betaFeedback

WorksEngineering

An Inspection Game for Monitor Design Against GNSS Timing Attacks in Synchronous Islanding

Abstract

Phasor measurement units (PMUs) take their time from GNSS, so a spoofer that slowly drags a receiver clock rotates the reported phase without touching the power waveform. The threat is acute in synchronous islanding, where a controller keeps an islanded generator in phase with the grid, using one PMU on each side of an open breaker, so that the two can be reconnected at any moment: a differential clock bias of about 0.560.56,ms can make the controller hold the generator 10∘10^\circ out of phase while it reports perfect synchronism. We cast the defense as a zero-sum inspection game. The defender, as inspector, chooses its timing monitors and audits the clocks periodically; the attacker, as inspectee, drags the two clocks to an unsafe bias by the next audit; the payoff is the evidence, the Kullback–Leibler divergence between the probability laws of the defender's record under attack and under normal operation. Our central result is that the evidence equals the cost of the cheapest innocent explanation of the attacked record. This single identity predicts which monitors work, turns the attacker's best response into a strictly convex quadratic program, and turns the defender's design into a convex program with a saddle point. Pairing explanations, which bound the evidence from above, with test statistics, which bound it from below, certifies audit periods in which every successful attack leaves at least a threshold of evidence. On a 50,Hz island, these evidence-based periods, at the matched-test threshold, are up to 8.4 times as long as those of a rule that checks each monitor separately, and the case study favors placing the better oscillator on the utility side of the breaker.

Introduction

A phasor measurement unit (PMU) measures the phase angle of a bus voltage against a reference cosine synthesized from its own clock, and it disciplines that clock to the Global Navigation Satellite System (GNSS). A clock error of τ\tau seconds therefore rotates the reported angle by 2πf0τ2\pi f_0\tau radians, where f0f_0 is the nominal frequency [1]; at f0=50f_0 = 50 Hz, one millisecond is 18∘18^\circ. A spoofer that captures the receiver can drag its clock and thereby shift the reported phase without touching the power system. Such attacks are inexpensive to mount [2], have been demonstrated against PMUs [3], and corrupt the applications built on synchrophasors [4].

The consequences are most severe where the reported phase is the controlled variable. In synchronous islanding [5, 6], a generator that has separated from the grid is held in phase with it, using one PMU on each side of the open breaker, so that the breaker can be closed at any moment. If the two PMU clocks disagree by τr−τg\tau_r-\tau_g seconds, the controller drives the reported phase difference to zero and thereby holds a true phase difference of 2πf0(τr−τg)2\pi f_0(\tau_r-\tau_g). Closing the breaker outside a synchronization window can cause a damaging transient. We adopt 10∘10^\circ, the strictest phase limit of [7 Table 5] (for resources above 1500 kVA), as the operating threshold of this study; at 50 Hz it corresponds to a clock disagreement of only 0.56 ms.

Without a model of normal operation, neither PMU can distinguish clock manipulation from a suitable physical disturbance. With the breaker open, a drifting clock on the utility side looks like a change of grid frequency, and one on the island side like a change of load; only the statistics of these disturbances make the two accounts distinguishable, and often barely so. Operators therefore audit the clocks periodically against an independent time source, and install monitors that keep their own time: holdover oscillators and, where available, a trusted phase reference. This raises a design question with a precise answer: how far apart may the audits be, so that every attack that succeeds between two audits leaves enough evidence to be detected, and which monitors lengthen that period the most?

Approach. We answer this question with a two-player inspection game [8], stated in Section II before any physics. The defender, the inspector, moves first: it chooses its monitors and announces them. The attacker, the inspectee, who knows the design and the audit schedule, then chooses how to drag the two clocks between two audits, subject to a rate limit, so that reconnection is unsafe at the next audit. The payoff is the evidence, the Kullback–Leibler (KL) divergence between the laws of the defender’s record with and without the attack; the attacker wants it small and the defender large. Under the Gaussian model, the evidence determines the detection probability of the likelihood-ratio test matched to a specified attack, which by the Neyman–Pearson lemma is the most powerful test against that attack. Audits are safe if every successful attack reveals at least a threshold of evidence. At the matched-test threshold for a chosen false-alarm probability and a 50% detection target, safety is necessary for any detector to meet that target against every successful attack, and sufficient for the test matched to each attack; a single detector for unknown attack profiles requires a composite test, which we leave to future work.

