The dempstershafer ds theory is an excellent method of information fusion. Because of the difference which is caused by the sensors, it is essential to deal with the evidence with a method of. In the absence of information, these methods must then make assumptions about the sensor data. The concept of multisensor information fusion model based on the dempstershafer theory is introduced. This naturally leads us to choose the dempstershafer approach as our first sensor fusion implementation algorithm. Thirdly, the belief function of the mean curvature. Multisensor fusion based on multiple classifier systems. Dempster shafer evidential theory, a probabilitybased data fusion classification algorithm, is useful when the sensors or more generally, the information sources contributing information cannot associate a 100 percent probability of certainty to their output decisions.
When a new event occurs, the nodes in the neighborhood of the event sense and then send many packets to the sink node. The dempster shafer ds theory is an excellent method of information fusion. This paper discusses the relationship between dempster shafer theory and the classical bayesian method, describes our sensor fusion research work using dempster shafer theory in comparison with. Introduction we build this paper on two previous papers, sensor fusion for context understanding 1 and sensor fusion using dempster shafer theory 2, presented at imtc2002.
A dempster shafer belief theoretic approach thanuka l. Another work in this area is the presentation of fuzzy logic approach to estimating tendencies in. The key contribution of this thesis is introducing the dempstershafer theory of. In a narrow sense, the term dempstershafer theory refers to the original conception of the theory by dempster and shafer. Sensor fusion for contextaware computing using dempster. In this paper, we apply the dempstershafer ds theory of evidence 31 to model imperfect sensor information since it can be used to model common types of sensor uncertainty e. In particular, the use of dempstershafer theory of evidence within markov models has brought some keys to several challenging difficulties that the conventional hidden markov models cannot handle.
Detailed applications of the dempstershafer theory can be found in eastman 1997. Novel data fusion algorithm based on eventdriven and. An introduction to bayesian and dempstershafer data fusion. Often used as a method of sensor fusion, dempstershafer theory is based on two ideas. Sensor data fusion for contextaware computing using dempster shafer theory huadong wu cmuritr0352 submitted in partial fulfillment of the requirements for the degree of doctor of philosophy in robotics thesis committee. This paper discusses the relationship between dempstershafer theory and the classical bayesian method, describes our sensor fusion research work using dempstershafer theory in comparison with the weighted sum of probability method. Firstly, the structured light imaging is used to acquire the 3d palmprint data. The aim of this paper is to investigate how to use the contextual knowledge in order to improve the fusion process. However, it is more common to use the term in the wider sense of the same general approach, as adapted to specific kinds of. Data fusion for fault diagnosis using dempstershafer theory. Pdf sensor fusion using dempstershafer theory for context.
An example of an operational dempstershafer id system is the canadian navy esm warning system, version 2 canews 42. The feasibility of this idea is demonstrated via a meetingparticipant. This naturally leads us to choose the dempster shafer approach as our first sensor fusion implementation algorithm. Novel paradigm for constructing masses in dempstershafer. Introduction we build this paper on two previous papers, sensor fusion for context understanding 1 and sensor fusion using dempstershafer theory 2, presented at imtc2002.
Multisensor image segmentation using dempstershafer fusion. The emergence of the benefits of ds methods were applied by robin murphy for robotic scene analysis 16. This paper proposed a novel 3d palmprint recognition algorithm by combining 3d palmprint features using ds fusion theory. Dempstershafer theory, on the other hand, handles this by assigning 0% belief to head and tail and assigning 100% belief to the set head, tail, meaning either head or tail. Dempsterschaffer theory for data fusion in communication. Dempstershafer theory for sensor f usion robin murph y. The dempstershafer theory, a generalization of the bayesian theory, is based on the idea of belief and as such can handle ignorance.
The principal task of such reasoning is data fusion, or evidence accumulation. The fusion result is a discrete value or a label, which is determined by the corresponding maximum probability values. The dempstershafer decision theory is essentially a generalized bayesian statistical theory. More generally, the ds approach allows for three kinds of answers. Sensor data fusion for contextaware computing using dempster. A dempster shafer theory based multiclass svm dsmsvm is constructed by using the combination rule of evidences. Pixel level data fusion using the dempstershafer theory was performed which consisted of assigning a degree of belief to a hypothesis and combining this information from multiple ndt systems. Dempsterschaffer theory for data fusion in communication networks prof. Introduction in his 1976 book, glenn shafer introduced the theory of evidence, later referred to as the dempster shafer theory dst. The concept of multisensor information fusion model based on the dempster shafer theory is introduced.
