Convergence Analysis of Dual Quantized Iterative Learning Control with Channel Fading

In the field of stochastic Iteration Learning Control (ILC) people tend to treat channel fading and data quantization as two separate random disturbances. In this paper, we give an analysis of the convergence of quantized ILC in the presence of channel fading. In addition, we design a dual quantized ILC, which can further reduce the deviation before and after channel fading while ensuring convergence. In the dual quantized ILC, two separate quantizers are set before and after the channel fading. The first quantizer is designed to relieve the transmission pressure; the second quantizer is designed to reduce the error caused by the channel fading. Here we use the multiplicative fading factor to describe the channel fading. We adjust the quantization interval of the quantizer to ensure that the dual quantized ILC has a greater probability of smaller errors than the quantized ILC. Illustrative simulations are provided to verify the theoretical results.