Key insight. The defender never observes the attack itself; it observes a record that varies randomly even without an attack. An attack is therefore hard to see exactly when the attacked record could plausibly have come from normal operation. We show that the evidence equals the cost of the cheapest innocent explanation: the least unusual account of normal operation (a slightly low grid frequency, a slightly heavy load, a slightly slow oscillator) that reproduces what the defender sees. Each monitor is thus characterized by what it can blame a clock drift on, and the whole analysis follows from this one identity.

Contributions.

  1. Model and game. A formal inspection game on an exactly discretized model of the island and its controller, with a first-order model of how the clocks enter the PMU reports and of the monitors, in which every quantity is typed by its domain.

  1. Evidence as explanation (Theorem 1). The evidence is a quadratic form that does not depend on the controller, and it equals the cost of the cheapest innocent explanation. Explanations bound it from above and test statistics from below, which sorts the monitors into three regimes.

  1. The attacker. The stealthiest attack is the unique solution of a strictly convex quadratic program. When the rate limit does not bind, it splits the bias between the two clocks in a harmonic ratio, and we characterize when a slow ramp or a late sprint comes close to its value.

  1. The defender. The design problem is a convex program and the game has a saddle point, so the defender loses nothing by announcing its design; under a stated approximation, and in the case study, a better oscillator belongs on the utility side.

  1. Certified audit periods. Lower bounds on the safe audit period that hold against every attack. For the case-study designs with a monitor they lie within 0.6% of the optimum, as bracketed by explicit feasible attacks, and at the matched-test threshold they are 1.1 to 8.4 times as long as those of a rule that checks each monitor separately.

Related work. Stealthy attacks on control systems have been studied from the viewpoints of detectability and identifiability [9] and of the resources an adversary needs [10]. When detectability is measured by the KL divergence, as here, [11, 12, 13] maximize the damage subject to a bound on the divergence. We solve the converse problem, which is to minimize the divergence subject to causing damage, under the structural constraints of clock manipulation (two clocks, a rate limit and an audit deadline), and we then optimize the monitors against the solution. GNSS spoofing of PMUs has been demonstrated [3] and its impact on grid applications analyzed [4]; common defenses compare the receiver clock with a local oscillator [14, 15]. We show that fusing these comparisons with other phasor data, and placing the better oscillator correctly, lengthens the evidence-based safe audit period severalfold in our case study. Inspection games [8] model an inspector who verifies with statistical tests that an inspectee does not violate an agreement, typically requiring that a violation be detected before a critical time, and the Neyman–Pearson lemma is their standard tool. Ours is an inspection game in which the inspector chooses the precision of its instruments rather than the times of its inspections, and the violation is a clock profile constrained by a rate limit. Unlike game-theoretic designs of resilient control [16], it fixes the controller and the alarm policy, which yields a convex formulation with closed-form bounds and special-case solutions; the monitor design is an instance of optimal experimental design [17].

Notation and conventions

Sets and matrices. N={1,2,…}\mathbb{N} = \{1,2,\ldots\}, Z≥0={0,1,2,…}\mathbb{Z}_{\ge0} = \{0,1,2,\ldots\}, and KN={0,1,…,N−1}\mathcal{K}_N = \{0,1,\ldots,N-1\} for N∈NN \in\mathbb{N}. Vectors are columns. Sn\mathbb{S}^n, S+n\mathbb{S}^n_+ and S++n\mathbb{S}^n_{++} are the sets of symmetric, positive semidefinite and positive definite n×nn \times n real matrices, and A⪰BA \succeq B means A−B∈S+nA-B \in\mathbb{S}^n_+. For A∈S+nA \in\mathbb{S}^n_+ and x∈Rnx \in\mathbb{R}^n, ∥x∥A2=xTAx\lVert x\rVert_A^2 = x^\mathsf{T}Ax. InI_n is the identity, diag⁡(⋅)\operatorname{diag}(\cdot) a (block) diagonal matrix, x+=max⁡{x,0}x_+ = \max\{x,0\}, ∣x∣|x| the largest integer not above xx, and 1S∈{0,1}\mathbf{1}_S \in\{0,1\} equals 1 if the statement SS holds. We set inf⁡∅=+∞\inf\varnothing= +\infty; for suprema of sets of nonnegative periods or indices, which lie in [0,∞)[0,\infty), we set sup⁡∅=0\sup\varnothing= 0.

