An Attention Mechanism Inspired Selective Sensing Framework for Physical-Cyber Mapping in Internet of Things

The increasing growth of big data is certainly challenging ubiquitous sensing in the Internet of Things (IoT) paradigm because of the limitations in sensing resources. Processing huge amounts of sensed data requires an enormous and unnecessary pool of resources. Both reasons strongly support the idea of adopting a selective sensing solution to handle the mapping between physical space and cyberspace and to lighten the load of data processing in IoT applications. Inspired by the ability of creatures that fleetly select the information of interest from a noisy environment and process them with limited attention resources, in this paper the biological attention mechanism is introduced to design a novel selective sensing framework called attention mechanism inspired selective sensing (AMiSS). In order to illustrate the functionality of the AMiSS platform, a use case scenario in reference to the security system of a modern transport station is presented. Further, we implement a proof-of-concept simulation using video-based object tracking to verify the feasibility and effectiveness of the AMiSS framework in IoT applications. Although it is just a narrow demonstration, the simulation still shows the effect of the AMiSS platform in reducing the amount of data processed by the higher layers.