liu.seSearch for publications in DiVA
Change search
Refine search result
1 - 30 of 30
CiteExportLink to result list
Permanent link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Rows per page
  • 5
  • 10
  • 20
  • 50
  • 100
  • 250
Sort
  • Standard (Relevance)
  • Author A-Ö
  • Author Ö-A
  • Title A-Ö
  • Title Ö-A
  • Publication type A-Ö
  • Publication type Ö-A
  • Issued (Oldest first)
  • Issued (Newest first)
  • Created (Oldest first)
  • Created (Newest first)
  • Last updated (Oldest first)
  • Last updated (Newest first)
  • Disputation date (earliest first)
  • Disputation date (latest first)
  • Standard (Relevance)
  • Author A-Ö
  • Author Ö-A
  • Title A-Ö
  • Title Ö-A
  • Publication type A-Ö
  • Publication type Ö-A
  • Issued (Oldest first)
  • Issued (Newest first)
  • Created (Oldest first)
  • Created (Newest first)
  • Last updated (Oldest first)
  • Last updated (Newest first)
  • Disputation date (earliest first)
  • Disputation date (latest first)
Select
The maximal number of hits you can export is 250. When you want to export more records please use the Create feeds function.
  • 1.
    Athalye, Akshay
    et al.
    Stony Brook University, USA.
    Savic, Vladimir
    Technical University of Madrid, Spain.
    Bolic, Miodrag
    University of Ottawa, Canada.
    Djuric, Petar M.
    Stony Brook University, USA.
    A Radio Frequency Identification System for accurate indoor localization2011In: Proc. of IEEE Intl. Conf. on Acoustics, Speech and Signal Processing (ICASSP), 2011, p. 1777-1780Conference paper (Refereed)
    Abstract [en]

    In this paper we present a novel Radio Frequency Identification (RFID) system for accurate indoor localization. The system is composed of a standard Ultra High Frequency (UHF), ISO-18006C compliant RFID reader, a large set of standard passive RFID tags whose locations are known, and a newly developed tag-like RFID component that is attached to the items that need to be localized. The new semi-passive component, referred to as sensatag (sense-a-tag), has a dual functionality wherein it can sense the communication between the reader and standard tags which are in its proximity, and also communicate with the reader like standard tags using backscatter modulation. Based on the information conveyed by the sensatags to the reader, localization algorithms based on binary sensor principles can be developed. We present results from real measurements that show the accuracy of the proposed system.

    Download full text (pdf)
    fulltext
  • 2.
    Athalye, Akshay
    et al.
    Stony Brook University, USA.
    Savic, Vladimir
    Signal Processing Application Group, Universidad Politecnica de Madrid, Madrid, Spain.
    Bolic, Miodrag
    University of Ottawa, Canada.
    Djuric, Petar M.
    Stony Brook University, USA.
    Novel Semi-Passive RFID System for Indoor Localization2013In: IEEE Sensors Journal, ISSN 1530-437X, E-ISSN 1558-1748, Vol. 13, no 2, p. 528-537Article in journal (Refereed)
    Abstract [en]

    In this paper, we present a novel semi-passive radio-frequency identification (RFID) system for accurate indoor localization. The system is composed of a standard ultra high frequency (UHF) ISO-18000-6C compliant RFID reader, a set of standard passive RFID tags whose locations are known, and a newly developed tag-like RFID component, which is attached to the items that need to be localized. The new semi-passive component, referred to as sensatag (sense-a-tag), has a dual functionality: it can sense and decode communication between the reader and standard tags in its proximity, and can communicate with the reader like standard tags using backscatter modulation. Based on the information conveyed by the sensatags to the reader, localization algorithms based on binary sensor principles can be developed. We conduct a number of experiments in a laboratory to quantify the performance of our system, including two real applications, one finding the exact placement of items on shelves, and the other estimating the direction of item movement.

    Download full text (pdf)
    fulltext
  • 3.
    Lindberg, Christopher
    et al.
    Chalmers University of Technology, Sweden.
    Muppirisetty, L. Srikar
    Chalmers University of Technology, Sweden.
    Dahlen, Karl-Magnus
    HiQ Consulting, Gothenburg, Sweden.
    Savic, Vladimir
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, The Institute of Technology.
    Wymeersch, Henk
    Chalmers University of Technology, Sweden.
    MAC Delay in Belief Consensus for Distributed Tracking2013In: IEEE Proc. of 10th Workshop on Positioning, Navigation and Communication (WPNC), IEEE , 2013, p. 1-6Conference paper (Refereed)
    Abstract [en]

    In target tracking applications where many sensors must have a common view of the target’s state, distributed particle filtering with belief consensus is an attractive solution. It allows for a fully distributed, scalable solution, guarantees exact consensus in connected networks, and convergences fast for network with high connectivity. However, for medium access control, high connectivity is detrimental, possibly leading to a different convergence/performance trade-off. We study the delay/performance trade-off of distributed particle filtering with belief consensus in the presence of time division medium access control. We found that for small networks, (i) the impact of max-consensus should be accounted for; (ii) a simple schedule combined with a large communication range gives the best delay/performance trade-off.

