Distributed Multi-Sensor Fusion of PHD Filters With Different Sensor Fields of View

Yi, Wei; Li, Guchong; Battistelli, Giorgio

doi:10.1109/tsp.2020.3021834

Cited by 63 publications

(33 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The most known mixture is the Gaussian mixture [13,7], which consists of a finite number of Gaussian distributions. Recently, it has been further shown that the arithmetic average (AA) fusion which has provided a compelling approach to multi-target density fusion/consensus over sensor networks [14,15,16,17,18,19,20,21] will also result in a mixture distribution.…”

mentioning

confidence: 99%

Why Mixture?

Li¹

2021

Preprint

View full text Add to dashboard Cite

From the most known Gaussian mixture to the cutting-edge multi-Bernoulli mixture of various forms, mixture offers a fundamental means to deal with uncertainties, which has led to a variety of appealing applications in the state estimation realm based on a single sensor or a sensor network. Like noise is often used to model unknown system input, one may use various hypotheses to deal with the uncertain state space model or data association. Meanwhile, consensus may be sought over the cross-correlated sensors. These all drive a need for representing the probability distribution by a mixture of properly weighted component distributions, which fuse the information gained from different models/hypotheses or from different sensors. This technical note presents information-theoretical results which answer how the averaging/mixture approach makes sense and how the fusing weights should be designed.

show abstract

mentioning

confidence: 99%

Why Mixture?

Li¹

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…-Example 2: In practical cases, sensors usually have limited FoVs, e.g., camera often have fixed square frames, and radars commonly have fan-shaped FoVs due to the radiation limitations. In [34], theoretical analysis was given to show that direct homogeneous fusion among sensors with different FoVs is unsuitable and can result severe performance degradation. Recently a lot of effort has been devoted to devise the countermeasures so as to deal with the different FoV issue [35]- [37], [40].…”

Section: Motivation Analysismentioning

confidence: 99%

“…In this case, the information confidence of an MOD cannot be sufficiently characterized using only a scalar coefficient, and severe fusion performance degradation can be observed due to the weighting bias [33]. This problem is especially evident in the fusion among sensors having different fields of view (FoVs), since the relative information confidence is different in different regions due to the non-overlapping FoVs [34]. Recently, the different FoV issue has been intensively studied, and effective countermeasures have been proposed for both unlabel MODs [34]- [38] and labelled RFS MODs [39]- [41].…”

Section: Introductionmentioning

confidence: 99%

“…This problem is especially evident in the fusion among sensors having different fields of view (FoVs), since the relative information confidence is different in different regions due to the non-overlapping FoVs [34]. Recently, the different FoV issue has been intensively studied, and effective countermeasures have been proposed for both unlabel MODs [34]- [38] and labelled RFS MODs [39]- [41]. However, these methods do not change the fundamental problem of the fusion mechanism.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Heterogeneous Multi-Sensor Fusion With Random Finite Set Multi-Object Densities

Chai

2021

IEEE Trans. Signal Process.

Self Cite

View full text Add to dashboard Cite

“…To improve the tracking performance and reduce the computational complexity, several gating strategies were introduced to PHD filter [44,45]. To fuse measurements from multiple receivers, several multi-sensor PHD filters were proposed [46,47]. Confronted with unknown background parameters, such as unknown target birth intensity, unknown detection probability, several refined PHD filters can estimate the multi-target state and background parameters simultaneously [48,49].…”

Section: Introductionmentioning

confidence: 99%

Improved probability hypothesis density filter for multi‐target tracking of non‐cooperative bistatic radar

Wang

Bao

Pan

2021

IET Radar Sonar & Navi

View full text Add to dashboard Cite

Non-cooperative bistatic radar refers to the passive bistatic radar using a non-cooperative radar as the illuminator of opportunity. Limited by the non-cooperation and bistatic configuration, multi-target tracking of the non-cooperative bistatic radar is confronted with challenges of low detection probability and high clutter. Based on the sequential Monte Carlo implementation of probability hypothesis density (PHD) filter, this study proposes an improved PHD (I-PHD) filter. Similar to the tracking gate in data association, whether each candidate target is detected at each time is determined and recorded, and the sequential probability ratio test distinguishes a real target and false target based on this record. To distinguish miss detection and target death, a new survival probability dependent on target position is defined, and the survival probability of a specific target drops rapidly when it is located near the boundary and moves outwards. To label targets and extract their states, a new method of track identification and state estimation is proposed. The simulation results illustrate that the proposed I-PHD filter can effectively track multiple targets of non-cooperative bistatic radar under low detection probability and high clutter and has superior performance than competing methods.

show abstract

Distributed Multi-Sensor Fusion of PHD Filters With Different Sensor Fields of View

Cited by 63 publications

References 47 publications

Why Mixture?

Why Mixture?

Heterogeneous Multi-Sensor Fusion With Random Finite Set Multi-Object Densities

Improved probability hypothesis density filter for multi‐target tracking of non‐cooperative bistatic radar

Contact Info

Product

Resources

About