12 October 2015 An improved method of fuzzy support degree based on uncertainty analysis
Author Affiliations +
Most multisensor association algorithms based on fuzzy set theory forms the opinion of fuzzy proposition using a simple triangular function. It does not take the randomness of measurements into account. Otherwise, the variance of sensors supposed to be known in the triangular function, but in fact the exact variance is difficult to acquire. This paper discuss about two situations with known and unknown variance of sensors. First, with known variance and known mean. This paper proposes a method, which use the probability ratio to calculate the fuzzy support degree. The interaction between the two objects is considered. Second, with unknown variance and known mean value, we replace the sample mean in the gray auto correlation function with the real sensor mean value to analysis the uncertainty which is the correlation coefficient between targets and measurements actually. In this way, it can deal with the case of small sample. Finally, form the opinion about the fuzzy proposition in terms of weighting the opinion of all the sensors based on the result of uncertainty analysis. Sufficient simulations on some typical scenarios are performed, and the results indicate that the method presented is efficient.
© (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Yuan Huang, Yuan Huang, Jing Wu, Jing Wu, Lihua Wu, Lihua Wu, Weidong Sheng, Weidong Sheng, } "An improved method of fuzzy support degree based on uncertainty analysis", Proc. SPIE 9639, Sensors, Systems, and Next-Generation Satellites XIX, 963922 (12 October 2015); doi: 10.1117/12.2194273; https://doi.org/10.1117/12.2194273


Fuzzy logic multisensor association algorithm
Proceedings of SPIE (July 27 1997)
On a new approach to reduction of data for ANFIS...
Proceedings of SPIE (October 15 2008)
Fuzzy logic approach for data association
Proceedings of SPIE (March 11 1999)

Back to Top