    Download full text (pdf)
    fulltext
  • 4.
    Penna, Federico
    et al.
    Politecnico di Torino, Italy.
    Wymeersch, Henk
    Chalmers University of Technology, Sweden.
    Savic, Vladimir
    Technical University of Madrid, Spain.
    Uniformly reweighted belief propagation for distributed Bayesian hypothesis testing2011In: Proc. of IEEE Statistical Signal Processing Workshop (SSP), 2011, p. 733-736Conference paper (Refereed)
    Abstract [en]

    Belief propagation (BP) is a technique for distributed inference in wireless networks and is often used even when the underlying graphical model contains cycles. In this paper, we propose a uniformly reweighted BP scheme that reduces the impact of cycles by weighting messages by a constant “edge appearance probability” ρ ≤ 1. We apply this algorithm to distributed binary hypothesis testing problems (e.g., distributed detection) in wireless networks with Markov random field models. We demonstrate that in the considered setting the proposed method outperforms standard BP, while maintaining similar complexity. We then show that the optimal ρ can be approximated as a simple function of the average node degree, and can hence be computed in a distributed fashion through a consensus algorithm.

    Download full text (pdf)
    fulltext
  • 5.
    Savic, Vladimir
    Technical University of Madrid, Spain.
    Nonparametric Message Passing Methods for Cooperative Localization and Tracking2012Doctoral thesis, monograph (Other academic)
    Abstract [en]

    The objective of this thesis is the development of cooperative localization and tracking algorithms using nonparametric message passing techniques. In contrast to the most well-known techniques, the goal is to estimate the posterior probability density function (PDF) of the position of each sensor. This problem can be solved using Bayesian approach, but it is intractable in general case. Nevertheless, the particle-based approximation (via nonparametric representation), and an appropriate factorization of the joint PDFs (using message passing methods), make Bayesian approach acceptable for inference in sensor networks. The well-known method for this problem, nonparametric belief propagation (NBP), can lead to inaccurate beliefs and possible non-convergence in loopy networks. Therefore, we propose four novel algorithms which alleviate these problems: nonparametric generalized belief propagation (NGBP) based on junction tree (NGBP-JT), NGBP based on pseudo-junction tree (NGBP-PJT), NBP based on spanning trees (NBP-ST), and uniformly-reweighted NBP (URW-NBP). We also extend NBP for cooperative localization in mobile networks. In contrast to the previous methods, we use an optional smoothing, provide a novel communication protocol, and increase the efficiency of the sampling techniques. Moreover, we propose novel algorithms for distributed tracking, in which the goal is to track the passive object which cannot locate itself. In particular, we develop distributed particle filtering (DPF) based on three asynchronous belief consensus (BC) algorithms: standard belief consensus (SBC), broadcast gossip (BG), and belief propagation (BP). Finally, the last part of this thesis includes the experimental analysis of some of the proposed algorithms, in which we found that the results based on real measurements are very similar with the results based on theoretical models.

    Download full text (pdf)
    fulltext
  • 6.
    Savic, Vladimir
    et al.
    Technical University of Madrid, Spain.
    Athalye, Akshay
    Stony Brook University, USA.
    Bolic, Miodrag
    University of Ottawa, Canada.
    Djuric, Petar M.
    Stony Brook University, USA.
    Particle filtering for indoor RFID tag tracking2011In: Proc. of IEEE Statistical Signal Processing Workshop (SSP), 2011, p. 193-196Conference paper (Refereed)
    Abstract [en]

    In this paper, we propose a particle filtering (PF) method for indoor tracking using radio frequency identification (RFID) based on aggregated binary measurements. We use an Ultra High Frequency (UHF) RFID system that is composed of a standard RFID reader, a large set of standard passive tags whose locations are known, and a newly designed, special semi-passive tag attached to an object that is tracked. This semi-passive tag has the dual ability to sense the backscatter communication between the reader and other passive tags which are in its proximity and to communicate this sensed information to the reader using backscatter modulation. We refer to this tag as a sense-a-tag (ST). Thus, the ST can provide the reader with information that can be used to determine the kinematic parameters of the object on which the ST is attached. We demonstrate the performance of the method with data obtained in a laboratory environment.

    Download full text (pdf)
    fulltext
  • 7.
    Savic, Vladimir
    et al.
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, The Institute of Technology.
    Ferrer-Coll, Javier
    University of Gävle, Sweden.
    Ängskog, Per
    University of Gävle, Sweden.
    Chilo, José
    University of Gävle, Sweden.
    Stenumgaard, Peter
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, The Institute of Technology. Swedish Defense Research Agency (FOI), Linköping, Sweden.
    Larsson, Erik G.
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, The Institute of Technology.
    Measurement Analysis and Channel Modeling for TOA-Based Ranging in Tunnels2015In: IEEE Transactions on Wireless Communications, ISSN 1536-1276, E-ISSN 1558-2248, Vol. 14, no 1, p. 456-467Article in journal (Refereed)
    Abstract [en]

