26 results about "Photonic network" patented technology

Various embodiments provide for electro-photonic networks, including a plurality of processing elements connected by bidirectional photonic channels, suited for implementing neural-network models. Weights of the model may be preloaded into memory of the processing elements based on assignments of neural nodes to processing elements implementing them, and routers of the processing elements can be configured to stream activations between the processing elements based on a predetermined flow of activations in the model.

The invention discloses a remote proving method for a trusted data space, which relates to the technical field of computer and data security, and comprises the following steps: initializing a trusted environment and generating a dynamic identity key for a heterogeneous terminal containing a CPU (Central Processing Unit), a GPU (Graphic Processing Unit) and an FPGA (Field Programmable Gate Array) based on a hardware security module and a physical unclonable function, and generating a challenge value by a verifier by using a quantum random number; the method comprises the following steps of: firstly, transmitting to a proving party through quantum encryption and a neuromorphic photonic network, constructing proving by the proving party by adopting a tensor decomposition zero-knowledge proving protocol, finishing multi-stage verification on a verification party in combination with a space-time cause and effect graph and quantum signature aggregation, and finally realizing dynamic trust evaluation and adaptive strategy adjustment through a quantum Bayesian network, reinforcement learning and biological feedback. According to the method, the remote attestation performance of the trusted data space is improved, dynamic trust evaluation is realized by means of the quantum Bayesian network and biological feedback, attacks are effectively resisted, real-time requirements are met, trust is accurately evaluated, and the development of the trusted data space is promoted.

Electro-photonic networks, including a plurality of processing elements connected by bidirectional photonic channels, suited for implementing neural-network models. Weights of the model may be preloaded into memory of the processing elements based on assignments of neural nodes to processing elements implementing them, and routers of the processing elements can be configured to stream activations between the processing elements based on a predetermined flow of activations in the model.

A photonic parallel network can be used to sample combinatorially hard distributions of Ising problems. The photonic parallel network, also called a photonic processor, finds the ground state of a general Ising problem and can probe critical behaviors of universality classes and their critical exponents. In addition to the attractive features of photonic networks—passivity, parallelization, high-speed and low-power—the photonic processor exploits dynamic noise that occurs during the detection process to find ground states more efficiently.

Methods and systems for routing single photons to selected destinations apply a single photon source optically coupled to a frequency demultiplexer of a photonic network. The frequency demultiplexer comprises plural outputs, each of the outputs is associated with a respective frequency band. A photon from the single photon source is directed to a selected one of the outputs by configuring the single photon source to emit photons having frequencies lying within the one of the frequency bands corresponding to the selected output and controlling the single photon source to emit the photon. The methods and systems may be applied to entangle a quantum state of a quantum system of the single photon source with another quantum system.

Multi-chip electro-photonic networks for interconnecting multiple circuit packages via photonic channels. Examples of a computing system include a first circuit package, a second circuit package, and one or more inter-chip bidirectional photonic channels interconnecting the first and second circuit packages. The first circuit package can include one or more memory nodes. The second circuit package can include multiple compute nodes and intra-chip bidirectional photonic channels interconnecting the compute nodes.

Multi-chip electro-photonic networks for interconnecting multiple circuit packages via photonic channels. Examples of a computing system include a first circuit package, a second circuit package, and one or more inter-chip bidirectional photonic channels interconnecting the first and second circuit packages. The first circuit package can include one or more memory nodes. The second circuit package can include multiple compute nodes and intra-chip bidirectional photonic channels interconnecting the compute nodes.

Multi-chip electro-photonic networks for interconnecting multiple circuit packages via photonic channels. Examples of a computing system include a first circuit package, a second circuit package, and one or more inter-chip bidirectional photonic channels interconnecting the first and second circuit packages. The first circuit package can include one or more memory nodes. The second circuit package can include multiple compute nodes and intra-chip bidirectional photonic channels interconnecting the compute nodes.

Described herein are embodiments of a photonic computing system comprising one or more processors in communication with disaggregated memory through one or more optical channels. The disaggregated memory comprises multiple memory units placed on a photonic substrate that includes a photonic network that can be programmed to configure which of the memory units can be accessed by each of the processor(s). The disaggregated memory includes a memory controller for reading and writing data to / from the memory units. The memory controller may be configured to perform processing in concert with the processor(s).

Multi-chip electro-photonic networks for interconnecting multiple circuit packages via photonic channels. Examples of a computing system include a first circuit package, a second circuit package, and one or more inter-chip bidirectional photonic channels interconnecting the first and second circuit packages. The first circuit package can include one or more memory nodes. The second circuit package can include multiple compute nodes and intra-chip bidirectional photonic channels interconnecting the compute nodes.

