Energy Detection Based Estimation of Channel Occupancy Rate with Adaptive Noise Estimation

Abstract

Recently, there has been growing interest in opportunistically utilizing the 2.4 GHz ISM-band. Numerous spectrum occupancy measurements covering the ISM-band have been performed to analyze the spectrum usage. However, in these campaigns the verification of the correctness of the obtained occupancy values for the highly dynamic ISMband has not been presented. In this paper, we propose and verify channel occupancy rate (COR) estimation utilizing energy detection mechanism with a novel adaptive energy detection threshold setting method. The results are compared with the true reference COR values. Several different types of verification measurements showed that our setup can estimate the COR values of 802.11 traffic well, with negligible overestimation. The results from real-time real-life measurements also confirm that the proposed adaptive threshold setting method enables accurate thresholds even in the situations where multiple interferers are present in the received signal.

Performance of the ETSI SmartBAN system in the interfered IEEE 802.15.6 channel

“…By using the channel occupancy data, a mathematical model of the interference was formulated for the frequency band SmartBAN operates. The detailed insight to the measurement campaigns and data analysis procedures are given in [7]- [10]. Table 1 gives the parameters applied in two measurement campaigns carried out in a daily surgery ward (DS) and accident & emergency ward (AEW).…”

ETSI TC SmartBAN: Overview of the wireless body area network standard

“…The impact of the interference is evaluated by three types of occupancy metrics which are calculated from the measured data: channel occupancy (CO), frequency band occupancy (FBO) and spectrum resource occupancy (SRO) [9]. These metrics allow to realize which is the intensity of the interference, its spreading over time and frequency, the most occupied hours per environment, the most occupied frequencies, the nature of the interference (bursty or continuous), etc.…”

Measurements and analysis of spectrum occupancy with several bandwidths

“…Even if part of the channel is occupied, the channel should be counted as occupied from the opportunistic radio point of view. Otherwise we may harm other users [7]. The proposed aggregation method to form a new binary matrix bn with 86400 columns for the analysis in nonoverlapping case is given as…”

“…The mentioned measurements have been performed with a fixed parameter setting, i.e., using the same detection threshold and measurement bandwidth for a certain band and location. Adaptive parameter settings regarding the threshold [7], resolution bandwidth and sweep time [8] have been proposed to increase the accuracy in the occupancy measurements.…”