    A robust and accurate positioning solution is required to increase the safety in GPS-denied environments. Although there is a lot of available research in this area, little has been done for confined environments such as tunnels. Therefore, we organized a measurement campaign in a basement tunnel of Linköping university, in which we obtained ultra-wideband (UWB) complex impulse responses for line-of-sight (LOS), and three non-LOS (NLOS) scenarios. This paper is focused on time-of-arrival (TOA) ranging since this technique can provide the most accurate range estimates, which are required for range-based positioning. We describe the measurement setup and procedure, select the threshold for TOA estimation, analyze the channel propagation parameters obtained from the power delay profile (PDP), and provide statistical model for ranging. According to our results, the rise-time should be used for NLOS identification, and the maximum excess delay should be used for NLOS error mitigation. However, the NLOS condition cannot be perfectly determined, so the distance likelihood has to be represented in a Gaussian mixture form. We also compared these results with measurements from a mine tunnel, and found a similar behavior.

    Download full text (pdf)
    fulltext
  • 8.
    Savic, Vladimir
    et al.
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, Faculty of Science & Engineering.
    Larsson, Erik G.
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, Faculty of Science & Engineering.
    Experimental Study of Indoor Tracking Using UWB Measurements and Particle Filtering2016In: 2016 IEEE 17TH INTERNATIONAL WORKSHOP ON SIGNAL PROCESSING ADVANCES IN WIRELESS COMMUNICATIONS (SPAWC), IEEE , 2016Conference paper (Refereed)
    Abstract [en]

    Target tracking with ultra-wideband (UWB) signals in indoor environments is a challenging problem due to the presence of multipath and non-line-of-sight conditions (NLOS). A solution to this problem is to use particle filtering (PF), which is able to handle both nonlinear models and non-Gaussian uncertainties that typically appear in the presence of NLOS. In this paper, we compare four different PF variants, that differ in terms of how  NLOS measurements are handled. According to our experimental results, based on the measurements from a basement tunnel,    multiple features from the UWB impulse response should be used, and  the ranging likelihood function should make use of both LOS and NLOS measurements. Standard time-of-arrival (TOA) based methods, even with NLOS rejection, are not good enough. Instead we advocate TOA-based algorithms that can actively mitigate errors due to NLOS.

    Download full text (pdf)
    fulltext
  • 9.
    Savic, Vladimir
    et al.
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, Faculty of Science & Engineering.
    Larsson, Erik G.
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, Faculty of Science & Engineering.
    Fingerprinting-Based Positioning in Distributed Massive MIMO Systems2015In: Proc. of IEEE 82nd Vehicular Technology Conference (VTC2015-Fall), Sept. 2015., Institute of Electrical and Electronics Engineers (IEEE), 2015Conference paper (Refereed)
    Abstract [en]

    Location awareness in wireless networks may enable many applications such as emergency services, autonomous driving and geographic routing. Although there are many available positioning techniques, none of them is adapted to work with massive multiple-in-multiple-out (MIMO) systems, which represent a leading 5G technology candidate. In this paper, we discuss possible solutions for positioning of mobile stations using a vector of signals at the base station, equipped with many antennas distributed over deployment area. Our main proposal is to use fingerprinting techniques based on a vector of received signal strengths. This kind of methods are able to work in highly-cluttered multipath environments, and require just one base station, in contrast to standard range-based and angle-based techniques. We also provide a solution for fingerprinting-based positioning based on Gaussian process regression, and discuss main applications and challenges.

    Download full text (pdf)
    fulltext
  • 10.
    Savic, Vladimir
    et al.
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, Faculty of Science & Engineering.
    Larsson, Erik G.
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, Faculty of Science & Engineering.
    Ferrer-Coll, Javier
    University of Gavle, Sweden.
    Stenumgaard, Peter
    Swedish Defense Research Agency (FOI), Linköping, Sweden.
    Kernel Methods for Accurate UWB-Based Ranging with Reduced Complexity2016In: IEEE Transactions on Wireless Communications, ISSN 1536-1276, E-ISSN 1558-2248, Vol. 15, no 3, p. 1783-1793Article in journal (Refereed)
    Abstract [en]

    Accurate and robust positioning in multipath environments can enable many applications, such as search-and-rescue and asset tracking. For this problem, ultra-wideband (UWB) technology can provide the most accurate range estimates, which are required for range-based positioning. However, UWB still faces a problem with non-line-of-sight (NLOS) measurements, in which the range estimates based on time-of-arrival (TOA) will typically be positively biased. There are many techniques that address this problem, mainly based on NLOS identification and NLOS error mitigation algorithms. However, these techniques do not exploit all available information in the UWB channel impulse response. Kernel-based machine learning methods, such as Gaussian Process Regression (GPR), are able to make use of all information, but they may be too complex in their original form. In this paper, we propose novel ranging methods based on kernel principal component analysis (kPCA), in which the selected channel parameters are projected onto a nonlinear orthogonal high-dimensional space, and a subset of these projections is then used as an input for ranging. We evaluate the proposed methods using real UWB measurements obtained in a basement tunnel, and found that one of the proposed methods is able to outperform state-of-the-art, even if little training samples are available.

