Fiber array optical systems that are used for free-space optical communications and phased laser beam projection applications depend on fast, closed-loop adaptive control to efficiently compensate for optical distortion caused by atmospheric turbulence. Current off-the-shelf systems are limited in performance, fiber array control channels, and flexibility. In this research project, we built a scalable and versatile platform for closed-loop controlling fiber-arrays with high update rate and rapid algorithm development. The controller platform consists of two parts: (I) an analysis and simulation software framework that is used for algorithm development and finding optimal parameters. Code written and simulated can then be directly ported to (II), a real-time hardware engine that can execute the algorithm in a real system and has the capability of controlling a large number of fiber arrays for testing and deployment in field conditions. The analysis and simulation software framework is used to simulate and predict how well the hardware will perform the implemented algorithm with a certain set of parameters and visualize the results of software or hardware runs. The optimized algorithms can be easily transferred back and forth to the hardware engine to run in real-time. The hardware platform is capable of standalone operation and is based on a System on chip (SoC) which has an ARM Central Processor Unit (CPU) and a floating-point Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), custom hardware modules that has large array of digital to analog convertors (DACs) and amplifiers for controlling fiber arrays, and another custom hardware module that has amplified ADC channels for reading back the quality metric value. Test results from the prototype system with 19 DAC channels and two analog to digital convertor (ADC) channels are presented and compared with simulated results.
展开▼
