Ondrej Sluciak
Vienna University of Technology
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Ondrej Sluciak.
IEEE Transactions on Signal Processing | 2012
Ondrej Hlinka; Ondrej Sluciak; Franz Hlawatsch; Petar M. Djuric; Markus Rupp
We consider distributed state estimation in a wireless sensor network without a fusion center. Each sensor performs a global estimation task-based on the past and current measurements of all sensors-using only local processing and local communications with its neighbors. In this estimation task, the joint (all-sensors) likelihood function (JLF) plays a central role as it epitomizes the measurements of all sensors. We propose a distributed method for computing, at each sensor, an approximation of the JLF by means of consensus algorithms. This “likelihood consensus” method is applicable if the local likelihood functions of the various sensors (viewed as conditional probability density functions of the local measurements) belong to the exponential family of distributions. We then use the likelihood consensus method to implement a distributed particle filter and a distributed Gaussian particle filter. Each sensor runs a local particle filter, or a local Gaussian particle filter, that computes a global state estimate. The weight update in each local (Gaussian) particle filter employs the JLF, which is obtained through the likelihood consensus scheme. For the distributed Gaussian particle filter, the number of particles can be significantly reduced by means of an additional consensus scheme. Simulation results are presented to assess the performance of the proposed distributed particle filters for a multiple target tracking problem.
international conference on acoustics, speech, and signal processing | 2011
Ondrej Hlinka; Ondrej Sluciak; Franz Hlawatsch; Petar M. Djuric; Markus Rupp
We propose a distributed implementation of the Gaussian particle filter (GPF) for use in a wireless sensor network. Each sensor runs a local GPF that computes a global state estimate. The updating of the particle weights at each sensor uses the joint likelihood function, which is calculated in a distributed way, using only local communications, via the recently proposed likelihood consensus scheme. A significant reduction of the number of particles can be achieved by means of another consensus algorithm. The performance of the proposed distributed GPF is demonstrated for a target tracking problem.
asilomar conference on signals, systems and computers | 2010
Ondrej Hlinka; Ondrej Sluciak; Franz Hlawatsch; Petar M. Djuric; Markus Rupp
We propose a distributed method for computing the joint (all-sensors) likelihood function (JLF) in a wireless sensor network. A consensus algorithm is used for a decentralized, iterative calculation of a sufficient statistic that describes an approximation to the JLF. After convergence of the consensus algorithm, the approximate JLF—which epitomizes the measurements of all sensors—is available at each sensor. This “likelihood consensus” method requires only communications between neighboring sensors. We implement the likelihood consensus method in a distributed particle filtering scheme. Each sensor runs a local particle filter that computes a global state estimate. The updating of the particle weights of each local particle filter uses the JLF. The performance of this distributed particle filter is demonstrated on a target tracking problem.
IEEE Signal Processing Letters | 2013
Ondrej Sluciak; Markus Rupp
We present novel distributed algorithms for estimating the number of nodes in a wireless sensor network without any a-priori knowledge or node preferences. The algorithms originate from distributed forms of Gram-Schmidt orthogonalization algorithms where the goal is to distributively find a set of orthogonal vectors. Using concepts from linear algebra, by finding the number of independent (orthogonal) vectors, we also find the number of nodes in a network.
international conference on acoustics, speech, and signal processing | 2014
Ondrej Hlinka; Ondrej Sluciak; Franz Hlawatsch; Markus Rupp
We propose an iterative extension of the covariance intersection (CI) algorithm for distributed data fusion. Our iterative CI (ICI) algorithm is able to disseminate local information throughout the network. We show that the ICI algorithm converges asymptotically to a consensus across all network nodes. We furthermore apply the ICI algorithm to distributed sequential Bayesian estimation and propose an ICI-based distributed particle filter (DPF). This DPF allows for spatially correlated measurement noises with unknown cross-correlations and does not require knowledge of the network size. The performance of the proposed DPF is assessed experimentally for a target tracking problem.
asilomar conference on signals, systems and computers | 2011
Ondrej Sluciak; Ondrej Hlinka; Markus Rupp; Franz Hlawatsch; Petar M. Djuric
We propose a sequential likelihood consensus (SLC) for a distributed, sequential computation of the joint (all-sensors) likelihood function (JLF) in a wireless sensor network. The SLC is based on a novel dynamic consensus algorithm, of which only a single iteration is performed per time step. We demonstrate the application of the SLC in a distributed particle filter with low communication requirements and low latency. Because the JLF is available at each sensor, the local particle filters at the individual sensors take into account the measurements of all sensors. The performance of the proposed distributed particle filter is assessed for a target tracking problem.
asilomar conference on signals, systems and computers | 2012
Ondrej Sluciak; Hana Straková; Markus Rupp; Wilfried N. Gansterer
We propose a novel distributed QR factorization algorithm for orthogonalizing a set of vectors in a wireless sensor network. The algorithm originates from the classical Gram-Schmidt orthogonalization which we formulate in a distributed way using the dynamic consensus algorithm. In contrast to existing distributed QR factorization algorithms, all elements of matrices Q and R are computed simultaneously and updated iteratively after each transmission. Assuming synchronous message broadcasting and communication only with neighboring nodes without any central computing unit (fusion center), we prove convergence of the algorithm. We investigate the algorithm in terms of numerical accuracy and we discuss the influence of the initial data distribution on the algorithm performance. Moreover, we provide a comparison with existing distributed QR algorithms in terms of communication cost and memory requirements, and we illustrate the comparison by simulations.
international conference on acoustics, speech, and signal processing | 2011
Ondrej Sluciak; Markus Rupp
Many models of wireless sensor networks (WSNs) assume a perfect synchronization along the graph of such network as a simplifying assumption. In our contribution we base our investigations of distributed algorithms solving consensus problems on more realistic, asynchronous networks in which nodes randomly transmit to their neighborhood. Following a linear algebraic approach we show conditions for convergence to a consensus and derive convergence properties in the mean and mean square sense.
international conference on acoustics, speech, and signal processing | 2010
Ondrej Sluciak; Thibault Hilaire; Markus Rupp
The major contribution of this paper is the presentation of a general unifying description of distributed algorithms allowing to map local, node-based algorithms onto a single global, network-based form. As a first consequence the new description offers to analyze their learning and steady-state behavior by classical methods. A further consequence is the analysis of implementation issues as they appear due to quantization in computing and communication links. Exemplarily, we apply the new method on several different averaging algorithms: the Push-Sum protocol, average consensus as well as its quantized form and furthermore examine the effects of quantization noise which is introduced by the bandwidth limited communication links and finite precision computation ability of every node. Statistical properties of these quantization noises are provided and verified by simulations.
ieee signal processing workshop on statistical signal processing | 2012
Ondrej Sluciak; Markus Rupp
In this contribution we present a stronger notion of almost sure convergence for a large class of consensus algorithms including also asynchronous updates. We introduce the concept of the so-called relaxed projection algorithms and show that many consensus algorithms can be interpreted as such relaxed projection updates. It is well known that such algorithms converge to a solution lying in the intersection of the projections. The convergence of such algorithms is, however, guaranteed only for deterministic ordering of the projections. Since we are interested in random data exchanges, we analyze the convergence in case of random orderings of the projections and show that the algorithms converge in the underrelaxed case even for time-varying and individual mixing parameters.