    Download full text (pdf)
    fulltext
  • 11.
    Savic, Vladimir
    et al.
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, The Institute of Technology.
    Larsson, Erik G.
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, The Institute of Technology.
    Ferrer-Coll, Javier
    University of Gävle, Sweden.
    Stenumgaard, Peter
    Swedish Defense Research Agency (FOI), Linköping, Sweden.
    Kernel Principal Component Analysis for UWB-Based Ranging2014In: IEEE International Workshop on Signal Processing Advances in Wireless Communications (SPAWC), 2014, 2014, p. 145-149Conference paper (Refereed)
    Abstract [en]

    Accurate positioning in harsh environments can enable many application, such as search-and-rescue in emergency situations. For this problem, ultra-wideband (UWB) technology can provide the most accurate range estimates, which are required for range-based positioning. However, it still faces a problem in non-line-of-sight (NLOS) environments, in which range estimates based on time-of-arrival (TOA) are positively biased. There are many techniques that try to address this problem, mainly based on NLOS identification and NLOS error mitigation. However, these techniques do not exploit all available information from the UWB channel impulse response. In this paper, we propose a novel ranging technique based on kernel principal component analysis (kPCA), in which the selected channel parameters are projected onto nonlinear orthogonal high-dimensional space, and a subset of these projections is then used for ranging. We tested this technique using UWB measurements obtained in a basement tunnel of Linkoping university, and found that it provides much better ranging performance comparing with standard techniques based on PCA and TOA.

    Download full text (pdf)
    fulltext
  • 12.
    Savic, Vladimir
    et al.
    Technical University of Madrid, Spain.
    Poblacion, Adrian
    Technical University of Madrid, Spain.
    Zazo, Santiago
    Technical University of Madrid, Spain.
    Garcia, Mariano
    Technical University of Madrid, Spain.
    An Experimental Study of RSS-Based Indoor Localization Using Nonparametric Belief Propagation Based on Spanning Trees2010In: Proc. of Intl. Conf. on Sensor Technologies and Applications, 2010, p. 238-243Conference paper (Refereed)
    Abstract [en]

    Nonparametric belief propagation (NBP) is the well-known method for cooperative localization in wireless sensor networks. It is capable to provide information about location estimation with appropriate uncertainty and to accommodate non-Gaussian distance measurement errors. However, the accuracy of NBP is questionable in loopy networks. Therefore, in this paper, we propose a novel approach, NBP based on spanning trees (NBP-ST) created by breadth first search (BFS) method. In addition, we propose a reliable indoor model based on obtained received-signal-strength (RSS) measurements in our lab. According to our experimental results, NBP-ST performs better than NBP in terms of accuracy and communication cost in the networks with high connectivity (i.e., highly loopy networks).

    Download full text (pdf)
    fulltext
  • 13.
    Savic, Vladimir
    et al.
    Technical University of Madrid, Spain.
    Poblacion, Adrian
    Technical University of Madrid, Spain.
    Zazo, Santiago
    Technical University of Madrid, Spain.
    Garcia, Mariano
    Technical University of Madrid, Spain.
    Indoor Positioning Using Nonparametric Belief Propagation Based on Spanning Trees2010In: EURASIP Journal on Wireless Communications and Networking, ISSN 1687-1472, E-ISSN 1687-1499, p. 1-12Article in journal (Refereed)
    Abstract [en]

    Nonparametric belief propagation (NBP) is one of the best-known methods for cooperative localization in sensor networks. It is capable of providing information about location estimation with appropriate uncertainty and to accommodate non-Gaussian distance measurement errors. However, the accuracy of NBP is questionable in loopy networks. Therefore, in this paper, we propose a novel approach, NBP based on spanning trees (NBP-ST) created by breadth first search (BFS) method. In addition, we propose a reliable indoor model based on obtained measurements in our lab. According to our simulation results, NBP-ST performs better than NBP in terms of accuracy and communication cost in the networks with high connectivity (i.e., highly loopy networks). Furthermore, the computational and communication costs are nearly constant with respect to the transmission radius. However, the drawbacks of proposed method are a little bit higher computational cost and poor performance in low-connected networks.

    Download full text (pdf)
    fulltext
  • 14.
    Savic, Vladimir
    et al.
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, The Institute of Technology.
    Wymeersch, Henk
    Chalmers University of Technology, Sweden.
    Simultaneous Localization and Tracking via Real-time Nonparametric Belief Propagation2013In: EIII  International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2013, IEEE , 2013Conference paper (Refereed)
    Abstract [en]

    Target tracking in wireless sensor networks is traditionally achieved by localization and tracking (LAT), where the sensors are first localized, and in a later stage the target is tracked. This approach is sub-optimal since the sensor-target observations are not used to refine the position estimates of the sensors. In contrast, simultaneous localization and tracking (SLAT) uses these observations to track the target while simultaneously localizing the sensors. In this paper, we propose a novel centralized SLAT method based on real-time nonparametric belief propagation, which has nearly the same complexity and the same communication cost as LAT, and can provide both sensors' and target's estimated distributions in non-Gaussian form.

