Abstract:
Data Fusion at edge computing plays an important role in IoT infrastructure and a lot
of research has already been carried out in this domain on privacy preservation of homogeneous
data fusion. However, there still remains a dire need to design a secure and
lightweight privacy preserving scheme for heterogeneous data fusion which should be
dynamic and adaptive in nature that can address the issues with authentication of the devices
connected to CSP for communication purposes and for fused data communication
while providing privacy preservation. This research focused to take data from multiple
sensors i.e. heterogeneous data and then use data fusion techniques to accurately identify
the action needed to be taken autonomously by the underlying machine. The main
objective of this study is to put forward a secure and efficient scheme to overcome these
prevailing issues. This thesis has presented a novel PPFHI scheme for heterogeneous
data fusion in IoT devices that can efficiently balance privacy and trust assessment while
requiring little overhead in terms of computation, communication, and storage to enable
distributed data fusion across the e-healthcare sector. Additionally, we have provided
in-depth theoretical research, and the findings have shown that the PPFHI scheme is
better compared to state-of-the-art schemes in many ways, including the accuracy of
fusion outcomes.