Probability. All random variables are defined on one probability space, and E\mathbb{E} is expectation. N(μ,C)\mathcal{N}(\mu,C) is the Gaussian law with mean μ∈Rn\mu\in\mathbb{R}^n and covariance C∈S+nC \in\mathbb{S}^n_+; Φ:R→(0,1)\Phi:\mathbb{R}\to(0,1) is the standard normal distribution function, and zp=Φ−1(p)z_p=\Phi^{-1}(p) for p∈(0,1)p\in(0,1). For two laws P1,P0\mathbb{P}_1,\mathbb{P}_0 on Rn\mathbb{R}^n, the KL divergence is DKL(P1∥P0)=∫log⁡dP1dP0 dP1D_{\mathrm{KL}}(\mathbb{P}_1\Vert\mathbb{P}_0)=\int\log\frac{\mathrm{d}\mathbb{P}_1}{\mathrm{d}\mathbb{P}_0}\,\mathrm{d}\mathbb{P}_1 if P1\mathbb{P}_1 is absolutely continuous with respect to P0\mathbb{P}_0, and +∞+\infty otherwise; it is measured in nats. We use three standard facts [18]:

(F1) Chain rule. For a random vector (U,V)(U,V), the divergence of its two joint laws is the divergence of the laws of UU plus the expected divergence, under P1\mathbb{P}_1, of the conditional laws of VV given UU.

(F2) Invariance. If f:Rn→Rnf:\mathbb{R}^n\to\mathbb{R}^n is a bijection such that ff and f−1f^{-1} are Borel measurable, then mapping both laws through ff does not change their divergence.

(F3) Gaussian laws. For μ1,μ0∈Rn\mu_1,\mu_0\in\mathbb{R}^n and C∈S++nC\in\mathbb{S}^n_{++}, DKL(N(μ1,C)∥N(μ0,C))=12∥μ1−μ0∥C−12D_{\mathrm{KL}}(\mathcal{N}(\mu_1,C)\Vert\mathcal{N}(\mu_0,C))=\frac{1}{2}\lVert\mu_1-\mu_0\rVert_{C^{-1}}^2.

Units. Time is in seconds (s), frequency in hertz (Hz) and angle in radians (rad); a clock drift is in seconds per second (s/s), a dimensionless ratio. Table I lists the main symbols with their domains, their units and the values used in Section VII.

