Answers are grounded in this paper and cite their sources.
The emergence of the Internet of Things (IoT), driven by applications such as smart cities and smart homes, demands wireless communication capable of handling large data volumes with efficiency. Low-powered area network (LPWAN) technologies such as LoRaWAN offer long-range, low-power connectivity, using adaptive transmission based on signal-to-noise ratio (SNR). However, increasing device density in unlicensed bands causes interference and network congestion. Cognitive radio frameworks address this problem by enabling dynamic parameter adaptation. In this paper, we propose a weighted moving average (WMA) and low-pass filter (LPF) mechanisms to enhance SNR estimation, reducing packet loss in LoRa networks and applications. We evaluated the feasibility of the proposed WMA and LPF mechanisms in a controlled experiment in a testbed environment. The results demonstrated the WMA reduced at 5.65% the network reconfiguration and improved the SNR at 39.09% compared to Sliding Change. The results also demonstrated that LPF improved the balance between network adaptability and stability compared to regression-based LR-ADR. LPF achieved a 59.3% higher average SNR gain post-reconfiguration, and reduced the number of frequency band shifts by 8.53% compared to LR-ADR. By smoothing transient noise while remaining responsive to persistent channel degradation, LPF enhanced link reliability, minimizing disruptive parameter changes.
No takes yet. Share an insight, caveat, or question.
Usman et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: