Facing the application demand of large bandwidth, low loss and high speed of future communication system, optical communication is rising rapidly and developing continuously because of its excellent communication performance. As the basic optical structure of silicon-based optical chips, resonant microloops are widely used in the manufacturing of integrated photonic devices with different functions because of their advantages such as simple structure, small size, mature technology, low production cost and photoelectric hybrid integration.
This research aims to break the performance optimization bottleneck caused by manual control of cascaded MRR, and focuses on the cascaded microloop adaptive control based on optimization algorithm. Starting from the key theories and technologies, this paper mainly introduces and combs the theoretical basis of microring resonator, the algorithm principle of adaptive control and the theoretical basis of microwave photonic filter, and establishes the cascaded microring model by using the transmission matrix method, analyzes its transmission spectrum and carries out adaptive control. The main work of this study is as follows:
-
The basic theory of the microcavity model and the adaptive control model is expounded, and the model simulation of single-ring and cascaded microrings based on the MATLAB simulation platform is successfully completed, and the model transmission characteristics of the symmetric structure and the asymmetric structure of the cascaded microring are analyzed and compared by taking the four-ring cascade as an example. The download rejection ratio of the symmetrical fourth-order microring model reaches 62.2dB, and the in-band fluctuation performance of the passband is better (< 0.5 dB), half height and full width of 624 GHz, with good filtering characteristics.
-
The performance of four intelligent optimization algorithms in the adaptive control of the microcavity model is discussed. We take the suppression ratio as a single optimization objective and set the stopping criterion as a running time limit of 500 units. The model based on the genetic algorithm and the model based on the nonlinear minimization algorithm have poor performance. The performance of the model based on the agent optimization algorithm is not so good. The rejection ratio of the downloader reaches 72dB, the insertion loss is 10 dB and the passband is very unstable. The model based on pattern search algorithm has excellent performance, and the rejection ratio of the cascaded microring download side reaches 96 dB.
-
The adaptive control of the four-loop cascade based on multi-objective integration was realized, and finally, the cascaded microring simulation model with the suppression ratio of the download end reaching 47dB, insertion loss of 1.5dB, in-band fluctuation of 0.28dB, half-height full width (FWHM) narrow to 25.125GHz and high Q value was obtained. The transmission spectrum is shown in FIG. 3.15.
-
Complete the experimental verification of the link based on the implementation of the microwave photon filter. The final adaptive control-based MPF has a half-height full width (FWHM) of 1.02GHz, a rejection ratio of 44dB, and a insertion loss of 35.7dB. Due to the improper parameter setting of bait-doped fiber amplifier EDFA and electric amplifier EA, the index of insertion loss is not ideal. However, from the two indexes of half-height full width and rejection ratio, we can see that the overall performance of MPF based on adaptive control is better, which validates the feasibility and effectiveness of the adaptive control work in this study.