The application belongs to the technical field of photonic link, and particularly relates to a photonic link fast frequency domain calculation method based on multimode coupling and sub-network growth, which comprises the following steps: S1, constructing a multimode scattering model of an optical device; S2, constructing a sub-network multimode scattering matrix; and S3, sub-network merging and optimization. The method is suitable for a photonic network containing any interconnected multi-port elements, can accurately characterize multimode propagation and coupling phenomena by using scattering parameters, can support any number of modes in a waveguide element and mode coupling effects, and can realize efficient calculation of photonic link frequency domain response through step-by-step solution of the multimode scattering matrix.

The present invention discloses a remote attestation method for a trusted data space, which relates to the field of computer and data security technology. First, a trusted environment is initialized for heterogeneous terminals containing CPUs, GPUs, and FPGAs based on hardware security modules and physically unclonable functions to generate dynamic identity keys. The verifier generates a challenge value using quantum random numbers, which is transmitted to the prover via quantum encryption and neuromorphic photonic networks. The prover constructs a proof using a tensor decomposition zero-knowledge proof protocol, combines a spatiotemporal causal graph with quantum signature aggregation to complete multi-level verification on the verifier, and finally implements dynamic trust evaluation and adaptive strategy adjustment through quantum Bayesian networks, reinforcement learning, and biofeedback. The present invention improves the remote attestation performance of trusted data spaces, implements dynamic trust evaluation with the help of quantum Bayesian networks and biofeedback, effectively resists attacks, meets real-time needs, accurately evaluates trust, and promotes the development of trusted data spaces.

A photonic system includes a photonic mesh comprising a first array of N input light guiding units located in a first plane and arranged in a spaced-apart substantially parallel relationship along a first axis, and a second array of M output light guiding units located in a second plane and arranged in a spaced-apart substantially parallel relationship along a second axis intersecting with the first axis. The optical coupler is configured for coupling a portion of an input light signal from the input light guiding unit into the output light guiding unit at the respective node, and is configured as a tunable modulation structure controllably switchable between various operative states by a tuning field, configured as a weighting signal to effect a corresponding intensity modulation of the portion of the light signal being coupled from the input light guiding unit into the output light guiding unit by said optical coupler.

The invention discloses an intelligent conference exhibition resource scheduling method and device based on space-time sequence analysis and a medium, and relates to the technical field of intelligent scheduling. The method comprises the steps that historical scheduling information is acquired firstly, and space-time feature vectors are extracted through high-dimensional space-time chaotic coding; selecting candidate scheduling points in combination with the spatial-temporal characteristics of the target event to generate an initial scheme; calculating a network evaluation scheme and calculating the transportation space-time cost through the photon neural morphology; establishing a space-time loss scheduling model according to the demand address, the candidate point and the transportation cost; and finally, a target model is obtained through a quantum-biological hybrid optimization algorithm, scheduling is completed, spatial-temporal feature extraction comprises chaotic characteristic analysis, candidate point selection gives consideration to similarity and resource quantity, an initial scheme introduces quantum coding, transportation cost is combined with a photon network and quantum calculation, and a loss model is constructed according to resource satisfaction and shortage situations. Quantum annealing and biological evolution are optimized and fused, and efficient and intelligent resource scheduling is achieved.

Systems and methods are provided for optically implemented Kolmogorov-Arnold Networks (KAN). Examples provide a photonic Kolmogorov-Arnold Network that includes a plurality of neurons and a plurality of synaptic edges. Each synaptic edge comprises a waveguide that optically couples a neuron of the plurality to another neuron of the plurality of neurons, and a nonlinear optical modulator formed on the waveguide, wherein the nonlinear optical modulator is configured to be tuned to a desired nonlinear activation function.

The device (2) comprises a medium (7) capable of supporting overlapping optical modes (10) such that, in response to spatially-structured illumination (4), a set of overlapping optical modes (10) in t

The invention discloses a photon network-on-chip routing method based on a PPO-GRU algorithm, and the method comprises the steps: firstly constructing a PNoC simulation array, and building a PNoC routing model based on deep reinforcement learning in combination with optical signal transmission characteristics; according to the model core, on the basis of a PPO algorithm, an invalid action shielding mechanism and a GRU module are fused, and the path searching efficiency is effectively improved. And meanwhile, by adopting dominant function normalization and reward scaling technologies, the stability and convergence reliability of model training are remarkably enhanced. By using the trained routing model, the low-loss optical routing path can be intelligently selected, so that the network performance, the reliability and the resource utilization rate of the PNoC are remarkably improved.

