event The 1st International Online Conference on Technologies
Published date
06 Nov, 2026
Academic Editor
author-avatarManoj Gupta
Citation
Volodymyr Pavlenko, Viktoriia Danylova, QoS-Aware Cooperative Multi-AP Wi-Fi Sensing for Coarse Indoor Awareness in OpenWrt-Based WLANs, in Proceedings of The 1st International Online Conference on Technologies, 11 November–13 November 2026, MDPI: Basel, Switzerland
Share
Email
Facebook
Twitter
LinkedIn

QoS-Aware Cooperative Multi-AP Wi-Fi Sensing for Coarse Indoor Awareness in OpenWrt-Based WLANs

Viktoriia Danylova 1
1. Department of Electronic Computing Machines, Institute of Computer Technologies, Automation and Metrology, Lviv Polytechnic National University, Lviv 79013, Ukraine
Abstract

Introduction: Wi-Fi sensing infers indoor changes from radio propagation, but single-link measurements are interference-sensitive. Multi-AP CSI fusion (Ding et al., 2023) and ReWiS combine CSI views for recognition; EfficientFi compresses CSI, whereas SenCom reuses communication packets. Unlike these systems and the published CSAN reliability/QoS workflow, this contribution is a design specification for a CSI-optional, QoS-aware OpenWrt controller; it is not an implemented or deployed system. Its distinction is one interface that normalizes native, companion-assisted, or absent CSI while exposing reliability and acquisition cost to admission, rather than a new fusion or recognition model.

Methods: The specification defines three routers, six telemetry classes (CSI, RSSI, packet counters, retry/loss, PHY-rate variation, and QoS/traffic load), and four output labels: stable/empty, presence/motion, static obstruction/environmental change, and uncertain. A deterministic fixture-based software check exercised all acquisition modes and CSI-loss fallback using no measured room or WLAN inputs.

Results: The deterministic fixture-based check validated use of all three router descriptors and retention of five conventional telemetry classes during complete CSI loss. This establishes interface and code-path consistency only, not sensing accuracy, WLAN performance, or OpenWrt integration. The provisional 10 s cycle trades response speed for a nominal 100-sample window per router at 10 samples/s and one decision per cycle, suiting coarse rather than real-time awareness. The provisional 5% ceiling is a single-digit admission bound chosen to leave 95% of modeled capacity outside sensing. Under the payload-only assumptions—three routers, 512-byte samples, 128 control bytes per router per cycle, and 6 Mb/s—the load is 2.05%, leaving 2.95 percentage points of headroom. These are design inputs, not measured latency, airtime, or QoS guarantees.

Conclusions: The contribution is a proposed controller method specified through an internally consistent contract that unifies CSI-present and CSI-free modes with explicit cost accounting. Hardware deployment, measured WLAN impact, and room-state accuracy remain future work.

Keywords
Wireless LAN (IEEE 802.11)
Wi-Fi sensing
Network controller
Resource allocation
Quality of service
Received signal strength indicator
Channel state information
Beyond Transactions: Leveraging Mobile Banking Technologies to Enhance Customer Satisfaction and Loyalty in Morocco
Explanation-Guided Rule Revision for Adaptive Symbolic Controllers