Comparative Analysis of the Uplink Bandwidth Allocation in Wireless Sensor Networks: DRPA and FL-WFQ Case Study
Abstract
An intelligent utilization of resources in wireless sensor networks (WSNs) significantly impacts overall energy efficiency and communication performance in constrained sensor nodes. Weighted Fair Queuing (WFQ) is commonly used to allocate uplink bandwidth according to queue weights. Recently, Federated Learning (FL) has been combined with WFQ to dynamically predict optimal queue weights under varying traffic and mobility conditions. This paper presents a comparative analysis of two allocation approaches: (i) Data-Rate-Proportional Allocation (DRPA) and (ii) Federated Learning-enabled WFQ (FL-WFQ). Simulation results show that FL-WFQ improves efficiency by up to 119.22% (average 49.38%) and fairness by up to 54.79% compared to DRPA under dynamic mobility scenarios.
Keywords
wireless sensor networks, edge computing, federated learning, weighted fair queuing
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
S. Pallantla and D. Haritha, "Comparative Analysis of the Uplink Bandwidth Allocation in Wireless Sensor Networks: DRPA and FL-WFQ Case Study," in Journal of Communications Software and Systems, vol. 22, no. 4, pp. 599-608, August 2026, doi: 10.24138/jcomss-2025-0297
@article{pallantla2026comparativeanalysis,
author = {Pallantla, Sravya and Haritha, D.},
title = {{Comparative Analysis of the Uplink Bandwidth Allocation in Wireless Sensor Networks: DRPA and FL-WFQ Case Study}},
journal = {Journal of Communications Software and Systems},
month = aug,
year = {2026},
volume = {22},
number = {4},
pages = {599--608},
doi = {10.24138/jcomss-2025-0297},
url = {https://doi.org/10.24138/jcomss-2025-0297}
}