the p-value as a new similarity function for spectral clustering in Sensor networks

Mael Bompais (1) , Hamza Ameur (1) , Dominique Pastor (2, 1) , Elsa Dupraz (3, 1)

1 IMT Atlantique - SC - Département Signal et Communications
2 Lab-STICC_IMTA_CID_TOMS
3 Lab-STICC_IMTA_CACS_COM

In this paper, we consider spectral clustering over data collected by a network of sensors. In this context, the spatial data distribution is not necessarily uniform and can further be affected by sensor noise. This is why we propose a new similarity measure for spectral clustering in sensor networks. This similarity function is derived as the p-value of an hypothesis test that has to decide whether two sensor measurements belong to the same cluster. Unlike other existing similarity measures, the p-value takes into account both the local data densities and the fact that the noise variance can vary from sensor to sensor. Simulation results show that the p-value leads to a better spectral clustering performance than the standard Gaussian kernel when there is some noise in the collected data.

Mots clés: Spectral clustering; sensor networks; Random Distortion Testing; p-value; similarity function

Domaines: Traitement du signal et de l'image [eess.SP]; Acoustique [physics.class-ph]

Mael Bompais, Hamza Ameur, Dominique Pastor, Elsa Dupraz. the p-value as a new similarity function for spectral clustering in Sensor networks. SSP 2018 : IEEE Statistical Signal Processing Workshop, Jun 2018, freiburg, Germany. ⟨10.1109/SSP.2018.8450769⟩. ⟨hal-01801577⟩