An estimation apparatus that estimates a Raman tilt of an optical transmission line in an all-photonics network, the estimation apparatus including an acquisition unit for acquiring a use state of each use wavelength in the optical transmission line and optical power of each use wavelength at an outlet of an adjacent node of the optical transmission line, a determination unit for determining the number of divisions, and a Raman tilt calculation unit for estimating the Raman tilt by using an integral expression in which each of subbands obtained by dividing an entire use band into the number of divisions is an integration step, the subbands having, as parameters, a use state of each use wavelength, optical power of each use wavelength, a loss coefficient determined by a fiber type of the optical transmission line, a link distance defining a distance between adjacent nodes, and a Raman coefficient.

The invention belongs to the technical field of photonic links, and particularly relates to a photonic link fast frequency domain calculation method based on multimode coupling and sub-network growth, which comprises the following steps: S1, constructing an optical device multimode scattering model; s2, constructing a sub-network multimode scattering matrix; s3, merging and optimizing the sub-networks; the method has the advantages that the method is suitable for the photonic network containing any interconnected multi-port element, scattering parameters are used for accurately representing multimode propagation and coupling phenomena, any number of modes in waveguide elements and the coupling effect between the modes are supported, and efficient calculation of the photonic link frequency domain response is achieved through step-by-step solution of a multimode scattering matrix.

Described herein are embodiments of a photonic computing system that includes one or more processors in communication with a split storage system over one or more optical channels. The split memory system includes a plurality of memory cells arranged on a photonic substrate, the photonic substrate including a photonic network that can be programmed to configure which of the memory cells is accessible by each of the processor (s).

A communication device communicates with another communication device via an all-photonics network (APN). The communication device includes circuitry that obtains characteristic information for each of a plurality of paths in the APN between the communication device and the another communication device. The circuitry also selects one path of the plurality of paths based on the characteristic information and based on data to be transmitted to the another communication device. Then the circuitry transmits the data via the one path selected.

An optoelectronic sensor includes a plurality of light sources for generating transmission light, including at least a first and a second light source. The optoelectronic sensor includes transmission optics for projecting transmission light into the field of view and at least one detector for detecting transmission light reflected by the object. A photonic network has a plurality of irradiation points to each of which transmission light, in particular in each case of exactly one light source, can be supplied. The transmission light exits into the transmission optics and ultimately exits into the field of view. A plurality of irradiation points that are arranged directly next to one another or that are directly adjacent are in this respect configured to irradiate transmission light into partial fields of view, in particular different partial fields of view and / or a plurality of partial fields of view arranged next to one another or are adjacent.

Multi-chip electro-photonic networks for interconnecting multiple circuit packages via photonic channels. Examples of a computing system include a first circuit package, a second circuit package, and one or more inter-chip bidirectional photonic channels interconnecting the first and second circuit packages. The first circuit package can include one or more memory nodes. The second circuit package can include multiple compute nodes and intra-chip bidirectional photonic channels interconnecting the compute nodes.

A bridge includes a plurality of interfaces including four interfaces and a photonic network. Each interface is an electrical interface configured to electrically connect a respective die to the bridge, each interface including a serializer and a deserializer for connecting to any die of the bridge through the interface. The photonic network comprises a directional photonic link from a serializer of each of the four interfaces to a deserializer of each of the remaining three interfaces, whereby the photonic network and the interfaces provide six bidirectional photonic communication paths terminating at the four interfaces, and each of the four interfaces is bidirectionally coupled to each of the other interfaces of the four interfaces.

The invention provides a multi-dimensional photonic network and a scheduling method and device thereof, electronic equipment and a storage medium, and a photonic switching matrix in the multi-dimensional photonic network comprises a photonic switch, a photonic router and a photonic link, and is used for switching, routing and relaying optical signals; the photonic link is also used for connecting different data centers and carrying out optical signal transmission among the different data centers; the AI calculation unit is used for determining target optical path information of the target optical signal according to the source end and the destination end of the target optical signal, the current topological structure of the multi-dimensional photonic network and the current resource state information; and the AI calculation unit is also connected with the photon routers and is used for controlling the photon routers in the multi-dimensional photon network according to the target optical path information and establishing a target optical path corresponding to the target optical path information. According to the invention, the technical problem that a network and a storage system are difficult to realize high-speed transmission and real-time scheduling of super-large-scale data in the prior art is solved.

A system and a method for controlling wavelength switching of open all-photonic network are provided and applied to an open all-photonic network and have wavelength switching protection and recovery functions to ensure that the network can quickly recover and provide reliable services when a failure occurs. During actual operation, information and data of the switching status can be collected in real time by a dynamic route monitoring module, and the configuration of an open all-photonic network transceiver, i.e., wavelength path switching, can be updated by a dynamic configuration module, so that efficient monitoring and management of quality assurance can be provided. A computer-readable medium for executing the method for controlling wavelength switching of open all-photonic network is also provided.