Opto-Electronic Sensor Network Powered Over Fiber for Harsh Industrial Applications

Abstract

Smart electronic devices powered and communicating optically over fiber provide several advantages for harsh industrial applications compared to systems using electrical connections. In spite of advantages such as electromagnetic immunity and high galvanic isolation, power-over-fiber technologies are still only used for special high-end applications due to their high costs compared to conventional power distribution. The main reason for this is the use of high-power laser diodes and photovoltaic cells for most power-over-fiber systems. Therefore, a new cost-effective power-over-fiber approach has been developed using only single standard light-emitting diodes (LEDs) and photodiodes for optical power supply significantly reducing the system's costs and making it suitable for mass-market applications such as the Internet of Things. In addition, the low optical power density inside the fibers makes the network intrinsically safe even in highly explosive environment. The use of incoherent and visible LED light also eliminates the risk of thermal damage to the human retina to occupational safety standards if a fiber is broken or during hot-plugging. A demonstration system has been built which consists of a base station optically powering up to four electronic sensor nodes. Between the base station and each sensor node, data can be exchanged via a bidirectional 2 MBaud optical link. This paper briefly explains the history and state of the art of power-over-fiber systems. The technical challenges of development as well as the steps made to build optically powered low-cost and low-power electronic sensor nodes which overcome the limitations and drawbacks of existing high power solutions are described. Optical, electrical, and computational aspects of the new system are elaborated on. In addition, energy measurements and power budget calculations are described to analyze and optimize the sensor nodes’ energy consumption during operation. Finally, the advantages of this new approach are exemplified for smart manufacturing applications.