A node and network level self-healing distributed wireless sensor architecture for greenhouse based plant monitoring systems

Visual Sensor Networks (VSN) has emerged as a new paradigm by giving a sensor the capability to perceive and analyze its surrounding. Creating an environment of "no doubt" is essential for a contact free plant monitoring system in a greenhouse environment equipped with VSN. Methods to monitor a system's health, and respond to hardware anomalies automatically are essential. With reconfigurable devices, such as Field Programmable Gate Arrays (FPGAs), designers are provided with a seductive tool to use as a basis for adopting the hardware's configuration to the needs of the application. In this study, we designed a distributed wireless architecture with self-configuration capability at network level and fault recovery capability at node level in the context of image processing algorithms (Discrete Cosine, Discrete Wavelet Transforms, and Color Space Conversion). The experimental results showed that the test bed comprised of wirelessly connected Xilinx Virtex5 FPGAs is able to achieve node level fault detection and healing within 2.18 seconds and network level healing in 5.43 seconds. The application of the proposed architecture in greenhouse based plant production is significant towards building a robust system for monitoring and sensing the climate conditions and plant status. Developing adaptive plant-response based climate control strategies in controlled environment plant production will also considerably contribute to resource savings (i.e. water, nutrients, and energy).