Such circumstances need their networks to be simultaneously reliable and eventdriven. Dempster shafer to update the occupancy grid and to generate raw conflict evidence. Ijccc was founded in 2006, at agora university, by ioan dzitac editorinchief, florin gheorghe filip editorinchief, and misujan manolescu managing editor. Mar 26, 2018 moreover, it should be remove redundant packets in order to lower the average energy consumption. A dempstershafer belief theoretic approach thanuka l. Dempstershafer ds theory is one of the main tools for reasoning about data obtained from multiple sources, subject to uncertain information 12, 16. Tracking methods using dempster shafer theory one of the earliest known works in applying dempster shafer ds methods to target tracking was by jean dezert for navigation 15, where the sensor is moving and the targets are stationary. When all of the required information is available, many data fusion methods provide a solid approach. Experimented their solution on nomadic super scout ii robot, they gave an example that despite of sensor fusion in robot localization using dsevidence theory with conflict detection using mahalanobis distance. Keywords dempster shafer evidence theory event, periodic trace, pattern fcm heterogeneous networks mipv6, home agent, binding update, ibgp, vharp, edge router model checking, atl, agents, jade, fsm sip seamless handover chaos collaborative filtering divisible load theory fuzzy ts model fuzzy logic systems genetic algorithm group decisionmaking multiattribute open, heterogeneous, dynamic and.
Data fusion andparameterestimation using qualitative models. The ds theory offers an alternative to the traditional probabilistic theory for the mathematical representation of uncertainty. An analytical framework for soft and hard data fusion. The use of locally weighted regression for the data fusion. System engineering 3team, the dempstershafer fusion approach is wellsuited for handling partial probability attributes, while the bayesian fusion approach is wellsuited for handling complete probability attributes. The theory of belief functions, also referred to as evidence theory or dempstershafer theory dst, is a general framework for reasoning with uncertainty, with understood connections to other frameworks such as probability, possibility and imprecise probability theories. The resulting information of the architecture is combined using similar sensor subset and dissimilar sensor subset. Contents list offigures xv listoftables xxi preface xxv chapter1 introduction 1 chapter2 multiplesensorsystemapplications,benefits, anddesign considerations 9 2. Furthermore, the reliability of a sensor can be updated by the fusion center using methods such as elouedi et al. A new multisensor information fusion model using dempster. Sensor fusion in robot localization using dsevidence theory. Fusionbased surveillance wsn deployment using dempster. Dempster in the context of statistical inference, the theory was later.
Faulttolerant interval estimation fusion by dempstershafer. Multisensor fusion based on multiple classifier systems for. Towards having computers understand human usersontext. The latest developments in markov models theory and their corresponding computational techniques have opened new rooms for image and signal modeling. Howthe information obtained from different sensing devices is combined to form a description of thesystem is the sensor fusion problem. Sensor data fusion using dsm theory for activity recognition.
System engineering 3team, the dempster shafer fusion approach is wellsuited for handling partial probability attributes, while the bayesian fusion approach is wellsuited for handling complete probability attributes. Data fusion andparameterestimation using qualitative. The ds theory of evidence is based on the classic works of dempster and shafer. Professor of electrical and computer engineering m. Sensor data fusion using dsm theory for activity recognition under uncertainty in homebased care. Data and information fusion, bayesian, dempstershafer, fuzzy logic, neural networks, visual. A data fusion algorithm based on eventdriven and dempstershafer evidence theory is proposed in this paper to reduce data packet quantities and reserve energy for wireless sensor networks upon detecting abnormal data. In this paper, we apply the dempster shafer ds theory of evidence 31 to model imperfect sensor information since it can be used to model common types of sensor uncertainty e.
Prioritizing intrusion analysis using dempstershafer theory. This paper discusses the relationship between dempster shafer theory and the classical bayesian method, describes our sensor fusion research work using dempster shafer theory in comparison with the weighted sum of probability method. Mobile robot selflocalization and local map alignment. Tracking methods using dempstershafer theory one of the earliest known works in applying dempstershafer ds methods to target tracking was by jean dezert for navigation 15, where the sensor is moving and the targets are stationary. Dempster shafer ds theory is one of the main tools for reasoning about data obtained from multiple sources, subject to uncertain information 12, 16. Byapplyingdempstersrule,wewouldconcludethat p1 hasabraintumour with probability 1. Moreover, it should be remove redundant packets in order to lower the average energy consumption. A dempstershafer theory based multiclass svm dsmsvm is constructed by using the combination rule of evidences. Dempstershafer to update the occupancy grid and to generate raw conflict evidence.
This data fusion method is based on the idea of belief and as such can handle ignorance. Expert systemsdempstershafer theory wikibooks, open books. One can assign a mass probability to the results from both ndt methods. Shafer theory of evidence using a large scale group decision making approach emrah koksalmis hezarfen aeronautics and space technologies institute, national defense university, istanbul, turkey. Overview of dempstershafer and belief function tracking methods.
