Context Aware Sensor Collaboration for Intelligent Wireless Communications an AI Approach to Moving Sensor Managemet

Abstract

Collaborative sensor data service is an emerging technology and it is beneficial to various applications including robotics, medicals, industry and military. Sensor collaborations improve technical difficultiesonthe verification and validation of sensor data or reduction of wireless sensor data transmission. However, typical approaches to sensor collaborations are less satisfactory. It is in part because the sensor calibrations are pre-fixed and therefore they are less adaptive to dynamic changes in environment. It is also because sensors are calibrated one time before deployment, and their collaborations do not take into consideration the dynamic movement of sensors. Their calibrations are not satisfactorily adaptive to environmental changes, or their collaborations are less efficient to cope with abrupt presence/absence of sensors. This paper proposes a two-tier deep learning technique to enable sensor devices to be adaptive and moving sensors to be collaborative. The contribution of this paper is an intelligent identification of environment changes and intelligent rearrangement of wireless sensor network.