Abstract
Distributed change-point detection has been a fundamental problem when performing real-time monitoring using sensor networks. We propose a distributed detection algorithm, where each sensor only exchanges CUSUM statistic with their neighbors based on the average consensus scheme, and an alarm is raised when local consensus statistic exceeds a prespecified global threshold. We provide theoretical performance bounds showing that the performance of the fully distributed scheme can match the centralized algorithms under some mild conditions. Numerical experiments demonstrate the good performance of the algorithm, especially, in detecting asynchronous changes.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Boyd, S., Diaconis, P., Xiao, L.: Fastest mixing Markov chain on a graph. SIAM Rev.D 46(4), 667–689 (2004)
Buldygin, V.V., Kozachenko, Y.V.: Sub-gaussian random variables. Ukr. Math. J. 32(6), 483–489 (1980)
Chen, J., Kim, S.H., Xie, Y.: \({\rm S}^3 T\): an efficient score-statistic for spatio-temporal surveillance (2017). arXiv:1706.05331
Fellouris, G., Sokolov, G.: Second-order asymptotic optimality in multisensor sequential change detection. IEEE Trans. Inf. Theory 62(6), 3662–3675 (2016)
Hadjiliadis, O., Zhang, H., Poor, H.V.: One shot schemes for decentralized quickest change detection. IEEE Trans. Inf. Theory 55(7), 3346–3359 (2009)
Karagiannis, G., Altintas, O., Ekici, E., Heijenk, G., Jarupan, B., Lin, K., Weil, T.: Vehicular networking: a survey and tutorial on requirements, architectures, challenges, standards and solutions. IEEE Commun. Surv. Tutor. 13(4), 584–616 (2011)
Kurt, M.N., Wang, X.: Multi-sensor sequential change detection with unknown change propagation pattern (2017). arXiv:1708.04722
Lakhina, A., Crovella, M., Diot, C.: Diagnosing network-wide traffic anomalies. In: ACM SIGCOMM Computer Communication Review, vol. 34, pp. 219–230. ACM (2004)
Li, S., Wang, X.: Order-2 asymptotic optimality of the fully distributed sequential hypothesis test (2016). arXiv:1606.04203
Liu, K., Mei, Y.: Improved performance properties of the CISPRT algorithm for distributed sequential detection. Submitted (2017)
Lorden, G.: Procedures for reacting to a change in distribution. The Annals of Mathematical Statistics pp. 1897–1908 (1971)
Ludkovski, M.: Bayesian quickest detection in sensor arrays. Seq. Anal. 31(4), 481–504 (2012)
Mei, Y.: Efficient scalable schemes for monitoring a large number of data streams. Biometrika 97(2), 419–433 (2010)
Page, E.S.: Continuous inspection schemes. Biometrika 41(1/2), 100–115 (1954)
Raghavan, V., Veeravalli, V.V.: Quickest change detection of a markov process across a sensor array. IEEE Trans. Inf. Theory 56(4), 1961–1981 (2010)
Sahu, A.K., Kar, S.: Distributed sequential detection for Gaussian shift-in-mean hypothesis testing. IEEE Trans. Signal Process. 64(1), 89–103 (2016)
Stankovic, S.S., Ilic, N., Stankovic, M.S., Johansson, K.H.: Distributed change detection based on a consensus algorithm. IEEE Trans. Signal Process. 59(12), 5586–5697 (2011)
Tartakovsky, A.G., Veeravalli, V.V.: An efficient sequential procedure for detecting changes in multichannel and distributed systems. In: Proceedings of the Fifth International Conference on Information Fusion, 2002, vol. 1, pp. 41–48. IEEE (2002)
Tartakovsky, A.G., Veeravalli, V.V.: Quickest change detection in distributed sensor systems. In: Proceedings of the 6th International Conference on Information Fusion, pp. 756–763 (2003)
Tartakovsky, A.G., Veeravalli, V.V.: Asymptotically optimal quickest change detection in distributed sensor systems. Seq. Anal. 27(4), 441–475 (2008)
Valero, M., Clemente, J., Kamath, G., Xie, Y., Lin, F.C., Song, W.: Real-time ambient noise subsurface imaging in distributed sensor networks. In: 2017 IEEE International Conference on Smart Computing (SMARTCOMP), pp. 1–8. IEEE (2017)
Xiao, L., Boyd, S.: Fast linear iterations for distributed averaging. Syst. Control Lett. 53(1), 65–78 (2004)
Xiao, L., Boyd, S., Lall, S.: A scheme for robust distributed sensor fusion based on average consensus. In: Proceedings of the 4th International Symposium on Information Processing in Sensor Networks, p. 9. IEEE Press (2005)
Xie, Y., Siegmund, D.: Sequential multi-sensor change-point detection. Ann. Stat. 41(2), 670–692 (2013)
Acknowledgements
We would like to thank Professor Ansgar Steland for the opportunity to submit an invited paper. This work was partially supported by NSF grants CCF-1442635, CMMI-1538746, DMS-1830210, an NSF CAREER Award CCF-1650913, and a S.F. Express award.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix
Appendix
For simplicity, we first indeed the sensors from 1 to N. Use vector \(\mathbf{L}^t\) to represent \(\left( L(\mathbf{x}_1^{t}),\ldots ,L(\mathbf{x}_N^{t})\right) ^\mathrm{T}\), vector \(\mathbf{y}^t\) to represent \(\left( y_1^t,\ldots ,y_N^t\right) ^\mathrm{T}\) and vector \(\mathbf{z}^t\) to represent \(\left( z_1^t,\ldots ,z_N^t\right) ^\mathrm{T}\). Now, our algorithm can be rewritten as
First, we prove some useful lemmas before reaching the main results.
