Higher order statistics (HOS) in signal processing: are they of any use?

Sabry-Rizk, M.; Romare, D.; Zgallai, Walid A.; Grattan, Kenneth T. V.; Hardiman, Paul; O’Riordan, J. L. H.

doi:10.1049/ic:19950729

Cited by 5 publications

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to pick precise onset time of ultrasonic further, higher order statistics method, which is an effective approach for processing non-Gaussian signals, is utilized to deal with this task. In this paper, kurtosis that is fourth-order cumulants was chosen as characteristic function because it is insensitive to the Gaussian noise and can achieve a higher accuracy with a smaller order and less calculation time [25,26].…”

Section: Kurtosis Characteristic Function Calculationmentioning

confidence: 99%

Determining Ultrasound Arrival Time by HHT and Kurtosis in Wind Speed Measurement

Liu

et al. 2021

Electronics

View full text Add to dashboard Cite

The determination of ultrasonic echo signal onset time is the core of performing the time difference method to calculate wind speed. However, in practical cases, background noise makes precise determination extremely difficult. This paper carries out research on the accurate determination of onset time, exploring the advantages of an improved method based on the combination of Hilbert-Huang Transform (HHT) and high-order statistics (kurtosis). Performing Hilbert-Huang Transform to the received wave is aimed at determining a rough arrival time, around which a fixed size of data is extracted as initial sample to avoid a false pick. Then the fourth-order kurtosis of a smaller sample, extracted successively by a moving window from the initial sample, is calculated. The minimum point corresponds to the initial onset time. This approach was tested on a real ultrasonic echo signal dataset, acquired in a wind tunnel with an ultrasonic anemometer. The proposed method showed satisfying results in both ideal cases and low signal-to-noise ratio (SNR) environment, compared with traditional onset time determination approaches, including Akaike Information Criterion (AIC-picker), Short-term Average over Long-term Average (STA/LTA), and Teager-Kaiser energy operator (TKEO). The experimental results acquired by the HHT-kurtosis method demonstrated that the proposed method possesses a high accuracy.

show abstract

Section: Kurtosis Characteristic Function Calculationmentioning

confidence: 99%