Parallel Frequency Function-Deep Neural Network for Efficient Approximation of Complex Broadband Signals.

In recent years, with the growing popularity of complex signal approximation via deep neural networks, people have begun to pay close attention to the spectral bias of neural networks-a problem that occurs when a neural network is used to fit broadband signals. An important direction taken to overcome this problem is the use of frequency selection-based fitting techniques, of which the representative work is called the PhaseDNN method, whose core idea is the use of bandpass filters to extract frequency bands with high energy concentration and fit them by different neural networks. Despite the method's high accuracy, we found in a large number of experiments that the method is less efficient for fitting broadband signals with smooth spectrums. In order to substantially improve its efficiency, a novel candidate-the parallel frequency function-deep neural network (PFF-DNN)-is proposed by utilizing frequency domain analysis of broadband signals and the spectral bias nature of neural networks. A substantial improvement in efficiency was observed in the extensive numerical experiments. Thus, the PFF-DNN method is expected to become an alternative solution for broadband signal fitting.

Electromagnetic Spectrum Allocation Method for Multi-Service Irregular Frequency-Using Devices in the Space-Air-Ground Integrated Network.

The management and allocation of electromagnetic spectrum resources is the inner driving force of the construction of the space-air-ground integrated network. Existing spectrum allocation methods are difficult to adapt to the scenario where the working bandwidth of multi-service frequency-using devices is irregular and the working priorities are different. In this paper, an orthogonal genetic algorithm based on the idea of mixed niches is proposed to transform the problem of frequency allocation into the optimization problem of minimizing the electromagnetic interference between frequency-using devices in the integrated network. At the same time, a system model is constructed that takes the minimum interference effect of low-priority-to-high-priority devices as the objective function and takes the protection frequency and natural frequency as the constraint conditions. In this paper, we not only introduce the thought of niches to improve the diversity of the population but also use an orthogonal uniform crossover operator to improve the search efficiency. At the same time, we use a standard genetic algorithm and a micro genetic algorithm to optimize the model. The global searchability and local search precision of the proposed algorithm are all improved. Simulation results show that compared with the existing methods, the proposed algorithm has the advantages of fast convergence, strong stability and good optimization effect.

Compact multi-functional frequency-selective absorber based on customizable impedance films.

Multi-functional metamaterial absorbers have attracted considerable attention for applications in the microwave frequency regime. In this paper, we report the design, fabrication, and characterization of frequency-selective absorbers, which exhibit substantial absorption property within a pre-defined frequency band, while at the same time behaving as a highly transparent screen in another targeted frequency band. The proposed designs consist of a symmetrically patterned indium tin oxide film acting as an absorbing layer, two dielectric substrates, and a cross-slot metal sheet frequency selective surface playing the role of a transmitting layer. In order to validate the functionalities of the designed absorbers, equivalent circuit models, full-wave numerical simulations and measurements are presented. The measured results, in good agreement with the numerical ones, show that the proposed designs realize 80% broadband absorption over the desired frequency range and possess a transparent window in a higher or lower frequency band for a wide range of incidence angles up to 60°. These performances suggest that the proposed designs are promising candidates for multi-functional scattering control and communication applications.