Correlation-based dynamic sampling for online high dimensional process monitoring

Nabhan, Mohammad; Mei, Yajun; Shi, Jianjun

doi:10.1080/00224065.2020.1726717

Cited by 15 publications

(4 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This observation led to the use of the non-parametric Spearman correlation test in our study (Menebo 2020 ; Şahin 2020 ; Tosepu et al 2020 ). In the correlation test, two indicators are calculated: the r -coefficient to indicate whether the correlation is positive or negative and the p value to reveal the significance of the correlation (Nabhan et al 2021 ). Note that the correlation tests were conducted with a 95% significance level.…”

Section: Correlation Analysismentioning

confidence: 99%

Impact of climate indicators on the COVID-19 pandemic in Saudi Arabia

Abdel-Aal

Eltoukhy

Nabhan

et al. 2021

Environ Sci Pollut Res

View full text Add to dashboard Cite

Section: Correlation Analysismentioning

confidence: 99%

Impact of climate indicators on the COVID-19 pandemic in Saudi Arabia

Abdel-Aal

Eltoukhy

Nabhan

et al. 2021

Environ Sci Pollut Res

View full text Add to dashboard Cite

“…A numerical procedure was proposed to approximate the optimal but intractable Bayesian solution based on the approximated posterior probabilities. Non-Bayesian change-point detection with spatial information under different setups has been studied in [24]- [33]. In [27], a setting where a moving anomaly affects one sensor at a time was considered.…”

Section: Introductionmentioning

confidence: 99%

Bayesian Quickest Detection of Propagating Spatial Events

Halme

Nitzan

Koivunen

2022

IEEE Trans. Signal Process.

View full text Add to dashboard Cite

Rapid detection of spatial events that propagate across a sensor network is of wide interest in many modern applications. In particular, in communications, radar, IoT, environmental monitoring, and biosurveillance, we may observe propagating fields or particles. In this paper, we propose Bayesian sequential single and multiple change-point detection procedures for the rapid detection of such phenomena. Using a dynamic programming framework we derive the structure of the optimal single-event quickest detection procedure, which minimizes the average detection delay (ADD) subject to a false alarm probability upper bound. The multi-sensor system configuration is arbitrary and sensors may be mobile. In the rare event regime, the optimal procedure converges to a more practical threshold test on the posterior probability of the change point. A convenient recursive computation of this posterior probability is derived by using the propagation characteristics of the spatial event. The ADD of the posterior probability threshold test is analyzed in the asymptotic regime, and specific analysis is conducted in the setting of detecting random Gaussian signals affected by path loss. Then, we show how the proposed procedure is easy to extend for detecting multiple propagating spatial events in parallel in a multiple hypothesis testing setting. A method that provides strict false discovery rate (FDR) control is proposed. In the simulation section, it is demonstrated that exploiting the spatial properties of the event decreases the ADD compared to procedures that do not utilize this information, even under model mismatch.

show abstract

“…Hence lemma 1 is proved.To further prove P1, we focus on the difference between local statistics W i,j i max − W i,j for ∀j ∈ U , where the increment value depends on ξ * m,j i max u min and ξ * m,j u min (we consider the positive CUSUM without loss of generality). Thus, it suffices to investigate (Nabhan et al, 2021Note both ξ * m,j and ξ * m,j i max asymptotically follow N (0, 1), thus their expectations are both 0. However, the term ξ * m,j i max − ξ * m,j are not necessarily identical in terms of m. This is because there might be some ms that ξ * m,j i max are jointly sampled with ξ * m,j (unobservable for the j i max ) while some that ξ * m,j i max are sampled independent with ξ * m,j (observable for j i max ).…”

mentioning

confidence: 99%

“…To further prove P1, we focus on the difference between local statistics W i,j i max − W i,j for ∀j ∈ U , where the increment value depends on ξ * m,j i max u min and ξ * m,j u min (we consider the positive CUSUM without loss of generality). Thus, it suffices to investigate (Nabhan et al, 2021…”

mentioning

confidence: 99%

Adaptive Sampling for Monitoring Multi-Profile Data with Within-and-between Profile Correlation

Xian

Wang

2023

Technometrics

View full text Add to dashboard Cite

Then ∀j ∈ U , we have W i,j i max ≥ W i,j , ∀i > m 1 . The lemma 1 can be proved by contradiction: Assume there ∃i > m 1 , s.t W i,j > W i,j i max for any j ∈ U . If profile j i max is selected to be in Ω(i + 1), then j must also be in Ω(i + 1), which contradicts to the definition of U . If profile j i max is not selected to be in Ω(i + 1), then none of the profiles in P/U can be in Ω(i + 1). In this case, all profiles in Ω(i + 1) are from U , which also contradicts to the definition of U . Hence lemma 1 is proved.To further prove P1, we focus on the difference between local statistics W i,j i max − W i,j for ∀j ∈ U , where the increment value depends on ξ * m,j i max u min and ξ * m,j u min (we consider the positive CUSUM without loss of generality). Thus, it suffices to investigate (Nabhan et al., 2021Note both ξ * m,j and ξ * m,j i max asymptotically follow N (0, 1), thus their expectations are both 0. However, the term ξ * m,j i max − ξ * m,j are not necessarily identical in terms of m. This is because there might be some ms that ξ * m,j i max are jointly sampled with ξ * m,j (unobservable for the j i max ) while some that ξ * m,j i max are sampled independent with ξ * m,j (observable for j i max ). Nevertheless, ξ * m,j i max − ξ * m,j are still independent

show abstract

Correlation-based dynamic sampling for online high dimensional process monitoring

Cited by 15 publications

References 28 publications

Impact of climate indicators on the COVID-19 pandemic in Saudi Arabia

Impact of climate indicators on the COVID-19 pandemic in Saudi Arabia

Bayesian Quickest Detection of Propagating Spatial Events

Adaptive Sampling for Monitoring Multi-Profile Data with Within-and-between Profile Correlation

Contact Info

Product

Resources

About