    Download full text (pdf)
    fulltext
  • 15.
    Savic, Vladimir
    et al.
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, The Institute of Technology.
    Wymeersch, Henk
    Chalmers University of Technology, Sweden.
    Larsson, Erik G.
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, The Institute of Technology.
    Simultaneous Sensor Localization and Target Tracking in Mine Tunnels2013In: IEEE Proc. of International Conference on Information Fusion, 2013, p. 1427-1433Conference paper (Refereed)
    Abstract [en]

    Mine tunnels are extensive labyrinths with irregularly-shaped walls, in which a hundreds of employees are working on extraction of valuable ores and minerals. Since the working conditions are extremely hazardous, a (wireless) sensor network (WSN) is deployed to increase the safety in tunnels. One of the most important applications of WSN is to track the personnel, mobile equipment and vehicles. However, the state-of-the-art algorithms assume that the positions of the sensors are perfectly known, which is not necessarily true due to the imprecise placement and/or possible sensor drops. Therefore, we propose an automatic approach for simultaneous refinement ofsensors’ positions (localization) and target tracking. We use a measurement model from a real mine, and apply a discrete variant of real-time belief propagation, which can efficiently solve this high-dimensional problem, and handle all non-Gaussian uncertainties typical for mining environments. Comparing with standard Bayesian target tracking and localization algorithms, both the sensors’ and the target’s estimates are improved.

    Download full text (pdf)
    fulltext
  • 16.
    Savic, Vladimir
    et al.
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, The Institute of Technology.
    Wymeersch, Henk
    Chalmers University of Technology, Gothenburg, Sweden.
    Larsson, Erik G.
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, The Institute of Technology.
    Target Tracking in Confined Environments with Uncertain Sensor Positions2016In: IEEE Transactions on Vehicular Technology, ISSN 0018-9545, E-ISSN 1939-9359, Vol. 65, no 2, p. 870-882Article in journal (Refereed)
    Abstract [en]

    To ensure safety in confined environments such as mines or subway tunnels, a (wireless) sensor network can be deployed to monitor various environmental conditions. One of its most important applications is to track personnel, mobile equipment and vehicles. However, the state-of-the-art algorithms assume that the positions of the sensors are perfectly known, which is not necessarily true due to  imprecise placement and/or dropping of sensors. Therefore, we propose an automatic approach for simultaneous refinement of sensors' positions and target tracking. We divide the considered area in a finite number of cells, define dynamic and measurement models, and apply a discrete variant of belief propagation which can efficiently solve this high-dimensional problem, and handle all non-Gaussian uncertainties expected in this kind of environments. Finally, we use ray-tracing simulation to generate an artificial mine-like environment and generate synthetic measurement data. According to our extensive simulation study, the proposed approach performs significantly better than standard Bayesian target tracking and localization algorithms, and provides robustness against outliers.

    Download full text (pdf)
    fulltext
  • 17.
    Savic, Vladimir
    et al.
    Technical University of Madrid, Spain.
    Wymeersch, Henk
    Chalmers University of Technology, Sweden.
    Penna, Federico
    Politecnico di Torino, Italy.
    Zazo, Santiago
    Technical University of Madrid, Spain.
    Optimized edge appearance probability for cooperative localization based on tree-reweighted nonparametric belief propagation2011In: Proc. of IEEE Intl. Conf. on Acoustics, Speech and Signal Processing (ICASSP), 2011, p. 3028-3031Conference paper (Refereed)
    Abstract [en]

    Nonparametric belief propagation (NBP) is a well-known particle-based method for distributed inference in wireless networks. NBP has a large number of applications, including cooperative localization. However, in loopy networks NBP suffers from similar problems as standard BP, such as over-confident beliefs and possible non-convergence. Tree-reweighted NBP (TRW-NBP) can mitigate these problems, but does not easily lead to a distributed implementation due to the non-local nature of the required so-called edge appearance probabilities. In this paper, we propose a variation of TRW-NBP, suitable for cooperative localization in wireless networks. Our algorithm uses a fixed edge appearance probability for every edge, and can outperform standard NBP in dense wireless networks.

    Download full text (pdf)
    fulltext
  • 18.
    Savic, Vladimir
    et al.
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, The Institute of Technology.
    Wymeersch, Henk
    Chalmers University of Technology, Gothenburg, Sweden.
    Zazo, Santiago
    Technical University of Madrid, Spain.
    Belief consensus algorithms for fast distributed target tracking in wireless sensor networks2014In: Signal Processing, ISSN 0165-1684, E-ISSN 1872-7557, Vol. 95, p. 149-160Article in journal (Refereed)
    Abstract [en]

    In distributed target tracking for wireless sensor networks, agreement on the target state can be achieved by the construction and maintenance of a communication path, in order to exchange information regarding local likelihood functions. Such an approach lacks robustness to failures and is not easily applicable to ad-hoc networks. To address this, several methods have been proposed that allow agreement on the global likelihood through fully distributed belief consensus (BC) algorithms, operating on local likelihoods in distributed particle filtering (DPF). However, a unified comparison of the convergence speed and communication cost has not been performed. In this paper, we provide such a comparison and propose a novel BC algorithm based on belief propagation (BP). According to our study, DPF based on metropolis belief consensus (MBC) is the fastest in loopy graphs, while DPF based on BP consensus is the fastest in tree graphs. Moreover, we found that BC-based DPF methods have lower communication overhead than data flooding when the network is sufficiently sparse.

