Open Access Open Access  Restricted Access Subscription Access
Cover Image

Sensor Networks with Fault Detection and Self Healing Using Soc based Fpga Architecture

Tapas Bapu B. R, Siddanna Gowd L. C.

Abstract


Wireless sensor network plays a very important role of successful applications in the engineering area. The tiny but excessive operative devices are featuring the industries with efficient online monitoring and control system applications. Recently FPGA based reconfigurable system for a wireless sensor network is proposed. When a node has a fault, it is commonly discarded and network is reorganized to ensure the normal operation of WSN. Sometimes the maintenance cost and function of the network will be increased .The conventional work is developed with a lightweight compression and a fault prediction technique. On the trigger for a fault occurrence of the system, the overall system is capable of switching to the back-up module. This was accomplished with the ease integration and flexibility feature offered by modern days FPGAs. On the current scenario SoC based FPGAs are fabricated to give wider features of industries. With this feature we can not only develop a self-healing WSN, but also a hardware accelerated power efficient WSN. In this work, we develop a novel hardware accelerated FPGA approach for the self-healing network of modular even-driven approach to address the power management issues with SoC based FPGA. Conventional event-driven approach were tested and implemented only on standalone FPGA without processor core inside. As our self-healing system demands a SoC based FPGA, and this work optimizes to deliver a power efficient network on a single system. The event-driven approach, compression and fault prediction are fully developed in Hardware platform, where as the self-healing controller logic is managed at processor level for faster performance. In evaluation, this design is proven to be power efficient self-healing system with faster performance through the processor based controller design.


Full Text:

PDF

Refbacks

  • There are currently no refbacks.