The Spatiotemporal Data Fusion (STDF) Approach: IoT-Based Data Fusion Using Big Data Analytics.

Area of Science:

• Computer Science
• Data Science
• Internet of Things

Background:

• Internet of Things (IoT) generates vast, heterogeneous sensory big data with unique features like trustworthiness, timing, and spatial characteristics.
• Traditional data fusion methods struggle with the complexity and scale of IoT data, posing challenges for collection and analysis.
• Existing approaches often overlook critical IoT data perspectives such as data expiry, trustworthiness, and spatiotemporal dynamics.

Purpose of the Study:

• To propose an IoT-based spatiotemporal data fusion (STDF) approach for low-level, in-data out fusion.
• To enable real-time spatial IoT source aggregation by addressing the unique challenges of IoT big data.
• To enhance data fusion performance by integrating traditional methods with IoT-specific considerations.

Main Methods:

• Leveraged big data analytics for traditional data fusion methods.
• Incorporated data expiry, trustworthiness, and spatial/temporal perspectives into the fusion process.
• Applied cluster sampling for data reduction at acquisition.
• Utilized k-means clustering for spatial analysis and Tiny AGgregation (TAG) for temporal aggregation at the processing server.

Main Results:

• Achieved significant data size reduction of 95%.
• Decreased processing time by 80%.
• Maintained high accuracy, reaching up to 90% on the largest dataset.

Conclusions:

• The proposed STDF approach effectively handles the complexities of IoT big data.
• STDF offers an efficient solution for real-time spatial IoT source aggregation.
• The method demonstrates optimal performance in data reduction, processing speed, and accuracy for IoT applications.