Mobile robot selflocalization and local map alignment with a. Eddy current and ultrasound data fusion using dempster. The dempstershafer ds theory provides an efficient framework to implement multisensor data fusion. Dempstershafer theory, with an example given of fusion using the cornerstone of the theory known as dempsters rule. To deal with the distributed multisource multiclass problem, the dsmsvm is trained corresponding to each information source, and then the dempstershafer theory is used to combine these learned dsmsvms. Although sensors may be of good quality and provide accurate readings, several factors may affect the signal level received by a sensor at a given location elouedi et al. Sensor fusion using dempstershafer theory for contextaware hci conference paper pdf available in conference record ieee instrumentation and measurement technology conference 1. Contextdependent combination of sensor information in. The use of dempstershafer theory helps to reduce uncertainty and imprecision in sensor representation and increases reliability. The ds theorys applications range from expert decision support systems to multiattribute decisionmaking and data fusion. Dempstershafer fusion for personnel detection application of dempstershafer theory with ultrasonic microdoppler and pir sensors. Yes, no, or dont know, the last option of allowing ignorance makes a big difference in.
Dempstershafer evidential theory, a probabilitybased data fusion classification algorithm, is useful when the sensors or more generally, the information sources contributing information cannot associate a 100 percent probability of certainty to their output decisions. Dempstershafer theory is based on the nonclassical idea of mass as opposed to the wellunderstood probabilities of bayes theory. Apperionavarro, loughborough university school of electronic, electrical and systems engineering learning outcomes 1. Alina bruma, nicoleta iftimie, rozina steigmann, adriana savin, raimond grimberg subject. Mar 22, 2012 the dempster shafer theory, a generalization of the bayesian theory, is based on the idea of belief and as such can handle ignorance. Associate professor of electrical and computer engineering. Dempstershafer evidence theory dset is a flexible and popular paradigm for multisource data fusion in wireless sensor networks wsns. Dempster schaffer theory for data fusion in communication networks prof. Often used as a method of sensor fusion, dempster shafer theory is based on two ideas. Secondly, two types of unique features, including mean curvature feature and gaussian curvature feature, are extracted. Faulttolerant interval estimation fusion by dempster. Faulttolerant interval estimation fusion by dempstershafer theory baohua li yunmin zhu. Data fusion for fault diagnosis using dempstershafer.
Reliable contextual information of remotely monitoredpatients should be generated to prevent hazardous situationsand to provide pervasive services in. Eddy current and ultrasound data fusion using dempster shafer theory author. Contextdependent combination of sensor information in probability0. Abstractthis paper proposes an approach to the fusion of multimodal sensor data for the purpose of personnel intrusion detection. In the absence of information, these methods must then make. Its new feature is that it allows distributing support for a proposition e. To deal with the distributed multisource multiclass problem, the dsmsvm is trained corresponding to each information source, and then the dempster shafer theory is used to combine these learned dsmsvms. Yet, most do not have a good way of dealing with ignorance. In particular, we investigate the use of dempstershafer based data fusion model and present techniques for processing of visual sensor data to facilitate the use of dempstershafer model. Overview of dempstershafer and belief function tracking. To understand how dempster schafer theory can fuse beliefs in anomaly from multiple sensors 2. In a narrow sense, the term dempster shafer theory refers to the original conception of the theory by dempster and shafer. Both the flexibility and the difficulty consist in defining the probability mass function. Mobile robot selflocalization and loca l map alignment with a dempster shafer sensor fusion algorithm hyunki lee, xingyong song and hyungsuck cho department of mechanical engineering, korea advanced ins titute of science and technology, daejeon, korea tel.
A dempstershafer method for multisensor fusion by bethany. Sensor grids, flexible communications networks and webbased services. Keywords dempstershafer evidence theory event, periodic trace, pattern fcm heterogeneous networks mipv6, home agent, binding update, ibgp, vharp, edge router model checking, atl, agents, jade, fsm sip seamless handover chaos collaborative filtering divisible load theory fuzzy ts model fuzzy logic systems genetic algorithm group decisionmaking multiattribute open. Because of the difference which is caused by the sensors, it is essential to deal with the. Detection of waterquality contamination events based on. Dempstershafer theory for sensor f usion in autonomous mobile rob ots robin r.
391 843 1126 1454 489 991 1070 1538 679 1065 193 1536 402 460 648 900 1234 74 923 442 1607 24 1395 959 538 1058 618 361 860 387 986 1549 1390 347 966 1314 162 201 992 1181 1179 758