A Parallel Implementation of Data Fusion Algorithm Using Gamma

Mello, Rui R.; Almeida, Rubens H. P. de; França, Felipe M. G.; Paillard, Gabriel

doi:10.1109/sbac-padw.2015.25

Cited by 4 publications

(3 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In 2015, another Gamma parallel implementation was proposed, now using a parallel hardware with support to Graphics Processing Unit (GPU) [20]. The Gamma paradigm can be applied to many application domains, for instance, Image Processing [21] and data fusion for target tracking [1], among others.…”

Section: B Gammamentioning

confidence: 99%

See 1 more Smart Citation

Exploring the Equivalence between Dynamic Dataflow Model and Gamma - General Abstract Model for Multiset mAnipulation

Mello

Araújo

Alves

et al. 2019

2019 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW)

Self Cite

View full text Add to dashboard Cite

With the increase of the search for computational models where the expression of parallelism occurs naturally, some paradigms arise as options for the next generation of computers. In this context, dynamic Dataflow and Gamma -General Abstract Model for Multiset mAnipulation) -emerge as interesting computational models choices. In the dynamic Dataflow model, operations are performed as soon as their associated operators are available, without rely on a Program Counter to dictate the execution order of instructions. The Gamma paradigm is based on a parallel multiset rewriting scheme. It provides a non-deterministic execution model inspired by an abstract chemical machine metaphor, where operations are formulated as reactions that occur freely among matching elements belonging to the multiset. In this work, equivalence relations between the dynamic Dataflow and Gamma paradigms are exposed and explored, while methods to convert from Dataflow to Gamma paradigm and vice versa are provided. It is shown that vertices and edges of a dynamic Dataflow graph can correspond, respectively, to reactions and multiset elements in the Gamma paradigm. Implementation aspects of execution environments that could be mutually beneficial to both models are also discussed. This work provides the scientific community with the possibility of taking profit of both parallel programming models, contributing with a versatility component to researchers and developers. Finally, it is important to state that, to the best of our knowledge, the similarity relations between both dynamic Dataflow and Gamma models presented here have not been reported in any previous work.

show abstract

Section: B Gammamentioning

confidence: 99%

“…Parallel computing has been widely used as a tool to obtain performance in a scenario where the von Neumann architecture is close to the hardware's performance exploitation limits [1]. In the other words, despite the validity of Moore's law, possible technological improvements are not being converted into performance in a proportional way.…”

Section: Introductionmentioning

confidence: 99%

Exploring the Equivalence between Dynamic Dataflow Model and Gamma - General Abstract Model for Multiset mAnipulation

Mello

Araújo

Alves

et al. 2019

2019 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW)

Self Cite

View full text Add to dashboard Cite

show abstract

“…This approach is a graph-based method proposed in [ 11 ], which is PP (Pair of Plot [ 12 ]) based. In 2015, Mello et al proposed the implementation based on the model Gamma [ 13 ]. The proposed solution was the first parallel implementation of the PPTS method, which employed three Gamma models, where two of them exploited the resources of a parallel hardware environment (using MPI protocol and GPU).…”

Section: Introductionmentioning

confidence: 99%

Cooperative-PHD Tracking Based on Distributed Sensors for Naval Surveillance Area

Lima¹,

Costa²

2022

Sensors

View full text Add to dashboard Cite

Brazil has an extensive coastline and Exclusive Economic Zone (EEZ) area, which are of high economic and strategic importance. A Maritime Surveillance System becomes necessary to provide information and data to proper authorities, and target tracking is crucial. This paper focuses on a multitarget tracking application to a large-scale maritime surveillance system. The system is composed of a sensor network distributed over an area of interest. Due to the limited computational capabilities of nodes, the sensors send their tracking data to a central station, which is responsible for gathering and processing information obtained by the distributed components. The local Multitarget Tracking (MTT) algorithm employs a random finite set approach, which adopts a Gaussian mixture Probability Hypothesis Density (PHD) filter. The proposed data fusion scheme, utilized in the central station, consists of an additional step of prune & merge to the original GM PHD filter algorithm, which is the main contribution of this work. Through simulations, this study illustrates the performance of the proposed algorithm with a network composed of two stationary sensors providing the data. This solution yields a better tracking performance when compared to individual trackers, which is attested by the Optimal Subpattern Assignment (OSPA) metric and its location and cardinality components. The presented results illustrate the overall performance improvement attained by the proposed solution. Moreover, they also stress the need to resort to a distributed sensor network to tackle real problems related to extended targets.

show abstract