Signal-comparison-based distributed estimation under decaying average data rate communications
asymptotic properties of parameter estimatorsconsensus algorithmdata ratedistributed estimation algorithmnetworksensorstochastic approximation
Asymptotic properties of parametric estimators (62F12) Parametric inference under constraints (62F30) Estimation in multivariate analysis (62H12) Distributed algorithms (68W15) Discrete event control/observation systems (93C65) Estimation and detection in stochastic control theory (93E10) Communication theory (94A05)
In this paper, the authors consider a network \( \mathcal{G} = \left( \mathcal{V}, \mathcal{E}, \mathcal{A} \right) \) with \( N \) sensors. The sensor \( i \in \mathcal{V} = \{ 1, \ldots, N \} \) receives information about the unknown parameter \( \theta \in \mathbb{R}^n \) from the observations \( \mathtt{y}_{i,k} \in \mathbb{R}^{m_i} \),\N\begin{align*}\N\mathtt{y}_{i,k} = H_{i,k} \theta + \mathtt{w}_{i,k},\N\end{align*}\Nwhere \( k \) is the time index, \( H_{i,k} \in \mathbb{R}^{m_i\times n} \) is the measurement matrix, \( \mathtt{w}_{i,k} \in \mathbb{R}^{m_i} \) is the observation noise. The communications between sensors is described by an undirected weighted graph \( \mathcal{G} = \left( \mathcal{V}, \mathcal{E}, \mathcal{A} \right) \), where \( \mathcal{V} = \{ 1, \ldots, N \} \) is the set of sensors, \( \mathcal{E} = \{ (i,j) : i,j \in \mathcal{V} \} \) is the edge set, \( (i,j) \in \mathcal{E} \) if and only if the sensor \( i \) and the sensor \( j \) can communicate with each other, \( \mathcal{A} = (a_{ij})_{N\times N} \) represents the symmetric weighted adjacency matrix of the graph whose elements are all non-negative while \( a_{ij} > 0 \) if and only if \( (i,j)\in \mathcal{E} \), \( \mathcal{N}_{i} = \{j : (i,j)\in\mathcal{E}\}\) is used to denote the neighbor set of the sensor \( i \). The goal of this paper is to cooperatively estimate the unknown parameter \( \theta \). Cooperative estimation requires information exchange between sensors, which brings communication cost. The authors propose a construction of the distributed estimation algorithm under the consensus+innovations framework (see [\textit{S. Kar} et al., IEEE Trans. Inf. Theory 58, No. 6, 3575--3605 (2012; Zbl 1365.94021)], where a consensus protocol is necessary to fuse the messages transmitted in the network. Therefore, the signal-comparison (SC) consensus algorithm is introduced as the foundation of the proposed distributed estimation algorithm (see [\textit{J. M. Ke} et al., CDC 2023, 8382--8387 (2023; \url{doi:10.1109/CDC49753.2023.10383784})]. The proposed SC-based algorithm has the advantages of high effectiveness and low communication cost. The estimates of the SC-based algorithm converge to the true value in the almost sure and mean square sense, and a polynomial almost sure convergence rate is also obtained. For the communication cost, the local and global average data rates decay to zero at a polynomial rate. The efficiency of the algorithm is demonstrated with the help of a simulation example
- $\mathcal {CIRFE}$: A Distributed Random Fields Estimator
- Analysis of normalized least mean squares-based consensus adaptive filters under a general information condition
- Communication constraints in the average consensus problem
- Compressive Diffusion Strategies Over Distributed Networks for Reduced Communication Load
- Consensus of Multi-Agent Systems Under Binary-Valued Measurements and Recursive Projection Algorithm
- Consensus Protocol for Multiagent Systems With Undirected Topologies and Binary-Valued Communications
- Decentralized Cooperative Online Estimation With Random Observation Matrices, Communication Graphs and Time Delays
- Differentially private distributed parameter estimation
- Distributed Average Consensus With Dithered Quantization
- Distributed consensus for multiagent systems with communication delays and limited data rate
- Distributed Consensus With Limited Communication Data Rate
- Distributed Estimation in Energy-Constrained Wireless Sensor Networks
- Distributed linear parameter estimation: asymptotically efficient adaptive strategies
- Distributed LMS estimation over networks with quantised communications
- Distributed Order Estimation of ARX Model under Cooperative Excitation Condition
- Distributed Parameter Estimation in Sensor Networks: Nonlinear Observation Models and Imperfect Communication
- Distributed Parameter Estimation Over Unreliable Networks With Markovian Switching Topologies
- Distributed Parameter Estimation With Quantized Communication via Running Average
- Distributed Recursive Estimation under Heavy-Tail Communication Noise
- Dynamic Event-Triggered Leader-Follower Consensus Control for MultiAgent Systems
- Event-triggered distributed estimation with decaying communication rate
- Finite-Bit Quantization for Distributed Algorithms With Linear Convergence
- Finite-Level Quantized Synchronization of Discrete-Time Linear Multiagent Systems with Switching Topologies
- Generating dithering noise for maximum likelihood estimation from quantized data
- scientific article; zbMATH DE number 1232374 (Why is no real title available?)
- Hybrid adaptive event-triggered consensus control with intermittent communication and control updating
- Mathematical analysis I. Translated from the 4th and 6th Russian editions by Roger Cooke and Octavio T. Paniagua
- Mitigating Quantization Effects on Distributed Sensor Fusion: A Least Squares Approach
- Quantization for decentralized learning under subspace constraints
- Quantized coordination algorithms for rendezvous and deployment
- System identification using binary sensors
This page was built for publication: Signal-comparison-based distributed estimation under decaying average data rate communications
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6994581)