Remark 1
Since \(\mathbf{W}{} \mathbf{1} = \mathbf{1}\), \(\mathbf{W}^\mathrm{T}=\mathbf{W}\) and \(z^0_v = y^0_v=0\), simple proof by m.i. can verify
Lemma 1
(Hoeffding Inequality) Let \(X_i\) be independent, mean zero, \(\sigma _i^2\)-sub-Gaussian random variables. Then for \(K>0\), \( \mathbb {P}(\sum _{i=1}^{n} X_{n}\ge K)\le \exp \left( -\frac{K^2}{2\sum _{i=1}^{n}\sigma _i^2}\right) . \)
Lemma 2
Consider a sequence of random variables \(X_k {\mathop {\sim }\limits ^{i.i.d.}}\mathcal {P}\), for \(k=1,2,\ldots ,t\). \(\mathcal {P}\) is a sub-Gaussian distribution and its mean and variance are defined as \(\mu _1<0\) and \(\sigma _1\), respectively. Given \(K>0\) large enough, we have
Proof
Case 1. For \(0 < t \le [-\frac{2K}{\mu _1}]\), by Hoeffding Inequality, we have
Using \(\frac{K-k\mu _1}{\sqrt{k}} \ge 2\sqrt{-K\mu _1}\) and \(t \le -\frac{2K}{\mu _1}\), we obtain
Case 2. For \([-\frac{2K}{\mu _1}]+1 \le t \), by (6), we have
Utilizing Hoeffding Inequality and \(k\ge [-\frac{2K}{\mu _1}]+1\), we obtain
Besides, for \(k \ge [-\frac{2K}{\mu _1}]+1\), we have
Then, from Hoeffding Inequality, (8) and (9), we derive
From (10), we know that the second term on the RHS of (7) is a small quantity compared with the first term provided K large enough, so we can neglect it to obtain
Note that \( \mathbb E_{f_1} [L(x_j^t)] = \mu _1 < 0 \), \( \mathbb E_{f_2} [L(x_j^t)] = \mu _2 > 0 \), \( \text{ Var }_{f_1} [L(x_j^t)] = \sigma _1^2 \), \( \text{ Var }_{f_2} [L(x_j^t)] = \sigma _2^2 \).
Given \(\varepsilon >0\) and \(p>0\), Define event
where b is the prespecified threshold in detection. Besides, we use \(\{T_s = p\}\) to represent the event that our algorithm detects the change at \(t=p\). We have the following lemma
Lemma 3
For any \(t \le p \), we have
Proof
Note that, by eigen-decomposition, \( \mathbf{W} = \frac{1}{N} \mathbf{1} \mathbf{1}^\intercal + \sum _{j=2}^N \lambda _j w_j w_j^\intercal . \) Throughout the proof, we assume under the condition that \(B(\varepsilon ,p)\) occurs. First, by the recursive form of our algorithm in (3), the result in (4) and the definition of \(B(\varepsilon ,p)\), for any sensor j, we have
where \(\lambda _2\) is the second largest eigenvalue modulus of \(\mathbf{W}\). If \(\{T_s = t\}\) happens, then \(z_{j}^{t}>b\) holds for some j, which, together with the inequality above, leads to \( \frac{\sum _{j=1}^{N} y_{j}^{t}}{N}> (1- \frac{ \sqrt{N}\varepsilon \lambda _{2}}{1- \lambda _{2}})b. \)
Lemma 4
Assume a sequence of independent random variables \(Y_1,\ldots ,Y_N\). Take any integer \(M>N\) and let
Then we have
Proof
\(\forall (y_1,\ldots ,y_N) \in \{(Y_1,\ldots ,Y_N): \sum _{j=1}^{N} Y_{j}> K\}\), take \( i_j = \left[ \frac{y_j M}{ \sum _{j=1}^{N} y_{j}}\right] \), for \(j=1,\ldots ,N \), where \(\left[ x\right] \) denotes the largest integer smaller than x. We can verify that \(M-N \le \sum _{j=1}^{N} i_j \le M\) and \(y_{j} > i_j K/M\). To see \(M-N \le \sum _{j=1}^{N} i_j\), notice that \(i_j\ge \frac{y_jM}{\sum _{j=1}^N y_j}-1\).Therefore, we have
Since \(Y_{j}\)’s are independent with each other, we obtain
1.1 5.1 Proof of Theorem 1
First, we calculate the probability that our algorithm stops within time p. The value of p is to be specified later.
where the last inequality is from Hoeffding Inequality and assumptions in Sect. 3. The value of \(\varepsilon \) is to be specified later.