    Download full text (pdf)
    fulltext
  • 19.
    Savic, Vladimir
    et al.
    Technical University of Madrid, Spain.
    Wymeersch, Henk
    Chalmers University of Technology, Sweden.
    Zazo, Santiago
    Technical University of Madrid, Spain.
    Distributed Target Tracking based on Belief Propagation Consensus2012In: Proc. of the 20th European Signal Processing Conference (EUSIPCO), 2012, p. 544-548Conference paper (Refereed)
    Abstract [en]

    Distributed target tracking in wireless sensor networks (WSN) is an important problem, in which agreement on the target state can be achieved using particle filters with standard consensus methods, which may take long to converge. We propose distributed particle filtering based on belief propagation (DPF-BP) consensus, a fast method for target tracking. According to our simulations, DPF-BP provides better performance than DPF based on standard belief consensus (DPF-SBC) in terms of disagreement in the network. However, in terms of root-mean square error, it can outperform DPF-SBC only for a specific number of consensus iterations.

    Download full text (pdf)
    fulltext
  • 20.
    Savic, Vladimir
    et al.
    Technical University of Madrid, Spain.
    Zazo, Santiago
    Technical University of Madrid, Spain.
    Belief Propagation Techniques for Cooperative Localization in Wireless Sensor Networks2011In: Handbook of Position Location: Theory, Practice, and Advances / [ed] Seyed A. (Reza) Zekavat, R. Michael Buehrer, Hoboken, NJ, USA: John Wiley & Sons, 2011, p. 837-869Chapter in book (Other academic)
    Abstract [en]

    A number of applications in wireless sensor networks (WSN) require sensor nodes to obtain their absolute or relative positions. Equipping every sensor with a GPS receiver may be expensive, energy prohibitive and limited to outdoor applications. Therefore, we consider cooperative localization where each sensor node with unknown location obtains its location by cooperating with neighboring sensor nodes. In this chapter, we apply probabilistic inference to the problem of cooperative localization. These techniques are capable to obtain, not only location estimates, but also a measure of the uncertainty of those estimates. Since these methods are computationally very expensive, we need to use message-passing methods, which are also known as belief propagation (BP) methods. BP is a way of organizing the global computation of marginal beliefs in terms of smaller local computations within the graph. It is one of the best-known probabilistic methods for distributed inference in statistical physics, articial intelligence, computer vision, error-correcting codes, localization, etc. The whole computation takes a time proportional to the number of links in the graph, which is significantly less than the exponentially large time that would be required to compute marginal probabilities naively. However, due to the presence of nonlinear relationships and highly non-Gaussian uncertainties the standard BP algorithm is undesirable. Nevertheless, a particle-based approximation via nonparametric belief propagation (NBP) makes BP acceptable for localization in sensor networks. In this chapter, after an introduction to cooperative localization, we describe BP/NBP techniques and its generalizations (GBP) for the loopy networks. Due to the poor performance of BP/NBP methods in loopy networks, we describe three improved methods: GBP based on Kikuchi approximation (GBP-K), nonparametric GBP based on junction tree (NGBP-JT), and NBP based on spanning trees (NBP-ST). The last one (NBP-ST) is currently a unique method which is computationally feasible in a large-scale WSN.

  • 21.
    Savic, Vladimir
    et al.
    Technical University of Madrid, Spain.
    Zazo, Santiago
    Technical University of Madrid, Spain.
    Nonparametric Belief Propagation Based on Spanning Trees for Cooperative Localization in Wireless Sensor Networks2010In: Proc. of IEEE 72nd Vehicular Technology Conference Fall (VTC 2010-Fall), 2010, p. 1-5Conference paper (Refereed)
    Abstract [en]

    Nonparametric belief propagation (NBP) is one of the best-known methods for cooperative localization in sensor networks. It is capable to provide information about location estimation with appropriate uncertainty and to accommodate non-Gaussian distance measurement errors. However, the accuracy of NBP is questionable in loopy networks. In this paper, we propose a novel approach, NBP based on spanning trees (NBP-ST) created by breadth first search (BFS) method. According to our simulation results, NBP-ST performs better than NBP in terms of accuracy, computational and communication cost in the networks with high connectivity (i.e., highly loopy networks).

    Download full text (pdf)
    fulltext
  • 22.
    Savic, Vladimir
    et al.
    Technical University of Madrid, Spain.
    Zazo, Santiago
    Technical University of Madrid, Spain.
    Nonparametric Boxed Belief Propagation for Localization in Wireless Sensor Networks2009In: IEEE Proc. of Intl. Conf. on Sensor Technologies and Applications, 2009, p. 520-525Conference paper (Refereed)
    Abstract [en]

    Obtaining estimates of each sensorpsilas position as well as accurately representing the uncertainty of that estimate is a critical step for effective application of wireless sensor networks (WSN). Nonparametric belief propagation (NBP) is a popular localization method which uses particle based approximation of belief propagation. In this paper, we present a new variant of NBP method which we call nonparametric boxed belief propagation (NBBP). The main idea is to constraint the area from which the samples are drawn by building a box that covers the region where anchorspsila radio ranges overlap. These boxes, which are created almost without any additional communication between nodes, are also used to filter erroneous samples of the beliefs. In order to decrease the computational and the communication cost, we also added incremental approach using Kullback-Leibler (KL) divergence as a convergence parameter. Simulation results show that accuracy, computational and communication cost are significantly improved.