SymbolDomainUnitMeaningValue
ii{r,g}\{r,g\}–side of the breaker: utility (rr) or island (gg)–
kk; KN\mathcal{K}_NZ\mathbb{Z}; {0,…,N−1}\{0,\ldots,N-1\}–frame index; frames of one audit window–
TsT_s(0,∞)(0,\infty)sframe length0.02
NN; T=NTsT=NT_sN\mathbb{N}; (0,∞)(0,\infty)–; saudit period in frames; in secondsto be chosen
LLZ≥0\mathbb{Z}_{\ge0}–number of frames recorded before the windowany
f0f_0; ω0=2πf0\omega_0=2\pi f_0(0,∞)(0,\infty)Hz; rad/snominal frequency50; 314.16
θi,k\theta_{i,k}; νi,k\nu_{i,k}; xi,kx_{i,k}R\mathbb{R}; R\mathbb{R}; R2\mathbb{R}^2rad; Hz; –phase; frequency deviation; state xi,k=(θi,k,νi,k)Tx_{i,k}=(\theta_{i,k},\nu_{i,k})^{\mathsf{T}} of side ii–
wkw_k; dkd_k; uku_kR\mathbb{R}Hzgrid-frequency step; load disturbance; frequency command–
qwq_w; qdq_d(0,∞)(0,\infty)Hz2\mathrm{Hz}^2variances of wkw_k and dkd_k10−810^{-8}; 10−610^{-6}
TgT_g; ρ\rho; β\beta(0,∞)(0,\infty); (0,1)(0,1); (0,Ts)(0,T_s)s; –; sgovernor time constant; e−Ts/Tge^{-T_s/T_g}; Tg(1−ρ)T_g(1-\rho)0.5; 0.9608; 0.0196
KθK_\theta; KPK_P; KIK_IR\mathbb{R}Hz/rad; –; 1/sgains of the PI controller stabilizing
φk\varphi_k; φmax⁡\varphi^{\max}R\mathbb{R}; (0,π)(0,\pi)radtrue phase difference across the breaker; reconnection window–; 0.1745 (10∘10^\circ)
τi,k\tau_{i,k}; γi,k\gamma_{i,k}R\mathbb{R}; [−γˉ,γˉ][-\bar{\gamma},\bar{\gamma}]s; s/sclock bias of PMU ii; drift (τi,k+1−τi,k)/Ts(\tau_{i,k+1}-\tau_{i,k})/T_s–
γˉ\bar{\gamma}; Δ=γˉTs\Delta=\bar{\gamma}T_s(0,∞)(0,\infty)s/s; srate limit of the drift; largest change of bias per frame2×10−52\times10^{-5}; 4×10−74\times10^{-7}
τc=φmax⁡/ω0\tau^c=\varphi^{\max}/\omega_0(0,1/(2f0))(0,1/(2f_0))sdamaging bias5.56×10−45.56\times10^{-4}
τ\tau; AN\mathcal{A}_NAN⊂R2N\mathcal{A}_N\subset\mathbb{R}^{2N}sclock profile; admissible profiles (1)–
θ~i,k\tilde{\theta}_{i,k}; ν~i,k\tilde{\nu}_{i,k}; yi,ky_{i,k}R\mathbb{R}; R\mathbb{R}; R2\mathbb{R}^2rad; Hz; –reported phase; reported frequency; yi,k=(θ~i,k,ν~i,k)Ty_{i,k}=(\tilde{\theta}_{i,k},\tilde{\nu}_{i,k})^{\mathsf{T}}–
vi,kv_{i,k}; vi,k′v'_{i,k}R\mathbb{R}rad; HzPMU noises, with standard deviations σθ,σν∈(0,∞)\sigma_\theta,\sigma_\nu\in(0,\infty)σθ=10−3\sigma_\theta=10^{-3}; σν=10−3\sigma_\nu=10^{-3}
oi,ko_{i,k}; ni,kn_{i,k}R\mathbb{R}sholdover reading; its noise, with standard deviation σo∈(0,∞)\sigma_o\in(0,\infty)σo=2×10−8\sigma_o=2\times10^{-8}
δi\delta_i; σi\sigma_iR\mathbb{R}; (0,∞)(0,\infty)s/sfrequency offset of oscillator ii; its standard deviationTCXO 3.236×10−73.236\times10^{-7}, OCXO 3.236×10−93.236\times10^{-9}, Rb 3.236×10−113.236\times10^{-11}
θkref\theta_k^{\mathrm{ref}}; eke_kR\mathbb{R}radtrusted utility-phase reference; its noise, with standard deviation σe∈(0,∞)\sigma_e\in(0,\infty)σe=10−3\sigma_e=10^{-3}
J\mathcal{J}; JJ⊆{r,g,e}\subseteq\{r,g,e\}; {0,1,2,3}\{0,1,2,3\}–installed monitors: holdover checks at PMU rr, gg and reference ee; their number ∣J∣\lvert\mathcal{J}\rvert–
λ\lambda; D\mathcal{D}(0,∞)J(0,\infty)^{\mathcal{J}}; D⊂(0,∞)J\mathcal{D}\subset(0,\infty)^{\mathcal{J}}(s/s)−2(\mathrm{s/s})^{-2}, rad−2\mathrm{rad}^{-2}design, λj=σj−2\lambda_j=\sigma_j^{-2} for j∈Jj\in\mathcal{J}, in (s/s)−2(\mathrm{s/s})^{-2} for j=r,gj=r,g and in rad−2\mathrm{rad}^{-2} for j=ej=e; affordable designs–
Y−\mathcal{Y}_{-}; Y\mathcal{Y}R(4+Je)L\mathbb{R}^{(4+J_e)L}; Rm\mathbb{R}^{m}–record before frame 0; record of frames 0,…,N−10,\ldots,N-1, with m=N(4+J)m=N(4+J)–
η\eta; ε\varepsilonRp\mathbb{R}^{p}; Rm\mathbb{R}^{m}–nuisance, with p=2N+2+Jop=2N+2+J_o; measurement noise of the window–
PP; RR; Ψ\Psi; MMS++p\mathbb{S}^{p}_{++}; S++m\mathbb{S}^{m}_{++}; S++m\mathbb{S}^{m}_{++}; S++2N\mathbb{S}^{2N}_{++}–covariance of η\eta; of ε\varepsilon; Ψ=R+FPFT\Psi=R+FPF^{\mathsf{T}}; M=GTΨ−1GM=G^{\mathsf{T}}\Psi^{-1}G–
PiP_i; sis_iS++2\mathbb{S}^{2}_{++}; (0,∞)(0,\infty)–; Hzsteady-state prediction covariance of side ii; frequency uncertaintysr=2.4×10−4s_r=2.4\times10^{-4}; sg=1.0×10−4s_g=1.0\times10^{-4}
E(τ)\mathcal{E}(\tau); Eλ⋆(N)\mathcal{E}^{\star}_\lambda(N)[0,∞)[0,\infty); [0,∞)[0,\infty)natsevidence of τ\tau; least evidence of a successful attack–
hh; α\alpha; BB(0,∞)(0,\infty); (0,12)(0,\frac{1}{2}); (1,∞)(1,\infty)nats; –; –evidence threshold; false-alarm probability; detector threshold, a likelihood ratio (on average at most one false alarm per BB frames)hZ=4.77h_Z=4.77, hB=15.3h_B=15.3; 10−310^{-3}; 4.3×1064.3\times10^6
Th⋆(λ)T_h^\star(\lambda)[0,∞][0,\infty]ssafe audit periodTable III