Denote \(\bar{b} = N(1- \frac{ \sqrt{N}\varepsilon \lambda _{2}}{1- \lambda _{2}})b\), then \(\bar{b}\) will also tend to infinity as b tends to infinity provided \(\varepsilon \) small enough. By Lemma 3, we have
By Lemma 4, we have
where \(B_j(\varepsilon , p) = \{ |L(\mathbf{x}_j^t)|< \varepsilon b, \text{ for } t=1,\ldots ,p \}\) and the value of M is to be specified later. If \(y_{j}^{t}>i_j \bar{b}/M\), then there must exist \(1\le k\le t \) such that \(y_{j}^{t} = \sum _{q=k}^{t} L(\mathbf{x}_j^q) \ge i_j \bar{b}/M\). So, we have
The influence of \(B_j(\varepsilon ,p)\) in (14) can be interpreted as truncating the original distribution of \(L(\cdot )\). It’s obvious that the new distribution is still sub-Gaussian. Besides, the mean and variance almost keep unchanged provided \(\varepsilon b\) large enough.
If \(i_j=0\), we just set the upper bound of the probability in (14) to be 1. If \(i_j\ne 0\), by Lemma 2, we have
Plugging (15) into (13), we get
Plugging (16) and (13) into (12), we obtain
Next, we will show that as b tends to infinity, the second term on the RHS of (17) is a small quantity in comparison with the first term if we choose the value of M and \(\varepsilon \) properly. Note that 2Np is a small quantity in comparison with \( p\left| C(M,N)\right| \left( -2\bar{b}/\mu _1\right) ^N\), so we only require \(\frac{2\bar{b}\mu _1}{\sigma _1^{2}}(1-\frac{N}{M}) \ge -\frac{(\varepsilon b - \mu _1)^2}{2\sigma _1^2}.\) Choose \(M=(N+1)^2\). Recall that \(\bar{b} = N(1- \frac{ \sqrt{N}\varepsilon \lambda _{2}}{1- \lambda _{2}})b\), the equation above can be rewritten as
To ensure that (19) holds as b tends to infinity, \(\varepsilon = 2\sqrt{-N\mu _1/b}\) is sufficient. Plugging the value of M and \(\varepsilon \) into (17) and neglecting the second term, we get
So \(\forall l > -\frac{4N(N^3+N^2+N) \sqrt{-\mu _1}\mu _1\lambda _{2}}{(N+1)^2(1-\lambda _{2})\sigma _1^2}\), if we choose \( p = \mathrm{exp}\left( -\frac{2(N^3+N^2+N)\mu _1 b}{(N+1)^2\sigma _1^{2}} - l\sqrt{b}\right) , \)
which together with the definition of \(\mathrm{ARL}\) leads to \( \mathrm{ARL} \ge p\), \(\forall l > -\frac{4N(N^3+N^2+N) \sqrt{-\mu _1}\mu _1\lambda _{2}}{(N+1)^2(1-\lambda _{2})\sigma _1^2}\). This leads to our desired result when b tends to infinity.
1.2 5.2 Proof of Lemma 1
First of all, note that \(\mathbb {P}\left( T_s =+\infty \right) =0\), so given \(\varepsilon >0\), we have
If \(T_s=t\), then we have that \(z_j^{t-1}<b\) holds for all j. Since \(\sum _{j=1}^{N} z_{j}^{t-1} = \sum _{j=1}^{N} y_{j}^{t-1}\), there must exist some \(y_j^{t-1}<b\). Therefore, we have
Note that \(y_j^{t-1}\ge \sum _{q=1}^{t-1}L(\mathbf{x}_j^q) \), together with Hoeffding Inequality, we get
When b is large enough, for any \(t>[\frac{b(1+\varepsilon )}{\mu _2}]\), utilizing the similar technique in (9), we get
Plugging (22) and (23) into (21), utilizing the similar technique in (10), we get
Note that \(\forall \varepsilon >0\), as b tends to infinity, the RHS of (24) would converge to zero. Therefore, by (20), we get \( \mathrm{EDD}\le \frac{b\big (1+o(1)\big )}{\mu _2}. \)
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Liu, Q., Zhang, R., Xie, Y. (2019). Distributed Change Detection via Average Consensus over Networks. In: Steland, A., Rafajłowicz, E., Okhrin, O. (eds) Stochastic Models, Statistics and Their Applications. SMSA 2019. Springer Proceedings in Mathematics & Statistics, vol 294. Springer, Cham. https://doi.org/10.1007/978-3-030-28665-1_13
Download citation
DOI: https://doi.org/10.1007/978-3-030-28665-1_13
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-28664-4
Online ISBN: 978-3-030-28665-1
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)