6 results about "Orthonormal basis" patented technology

This invention discloses a method for recognizing orbital angular momentum superposition states based on conditional generative adversarial networks (GANs). The improved GAN effectively recovers and identifies severely distorted vortex beams. A skip connection mechanism is introduced into the generator's network structure to preserve low-level detail features and improve image generation quality. Furthermore, a peak signal-to-noise ratio (PSNR) loss function is introduced into the generator's loss function, forcing the generated image to maintain a higher pixel-level similarity to the real image. Convolutional attention modules are introduced into both the generator and discriminator based on the improved GAN, enhancing the network's ability to extract key information. This method effectively recovers severely distorted vortex beams affected by strong turbulence and long-distance propagation conditions, ultimately achieving high recognition accuracy.

Methods, systems and apparatus for simulating a physical system described by an electronic structure Hamiltonian expressed in an orthonormal basis. In one aspect, a method includes decomposing the electronic structure Hamiltonian into a sum of sub-Hamiltonians, wherein each sub-Hamiltonian in the sum of sub-Hamiltonians is expressed in one of multiple bases; simulating evolution of the physical system using the decomposed electronic structure Hamiltonian; and using the simulated evolution of the physical system using the decomposed electronic structure Hamiltonian to determine properties of the physical system.

A method and network node for interference-aware dimension reduction via orthonormal basis selection are disclosed. According to one aspect, a method includes determining a beamformer for each wireless device (WD), each beamformer being based at least in part on an interference covariance matrix; determining a power in each dimension reduction basis vector based at least in part on the determined beamformers; and selecting a subset of the dimension reduction basis vectors based on the determined power. A number of selected dimension reduction basis vectors is less than a number of antenna elements of the network node. The method also includes multiplying a received signal vector by a dimension reduction matrix having as rows the selected subset of the dimension reduction basis vectors. The multiplying results in a reduced signal vector having a dimension equal to the number of selected rows of the dimension reduction matrix.

A method for determining at least one characteristic of an electromagnetic signal using a quantum probe, the method including determining a resonant frequency of at least one qubit of at least one quantum probe, applying at least one pulse sequence to the at least one qubit, wherein the at least pulse sequence has a carrier frequency of the resonant frequency and is associated with a filter function that comprises an orthonormal basis, capturing a plurality of images associated with electromagnetic radiation emitted by the at least one qubit after applying the at least one pulse sequence to the at least one qubit, and determining at least one characteristic of the electromagnetic signal based on the plurality of images.

The invention belongs to the technical field of optical communication, and particularly relates to an orbital angular momentum pattern recognition method based on deep learning. According to the OAM pattern recognition method, the efficiency and precision of model training are remarkably improved by introducing the adaptive learning rate scheduling and gradient cutting technology. In the experiment, the Two Phase Scheduler and Dynamic Range SGD optimizers provided by the invention can ensure the steady convergence of the model in different initial learning rate ranges, so that the training process is obviously accelerated, and the common convergence oscillation problem is effectively avoided. According to the method, a high-precision recognition result can be provided in an OAM speckle classification task, good generalization ability is shown in the face of various different scattering environments, and an efficient and robust solution is provided for the OAM mode recognition problem.

The invention discloses an orbital angular momentum superposition state identification method based on a conditional generative adversarial network. Vortex beams which are seriously distorted are effectively recovered and identified through the improved conditional generative adversarial network. A jump connection mechanism is introduced into a network structure of the generator, so that bottom detail features can be reserved, and the generation quality of the image is improved. In addition, a peak signal-to-noise ratio loss function is introduced into the loss function of the generator, and the generated image and the real image are forced to keep higher similarity at the pixel level. On the basis of the improved conditional generative adversarial network, convolution attention modules are introduced into a generator and a discriminator, so that the key information extraction capability of the network is improved. According to the method provided by the invention, the vortex beam which is seriously distorted due to the influence of strong turbulence and long-distance propagation conditions is effectively recovered, and finally relatively high identification accuracy is obtained.