A health- and link-aware adaptive data-rate method for enhancing quality of service in wireless body area networks
Wireless Body Area Networks (WBANs) must provide reliable, low-latency, and energy-efficient communication despite rapid changes in physiological urgency, body-induced channel variation, and heterogeneous sensor quality. Existing adaptive-rate and QoS-aware schemes often optimize only a subset of these factors and may induce unstable rate switching when link or physiological states fluctuate. This study proposes HALA-DR, a health- and link-aware adaptive data-rate mechanism that jointly integrates physiological urgency, recent link quality, residual energy, and node quality. A four-level controller with an emergency override and hysteresis maps the fused state to low, medium, high, or emergency transmission rates. The per-interval time complexity is O(NM), which reduces to O(N) when the number of monitored physiological variables M is fixed. A reproducible MATLAB evaluation was conducted using an IEEE 802.15.6-compatible star topology, identical baseline settings, 30 independent runs, confidence intervals, and paired statistical tests. In the 15-node scenario, HALA-DR achieved 236 kbps throughput, a 95.1% packet delivery ratio, 45 ms end-to-end latency, 0.94 mJ per successfully delivered packet, and 118 h network lifetime. Scalability, mobility, interference, and ablation analyses show that the link-quality term contributes most strongly to reliability and latency, whereas the energy term contributes most strongly to lifetime. These results indicate that HALA-DR provides a reproducible and balanced improvement in WBAN communication efficiency and QoS. Not applicable.
Authors
- B. Ye
- Qiliang Liu (ORCID: https://orcid.org/0000-0002-2548-7359)
- Wei Zhou
Institutions
- Hubei University of Arts and Science (CN)
Publication Details
- Journal
- Discover Computing
- Published
- 2026-09-15
- DOI
- https://doi.org/10.1007/s10791-026-10513-2
- Primary Topic
- Wireless Body Area Networks
- Type
- article
- Field-Weighted Citation Impact
- 0.00