Table I. TABLE I. MAIN SYMBOLS, THEIR DOMAINS AND UNITS, AND THE VALUES USED IN SECTION VII. A DASH IN THE UNIT COLUMN MARKS A DIMENSIONLESS OR COMPONENT-DEPENDENT QUANTITY.

References

  1. [1]IEEE Standard for Synchrophasor Measurements for Power Systems, IEEE Std C37.118.1-2011, 2011 (superseded in part by IEC/IEEE 60255-118-1:2018).DOI
  2. [2]T. E. Humphreys, B. M. Ledvina, M. L. Psiaki, B. W. O’Hanlon, and P. M. Kintner, Jr., “Assessing the spoofing threat: Development of a portable GPS civilian spoofer,” in Proc. ION GNSS, Savannah, GA, 2008, pp. 2314–2325.
  3. [3]D. P. Shepard, T. E. Humphreys, and A. A. Fansler, “Evaluation of the vulnerability of phasor measurement units to GPS spoofing attacks,” Int. J. Crit. Infrastruct. Prot., vol. 5, no. 3–4, pp. 146–153, 2012.DOI
  4. [4]Z. Zhang, S. Gong, A. D. Dimitrovski, and H. Li, “Time synchronization attack in smart grid: Impact and analysis,” IEEE Trans. Smart Grid, vol. 4, no. 1, pp. 87–98, 2013.DOI
  5. [5]R. J. Best, D. J. Morrow, D. J. McGowan, and P. A. Crossley, “Synchronous islanded operation of a diesel generator,” IEEE Trans. Power Syst., vol. 22, no. 4, pp. 2170–2176, 2007.DOI
  6. [6]R. J. Best, D. J. Morrow, D. M. Laverty, and P. A. Crossley, “Techniques for multiple-set synchronous islanding control,” IEEE Trans. Smart Grid, vol. 2, no. 1, pp. 60–67, 2011.DOI
  7. [7]IEEE Standard for Interconnection and Interoperability of Distributed Energy Resources with Associated Electric Power Systems Interfaces, IEEE Std 1547-2018, 2018.DOI
  8. [8]R. Avenhaus, B. von Stengel, and S. Zamir, “Inspection games,” in Handbook of Game Theory with Economic Applications, vol. 3, R. J. Aumann and S. Hart, Eds. Amsterdam: Elsevier, 2002, ch. 51, pp. 1947–1987.
  9. [9]F. Pasqualetti, F. Dörfler, and F. Bullo, “Attack detection and identification in cyber-physical systems,” IEEE Trans. Autom. Control, vol. 58, no. 11, pp. 2715–2729, 2013.DOI
  10. [10]A. Teixeira, I. Shames, H. Sandberg, and K. H. Johansson, “A secure control framework for resource-limited adversaries,” Automatica, vol. 51, pp. 135–148, 2015.arxiv.org/abs/1212.0226
  11. [11]C.-Z. Bai, F. Pasqualetti, and V. Gupta, “Data-injection attacks in stochastic control systems: Detectability and performance tradeoffs,” Automatica, vol. 82, pp. 251–260, 2017.arxiv.org/abs/1704.00748
  12. [12]Y. Mo and B. Sinopoli, “On the performance degradation of cyber-physical systems under stealthy integrity attacks,” IEEE Trans. Autom. Control, vol. 61, no. 9, pp. 2618–2624, 2016.DOI
  13. [13]E. Kung, S. Dey, and L. Shi, “The performance and limitations of ε-stealthy attacks on higher order systems,” IEEE Trans. Autom. Control, vol. 62, no. 2, pp. 941–947, 2017.DOI
  14. [14]D. P. Shepard and T. E. Humphreys, “Characterization of receiver response to spoofing attacks,” in Proc. ION GNSS, Portland, OR, 2011, pp. 2608–2618.
  15. [15]M. L. Psiaki and T. E. Humphreys, “GNSS spoofing and detection,” Proc. IEEE, vol. 104, no. 6, pp. 1258–1270, 2016.DOI