    Download full text (pdf)
    fulltext
  • 23.
    Savic, Vladimir
    et al.
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, The Institute of Technology.
    Zazo, Santiago
    Technical University of Madrid, Spain.
    Nonparametric generalized belief propagation based on pseudo-junction tree for cooperative localization in wireless networks2013In: EURASIP Journal on Advances in Signal Processing, ISSN 1687-6172, E-ISSN 1687-6180, Vol. 16Article in journal (Refereed)
    Abstract [en]

    Non-parametric belief propagation (NBP) is a well-known message passing method for cooperative localization in wireless networks. However, due to the over-counting problem in the networks with loops, NBP’s convergence is not guaranteed, and its estimates are typically less accurate. One solution for this problem is non-parametric generalized belief propagation based on junction tree. However, this method is intractable in large-scale networks due to the high-complexity of the junction tree formation, and the high-dimensionality of the particles. Therefore, in this article, we propose the non-parametric generalized belief propagation based on pseudo-junction tree (NGBP-PJT). The main difference comparing with the standard method is the formation of pseudo-junction tree, which represents the approximated junction tree based on thin graph. In addition, in order to decrease the number of high-dimensional particles, we use more informative importance density function, and reduce the dimensionality of the messages. As by-product, we also propose NBP based on thin graph (NBP-TG), a cheaper variant of NBP, which runs on the same graph as NGBP-PJT. According to our simulation and experimental results, NGBP-PJT method outperforms NBP and NBP-TG in terms of accuracy, computational, and communication cost in reasonably sized networks.

    Download full text (pdf)
    fulltext
  • 24.
    Savic, Vladimir
    et al.
    Technical University of Madrid, Spain.
    Zazo, Santiago
    Technical University of Madrid, Spain.
    Pseudo-junction tree method for cooperative localization in wireless sensor networks2010In: IEEE Proc. of Intl. Conf. on Information Fusion (FUSION), 2010, p. 1-8Conference paper (Refereed)
    Abstract [en]

    Nonparametric belief propagation (NBP) is well-known probabilistic method for cooperative localization in sensor networks. However, due to the double counting problem, NBP convergence is not guaranteed in the networks with loops or even if NBP converges, it could provide us less accurate estimates. The well-known solution for this problem is nonparametric generalized belief propagation based on junction tree (NGBP-JT). However, there are two problems: how to efficiently form the junction tree in an arbitrary network, and how to decrease the number of particles while keeping the good performance. Therefore, in this paper, we propose the formation of pseudo-junction tree (PJT), which represents the approximated junction tree based on thin graph. In addition, in order to decrease the number of particles, we use a set of very strong constraints. The resulting localization method, NGBP based on PJT (NGBP-PJT), overperforms NBP in terms of accuracy and communication cost in any arbitrary network.

    Download full text (pdf)
    fulltext
  • 25.
    Savic, Vladimir
    et al.
    Technical University of Madrid, Spain.
    Zazo, Santiago
    Technical University of Madrid, Spain.
    Reducing Communication Overhead for Cooperative Localization Using Nonparametric Belief Propagation2012In: IEEE Wireless Communication Letters, ISSN 2162-2337, Vol. 1, no 4, p. 308-311Article in journal (Refereed)
    Abstract [en]

    A number of methods for cooperative localization has been proposed, but most of them provide only location estimate, without associated uncertainty. On the other hand, nonparametric belief propagation (NBP), which provides approximated posterior distributions of the location estimates, is expensive mostly because of the transmission of the particles. In this paper, we propose a novel approach to reduce communication overhead for cooperative positioning using NBP. It is based on: i) communication of the beliefs (instead of the messages), ii) approximation of the belief with Gaussian mixture of very few components, and iii) censoring. According to our simulations results, these modifications reduce significantly communication overhead while providing the estimates almost as accurate as the transmission of the particles.

    Download full text (pdf)
    fulltext
  • 26.
    Savic, Vladimir
    et al.
    Technical University of Madrid, Spain.
    Zazo, Santiago
    Technical University of Madrid, Spain.
    Sensor Localization using Generalized Belief Propagation in Networks with Loops2009In: Proc. of the 17th European Signal Processing Conference (EUSIPCO), 2009, p. 75-79Conference paper (Refereed)
    Abstract [en]

    Belief propagation (BP), also called “sum-product algorithm”, is one of the best-known graphical model for inference in statistical physics, artificial intelligence, computer vision, etc. Furthermore, a recent research in distributed sensor network localization showed us that BP is an efficient way to obtain sensor location as well as appropriate uncertainty. However, BP convergence is not guaranteed in a network with loops. In this paper, we propose localization using generalized belief propagation based on junction tree (GBP-JT) method. We illustrate it in a network with loop where BP shows poor performance. In fact, we compared estimated locations with Nonparametric Belief Propagation (NBP) algorithm. According to our simulation results, GBP-JT resolved the problems with loops, but the price for this is unacceptable large computational cost. The main conclusion is that this algorithm could be used with some approximation which keeps improved accuracy and significantly decreases the computational cost.

