Fog Computing Architecture-Based Data Reduction Scheme for WSN

Wireless Sensor Networks (WSNs) are widely used in important areas such as smart cities, security detection, and environmental monitoring. However, as the volume of sensor data increases exponentially, intelligent data reduction is required on edge devices. Data reduction technology can reduce network congestion and the transmission power consumption of sensor nodes, and extend the life of WSNs. Inspired by fog computing, this paper proposes a data compression scheme based on fog computing architecture, which uses the synchronous prediction model to achieve data reduction. The fog node constructs a synchronous prediction model by using an auto-regressive analysis method and fits the characteristics of the sensor data so that the data stream can be reduced from the source node. Considering the complexity of the WSN deployment environment, an environment adaptive variable threshold optimization strategy is proposed to improve the availability and performance of the model. The scheme is then evaluated by simulation with the real world data. The simulation results demonstrate that the scheme can effectively reduce the data traffic of WSNs and the energy consumption of sensor nodes.