Interior Time-Frequency Domain Sensor Positioning in Strong Mobility-Oriented Human-Centric WSNs

Qian Dong*, Jinbao Xia*, Jianjun Wen, Mi Lu

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


The strong mobility characteristic of wireless sensor nodes has attracted extensive attention because it can lead to real-time changes in node locations and the attenuation, instability, and even disruption of link quality. Focusing on these issues, this paper studies the reliability of adopting received signal strength indication (RSSI) technology to locate mobile nodes in a human-centric interior environment. To this end, this paper conducts dynamic and static experiments, analyzes the spatiotemporal performance of mobile RSSI, investigates the influence of a human body on link quality, draws the standard curve to show the one-to-one correspondence between distance and RSSI in a static scene, and proposes six denoising methods to alleviate severe RSSI fluctuation in a dynamic scene. The denoising effect is verified by comparing the consistency between the denoised mobile RSSI and static standard RSSI. The comparison is achieved by computing the root mean square error (RMSE) of RSSI in the time domain and transforming the result into the noise fast Fourier transform (FFT) spectrum in the frequency domain. Though the RMSE is reduced by 39.7% using the overall optimized filtering method, while the amplitude of noise FFT is reduced to an average of 0.02 dBm, due to the non-monotonic decrease of RSSI when the distance increases, one RSSI may correspond to multiple distances, with these distances even differing by 12.8 m. Because this number is too large for most applications, the human-centric sensor positioning using only RSSI technology is unreliable in mobile interior wireless sensor networks.

Original languageEnglish
Article number36
JournalHuman-centric Computing and Information Sciences
Publication statusPublished - 2023


  • Denoising
  • Filtering
  • Mobility
  • RSSI Localization
  • Sensor positioning
  • Wireless Sensor Networks


Dive into the research topics of 'Interior Time-Frequency Domain Sensor Positioning in Strong Mobility-Oriented Human-Centric WSNs'. Together they form a unique fingerprint.

Cite this