Deep‐Learning‐Based Pilot Optimization for Near‐Field Channel Estimation in Ultra‐Massive MIMO

ABSTRACT As sixth‐generation (6G) communication technology evolves, the increase in frequency and the number of antennas has made traditional far‐field channel estimation methods less effective. This paper proposes a deep neural network (DNN)‐based method to optimize pilot signals for near‐field channel estimation in ultra‐massive multiple‐input multiple‐output (MIMO) systems. By optimizing the pilot signals, the method can accurately estimate the distance and angle of scatterers, addressing the challenges of traditional sparse estimation techniques. Simulation results demonstrate that this approach improves estimation accuracy compared to conventional methods.