Sequential Estimation and Diffusion of Information Over Networks: A Bayesian Approach With Exponential Family of Distributions

Dedecius, Kamil; Djurić, Petar M.

doi:10.1109/tsp.2016.2641380

Cited by 39 publications

(27 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Under the probability perspective, it is interesting to utilize the well-known Bayesian interpretation to disclose 1) what kind of posterior probability distribution of unknown states is reasonable in the formalization of distributed Kalman filtering and 2) how it is compared to the optimal posterior distribution of states conditioned on all present and past network data [47]. Usually, the locally available measurements Y i,k = {y j,k , j ∈ N i ∪{i}} together with mutual independence assumptions are first used to update the desired posterior PDFp i (x k |Y i,k ), and then the obtained PDFs are assimilated into a PDF [48] via optimizing average KL divergence [49] or Logarithmic opinion [50] where ∑ α j = 1. Recently, various alternative schemes are developed to overcome the deficiency of above fusion approach in communication and computation.…”

Section: Filter Structurementioning

confidence: 99%

A Survey on Model-Based Distributed Control and Filtering for Industrial Cyber-Physical Systems

Ding

Han

Wang

et al. 2019

IEEE Trans. Ind. Inf.

427

124

View full text Add to dashboard Cite

Industrial cyber-physical systems (CPSs) are largescale, geographically dispersed and life-critical systems, in which lots of sensors and actuators are embedded and networked together to facilitate real-time monitoring and closed-loop control. Their intrinsic features in geographic space and resources put forward to urgent requirements of reliability and scalability for designed filtering or control schemes. This paper presents a review of the state-of-the-art of distributed filtering and control of industrial CPSs described by differential dynamics models. Special attention is paid to sensor networks, manipulators and power systems. For real-time monitoring, some typical Kalmanbased distributed algorithms are summarized and their performances on calculation burden, communication burden as well as scalability are discussed in depth. Then, the characteristics of non-Kalman cases are further disclosed in light of constructed filter structures. Furthermore, the latest development is surveyed for distributed cooperative control of mobile manipulators and distributed model predictive control in industrial automation systems. By resorting to droop characteristics, representative distributed control strategies classified by controller structures are systematically summarized for power systems with the requirements of power sharing, and voltage and frequency regulation. In addition, distributed security control of industrial CPSs is reviewed when cyber-attacks are taken into consideration. Finally, some challenges are raised to guide the future research. Index Terms-Industrial cyber-physical systems; distributed filtering; distributed control; power schedule; droop characteristics. I. INDUSTRIAL CYBER-PHYSICAL SYSTEMS A cyber-physical system (CPS), a fast-growing research area, is a highly integrated system of physical components involving sensors, actuators and various equipments, as well as cyber possessing ubiquitous computation and efficient communication. From the engineering point of view, CPSs are considered as the most promising industrial systems including transportation networks, energy systems, water/gas distribution networks, and unmanned factories. For example, multiple industrial robots with an inertial navigation device or various sensors are programmed for the movement along a programmed trajectory to cooperatively complete production tasks [1], [2]. The main advantage of these systems is that the tight

show abstract

Section: Filter Structurementioning

confidence: 99%

A Survey on Model-Based Distributed Control and Filtering for Industrial Cyber-Physical Systems

Ding

Han

Wang

et al. 2019

IEEE Trans. Ind. Inf.

427

124

View full text Add to dashboard Cite

show abstract

“…With regard to the type of information disseminated, three main categories of protocols exist; we note that there are protocols such as diffusion [8], [9] that may belong to either. Our approach falls into the last category: 1) Measurement/Likelihood.…”

Section: Introductionmentioning

confidence: 99%

Partial Consensus and Conservative Fusion of Gaussian Mixtures for Distributed PHD Fusion

Corchado

Sun

2019

IEEE Trans. Aerosp. Electron. Syst.

View full text Add to dashboard Cite

We propose a novel consensus notion, called "partial consensus", for distributed GM-PHD (Gaussian mixture probability hypothesis density) fusion based on a peer-to-peer (P2P) sensor network, in which only highly-weighted posterior Gaussian components (GCs) are disseminated in the P2P communication for fusion while the insignificant GCs are not involved. The partial consensus does not only enjoy high efficiency in both network communication and local fusion computation, but also significantly reduces the affect of potential false data (clutter) to the filter, leading to increased signal-to-noise ratio at local sensors. Two "conservative" mixture reduction schemes are advocated for fusing the shared GCs in a fully distributed manner. One is given by pairwise averaging GCs between sensors based on Hungarian assignment and the other is merging close GCs based a new GM merging scheme. The proposed approaches have a close connection to the conservative fusion approaches known as covariance union and arithmetic mean density. In parallel, average consensus is sought on the cardinality distribution (namely the GM weight sum) among sensors. Simulations for tracking either a single target or multiple targets that simultaneously appear are presented based on a sensor network where each sensor operates a GM-PHD filter, in order to compare our approaches with the benchmark generalized covariance intersection approach. The results demonstrate that the partial, arithmetic average, consensus outperforms the complete, geometric average, consensus.

show abstract

“…Further, ξ is a hyperparameter of the same size as T (y, z), ν ∈ R + is a scalar hyperparameter, and q(ξ, ν) is a known function. Taking advantage of this distribution under conjugacy, the Bayesian update (14) takes the form [30]:…”

Section: Bayesian Real-time Solutionmentioning

confidence: 99%

“…, the combination step to achieve a non-negative minimal KLD takes the form [30] (19) where coefficients {c nm,i , ∀m ∈ N n } are unit |N n |-simplex weights expressing the degree of belief of customer n in customer {m ∈ N n }'s information. So, we can formulate the combination phase as:…”

Section: Bayesian Real-time Solutionmentioning

confidence: 99%

See 1 more Smart Citation

A Bayesian Real-Time Electric Vehicle Charging Strategy for Mitigating Renewable Energy Fluctuations

Latifi

Khalili

Rastegarnia

et al. 2019

IEEE Trans. Ind. Inf.

View full text Add to dashboard Cite

A novel pricing and scheduling mechanism is proposed here for Plug-in electric vehicles (PEVs) charging/discharging to track and synchronize with a renewable power generation pattern. Moreover, the proposed mechanism can be used in the demand-side management and ancillary service applications, respectively for the peak shaving and frequency regulation responding. We design a fully distributed stochastic optimization mechanism using Bayesian pure strategic repeated game by which the PEVs optimally schedule their demands. We also use a mixed Bayesian-diffusion Kalman filtering strategy for the customers to collaboratively estimate and track the stochastic price and regulation signals for the upcoming scheduling window. In the proposed paper all the characteristics of the PEVs, as well as the uncertainty about their deriving patterns are considered. As our framework converges to an equilibrium even with incomplete information, is agent-based, and the agents share the information only with their optional neighbors, it is scale-free, robust, and secure. Index Terms-Bayesian game scheduling, diffusion Kalman filtering, plug-in electric vehicles, renewable energy resources.

show abstract

Sequential Estimation and Diffusion of Information Over Networks: A Bayesian Approach With Exponential Family of Distributions

Cited by 39 publications

References 52 publications

A Survey on Model-Based Distributed Control and Filtering for Industrial Cyber-Physical Systems

A Survey on Model-Based Distributed Control and Filtering for Industrial Cyber-Physical Systems

Partial Consensus and Conservative Fusion of Gaussian Mixtures for Distributed PHD Fusion

A Bayesian Real-Time Electric Vehicle Charging Strategy for Mitigating Renewable Energy Fluctuations

Contact Info

Product

Resources

About