Software and Datasets

Software

  • A methodology for the design and porting of Convolutional Neural Networks (CNNs) to embedded systems has been proposed in the paper “Moving Convolutional Neural Networks to Embedded Systems: the AlexNet and VGG-16 case” by C. Alippi, S. Disabato and M. Roveri published in the 17th ACM/IEEE Conference on Information Processing in Sensor Networks (ISPN 2018) and is made available to the scientific community as a Matlab Toolbox at the following link.
  • A CPM-based Change Detection Test for Big Data: A MATLAB demo of the CDT meant to operate in Big Data Scenarios  is available for the download at the following link: CDT_BigData_demo. Additional information about thresholds and ARL of the CPM-based CDT for Big Data could be found at the following web page.
  • Matlab Toolbox about the Cognitive Fault Detection and Diagnosis System described in the paper “A self-building and cluster-based cognitive fault diagnosis system for sensor networks.” by Alippi, Cesare, Manuel Roveri, and Francesco Trovò published in IEEE Transactions on Neural Networks and Learning Systems, Vol. 25, No. 6, pp. 1021-1032, June 2014.

Datasets

  • POLIMI Rock Collapse Forecasting: Sensor measurements coming from a wireless-wired monitoring system for rock collapse forecasting deployed on the Italian Alps. The dataset refers to 3 temperature sensors. The sampling period is 600 seconds and the length of the dataset is 12000 samples. Download Matlab file (166 KB).

    ForecastTo use this dataset please cite the following paper: C. AIippi, R. Camplani, C. Galperti, A. Marullo, M. Roveri, “A high frequency sampling monitoring system for environmental and structural applications”, ACM Transactions on Sensor Networks, Vol. 9, No. 4, Art. 41, 32 pages, July 2013.
  • POLIMI Landslide Forecasting: Sensor measurements coming from a wireless monitoring system for landslide forecasting deployed on the Italian Alps. The dataset refers to 14 sensors (encompassing temperature, clinometer and crackmeter sensors). The sampling period is 1h and the length of the dataset is 2844 samples. Download Matlab file (272 KB).
    LandslideTo use this dataset please cite the following paper: C. AIippi, R. Camplani, C. Galperti, A. Marullo, M. Roveri, “A high frequency sampling monitoring system for environmental and structural applications”, ACM Transactions on Sensor Networks, Vol. 9, No. 4, Art. 41, 32 pages, July 2013.