    Download full text (pdf)
    fulltext
  • 27.
    Savic, Vladimir
    et al.
    Technical University of Madrid, Spain.
    Zazo, Santiago
    Technical University of Madrid, Spain.
    Sensor localization using nonparametric generalized belief propagation in network with loop2009In: IEEE Proc. of Intl. Conf. on Information Fusion (FUSION), 2009Conference paper (Refereed)
    Abstract [en]

    Belief propagation (BP) is one of the best-known graphical model for inference in statistical physics, artificial intelligence, computer vision, etc. Furthermore, a recent research in distributed sensor network localization showed us that BP is an efficient way to obtain sensor location as well as appropriate uncertainty. However, BP convergence is not guaranteed in a network with loops. In this paper, we propose localization using generalized belief propagation based on junction tree method (GBP-JT) and nonparametric (particle-based) approximation of this algorithm (NGBP-JT). We illustrate it in a network with loop where BP shows poor performance. In fact, we compared estimated locations with nonparametric belief propagation (NBP) algorithm. According to our simulation results, GBP-JT resolved the problems with loops, but the price for this is unacceptable large computational cost. Therefore, our approximated version of this algorithm, NGBP-JT, reduced significantly this cost, with little effect on accuracy.

    Download full text (pdf)
    fulltext
  • 28.
    Wymeersch, Henk
    et al.
    Chalmers University of Technology, Sweden.
    Penna, Federico
    Politecnico di Torino, Italy.
    Savic, Vladimir
    Technical University of Madrid, Spain.
    Uniformly reweighted belief propagation: A factor graph approach2011In: Proc. of IEEE Intl. Symp. on Information Theory Proceedings (ISIT), 2011, p. 2000-2004Conference paper (Refereed)
    Abstract [en]

    Tree-reweighted belief propagation is a message passing method that has certain advantages compared to traditional belief propagation (BP). However, it fails to outperform BP in a consistent manner, does not lend itself well to distributed implementation, and has not been applied to distributions with higher-order interactions. We propose a method called uniformly-reweighted belief propagation that mitigates these drawbacks. After having shown in previous works that this method can substantially outperform BP in distributed inference with pairwise interaction models, in this paper we extend it to higher-order interactions and apply it to LDPC decoding, leading performance gains over BP.

    Download full text (pdf)
    fulltext
  • 29.
    Wymeersch, Henk
    et al.
    Chalmers University of Technology, Sweden.
    Penna, Federico
    Politecnico di Torino, Italy.
    Savic, Vladimir
    Technical University of Madrid, Spain.
    Uniformly Reweighted Belief Propagation for Estimation and Detection in Wireless Networks2012In: IEEE Transactions on Wireless Communications, ISSN 1536-1276, E-ISSN 1558-2248, Vol. 11, no 4, p. 1587-1595Article in journal (Refereed)
    Abstract [en]

    In this paper, we propose a new inference algorithm, suitable for distributed processing over wireless networks. The algorithm, called uniformly reweighted belief propagation (URW-BP), combines the local nature of belief propagation with the improved performance of tree-reweighted belief propagation (TRW-BP) in graphs with cycles. It reduces the degrees of freedom in the latter algorithm to a single scalar variable, the uniform edge appearance probability ρ. We provide a variational interpretation of URW-BP, give insights into good choices of ρ, develop an extension to higher-order potentials, and complement our work with numerical performance results on three inference problems in wireless communication systems: spectrum sensing in cognitive radio, cooperative positioning, and decoding of a low-density parity-check (LDPC) code.

    Download full text (pdf)
    fulltext
  • 30.
    Wymeersch, Henk
    et al.
    Chalmers University of Technology, Sweden.
    Penna, Federico
    Fraunhofer Heinrich Hertz Institute, Berlin, Germany.
    Savic, Vladimir
    Linköping University, Department of Electrical Engineering, Communication Systems. Linköping University, The Institute of Technology.
    Zhao, Jun
    Chalmers University of Technology, Sweden.
    Comparison of reweighted message passing algorithms for LDPC decoding2013In: IEEE International Conference on Communications (ICC), 2013, IEEE , 2013, p. 3264-3269Conference paper (Refereed)
    Abstract [en]

    Low density parity check (LDPC) codes can be decoded with a variety of decoding algorithms, offering a trade-off in terms of complexity, latency, and performance. We describe seven distinct LDPC decoders and provide a performance comparison for a practical regular LDPC code. Our simulations indicate that the best performance/latency trade-off is achieved by one version ofthe reweighted max-product decoder. When latency is not an issue, the traditional sum-product decoder yields the best performance.

    Download full text (pdf)
    fulltext
1 - 30 of 30
CiteExportLink to result list
Permanent link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf