Kun Xu

TL;DR: In this paper, a signal modulation format identification method and device based on a photonic neural network, electronic equipment and a storage medium is presented, which comprisesthe steps of: acquiring a characteristic signal of a to-be-identified signal, wherein the characteristic signal is obtained after the to be-identified signals are sampled; inputting the characteristicsignal into a photonics chip, where the photonic chip is used for completing a linear operation in a neural network algorithm for signal modality format identification; and acquiring an output result after the photon carries out the linear operation

TL;DR: An adjustment-free linearization approach where high dynamic range could be kept during RF frequency tuning is proposed and demonstrated and the spurious-free dynamic range (SFDR) is kept within 1-Hz bandwidth without attendant optimization of link parameters.

TL;DR: This work utilizes optimal transport based pixel matching to design a more sophisticated loss function for inverse rendering applications and has demonstrated superior convergence behavior compared to state-of-the-art baselines.

TL;DR: In this paper, the authors proposed a multi-frequency wireless location method, which includes that a mobile terminal transmits at least two kinds of wireless signals different in frequency; a receiver receives the wireless signals and transmits to a signal processor, and finally, the signal processor processes the wireless signal and obtain the distance between the mobile terminal and the receiver.