Threshold Optimization For Energy Detection in Spectrum Sensing of Wireless Sensor Networks
Efficient spectrum utilization is critical for the evolving landscape of wireless communication. This paper presents energy detectors (ED) as a fundamental technique for identifying spectrum holes by examining their performance under different noise levels from data derived from USRP in 2.4 GHz environment. A theoretical model is developed to analyse the influence of noise variance, threshold, signal-to-noise ratio (SNR), and the number of samples on detection performance. Using a noise floor reference of –65 dBm, a noise variance of 9.99 × 10⁻²² is derived, which resulted in 7.38% spectrum usage and 92.62% availability which resulted in significant opportunities for secondary user access. Further empirical evaluation using USRP data revealed that lowering the noise floor improved detection sensitivity and Pd, due to the lower noise variance, which raises the effective SNR, with Pd = 0.6820 with N = 8000 at Pfa = 0.01. The results obtained show detection performance is highly dependent on accurate noise floor estimation, threshold tuning, and sample size. Lower noise floors led to improved sensitivity for better detection of weak signals and thus improved spectrum utilization. These findings highlight the need for adaptive and noise-aware spectrum sensing strategies in hybrid detection systems for dynamic spectrum access.
Authors
- Nor Khairiah Ibrahim (ORCID: https://orcid.org/0000-0002-1692-8140)
- Nur Hanis Abdul Rani (ORCID: https://orcid.org/0009-0009-1743-9048)
- Mas Haslinda Mohamad
Institutions
- Technical University of Malaysia Malacca (MY)
- University of Kuala Lumpur (MY)
Publication Details
- Journal
- International Journal of Research and Innovation in Social Science
- Published
- 2026-11-16
- DOI
- https://doi.org/10.47772/ijriss.2026.101000701
- Primary Topic
- Cognitive Radio Networks and Spectrum Sensing
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Universiti Teknikal Malaysia Melaka