  16. [16]Q. Zhu and T. Başar, “Game-theoretic methods for robustness, security, and resilience of cyberphysical control systems: Games-in-games principle for optimal cross-layer resilient control systems,” IEEE Control Syst. Mag., vol. 35, no. 1, pp. 46–65, 2015.DOI
  17. [17]F. Pukelsheim, Optimal Design of Experiments, ser. Classics in Applied Mathematics, vol. 50. Philadelphia, PA: SIAM, 2006.
  18. [18]T. M. Cover and J. A. Thomas, Elements of Information Theory, 2nd ed. Hoboken, NJ: Wiley, 2006.
  19. [19]A. A. Girgis and W. L. Peterson, “Adaptive estimation of power system frequency deviation and its rate of change for calculating sudden power system overloads,” IEEE Trans. Power Del., vol. 5, no. 2, pp. 585–594, 1990.DOI
  20. [20]R. Ferrero, P. A. Pegoraro, and S. Toscani, “Synchrophasor estimation for three-phase systems based on Taylor extended Kalman filtering,” IEEE Trans. Instrum. Meas., vol. 69, no. 9, pp. 6723–6730, 2020.DOI
  21. [22]P. M. Anderson and M. Mirheydar, “A low-order system frequency response model,” IEEE Trans. Power Syst., vol. 5, no. 3, pp. 720–729, 1990.DOI
  22. [23]M. A. Lombardi, “Fundamentals of time and frequency,” in The Mechatronics Handbook, R. H. Bishop, Ed. Boca Raton, FL: CRC Press, 2002, ch. 17.
  23. [24]A. Derviškadić, R. Razzaghi, Q. Walger, and M. Paolone, “The White Rabbit time synchronization protocol for synchrophasor networks,” IEEE Trans. Smart Grid, vol. 11, no. 1, pp. 726–738, 2020.DOI
  24. [25]L. Narula and T. E. Humphreys, “Requirements for secure clock synchronization,” IEEE J. Sel. Topics Signal Process., vol. 12, no. 4, pp. 749–762, 2018.DOI
  25. [26]J. W. Pierre, D. J. Trudnowski, and M. K. Donnelly, “Initial results in electromechanical mode identification from ambient data,” IEEE Trans. Power Syst., vol. 12, no. 3, pp. 1245–1251, 1997.DOI
  26. [27]M. Brown, M. Biswal, S. Brahma, S. J. Ranade, and H. Cao, “Characterizing and quantifying noise in PMU data,” in Proc. IEEE PES General Meeting, Boston, MA, 2016, pp. 1–5.
  27. [28]S. Wang, J. Zhao, Z. Huang, and R. Diao, “Assessing Gaussian assumption of PMU measurement error using field data,” IEEE Trans. Power Del., vol. 33, no. 6, pp. 3233–3236, 2018.DOI
  28. [29]B. D. O. Anderson and J. B. Moore, Optimal Filtering. Englewood Cliffs, NJ: Prentice-Hall, 1979.
  29. [30]L. L. Scharf and B. Friedlander, “Matched subspace detectors,” IEEE Trans. Signal Process., vol. 42, no. 8, pp. 2146–2157, 1994.
  30. [31]P. R. Kumar and P. Varaiya, Stochastic Systems: Estimation, Identification, and Adaptive Control. Englewood Cliffs, NJ: Prentice-Hall, 1986.
  31. [32]A. Tartakovsky, I. Nikiforov, and M. Bassaville, Sequential Analysis: Hypothesis Testing and Changepoint Detection. Boca Raton, FL: CRC Press, 2014.
  32. [33]R. T. Rockafellar, Convex Analysis. Princeton, NJ: Princeton Univ. Press, 1970.
  33. [34]M. Sion, “On general minimax theorems,” Pacific J. Math., vol. 8, no. 1, pp. 171–176, 1958.DOI

Paper details

vlnc:W_ds2gjcwu
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Engineering

Not recorded for this paper

View statement
Looking for related papers…
Contents