Abstract: Systems and methods for configuring and managing a network device with a transformer model under control of a network control function (NCF) are disclosed. A processor of the NCF receives a request that identifies a network management task and associated service targets. The processor forms a token set of schema-defined tokens that represent network context, applies positional encodings to generate an ordered token sequence, and invokes the transformer model to produce a configuration patch. The configuration patch is validated against a schema-constrained decoder that enforces device grammar and is applied to the target network device. Device state and telemetry are read back to obtain a read-back state, which is evaluated against the service targets.

Abstract: Systems and methods for configuring and managing a network device with a transformer model under control of a network control function (NCF) are disclosed. A processor of the NCF receives a request that identifies a network management task and associated service targets. The processor forms a token set of schema-defined tokens that represent network context, applies positional encodings to generate an ordered token sequence, and invokes the transformer model to produce a configuration patch. The configuration patch is validated against a schema-constrained decoder that enforces device grammar and is applied to the target network device. Device state and telemetry are read back to obtain a read-back state, which is evaluated against the service targets.

Abstract: Various methods and edge computing systems are disclosed that implement a versatile elastic edge compute (vEEC) system that includes a computing mesh including two or more edge computing nodes (ECNs). The ECNs may be configured to identify ECNs in the vEEC and their capabilities, determine resource requirements for one or more software applications or tasks within the vEEC system, dynamically scale network resources based on the determined requirements, resolve network congestion by redistributing tasks among ECNs based on network traffic analysis, implement failover to cloud resources for ECNs that face resource limitations, offload computational tasks from edge devices, monitor network performance and resource utilization for adjustments, and refine resource allocation models and system configurations based on feedback and performance metrics.

Abstract: An edge computing system configured to dynamically offload tasks from a user device to an edge device. The edge device may receive a request to download or run a software application suitable processing a sensory feed collected by the user device. The edge device may determine operating requirements of the software application, determine the internal capabilities of the edge device, and compare the determined operating requirements of the software application to the determined capabilities of the edge device to determine whether the edge device is capable of running the entire software application. The edge device may also determine whether the edge device is capable of running a portion of the software application in response to determining that the edge device is not capable of running the entire software application. The edge device may send a message identifying portions that could be run on the edge device.

Abstract: A method for establishing connections and forming groups in an edge computing system includes detecting and identifying devices attempting to connect to the network using a processor. The method involves authenticating detected devices with a common pre-shared key (PSK) stored in memory, forming groups of connected devices based on predefined criteria, and sharing the PSK within each group via a secured channel. It also includes creating a subnetwork or private LAN for each subscriber using network configuration data, assigning virtual pre-shared keys (vPSKs) to devices based on service requirements, determining device capabilities by analyzing received device-specific information, and identifying supported applications based on device capabilities and application compatibility data stored in memory.

Abstract: An edge computing system configured to dynamically distribute in an Elastic Edge Computing Platform (ECP) that includes one or more processors or devices configured to perform dynamic grouping operations for the automatic formation of logical groups that may be used for more efficient resource sharing among edge nodes and devices. A processor in the ECP system may collect information about edge nodes and associated end devices within the ECP system, use the collected information to determine characteristics of the edge nodes and end devices, determine a group objective for each of groups, request membership to one of available groups, cluster the edge nodes based on the determined characteristics and group memberships, and dynamically adjust the group configurations or group memberships based on changes in conditions in the ECP system.

Abstract: This disclosure describes methods for distributed machine learning across edge nodes. An edge node acquires new data and checks with a federated server for an updated large language model (LLM) or small learning model (SLM). If available, it downloads the update; otherwise, it uses the existing model. The edge node then trains the LLM/SLM on its new data and sends the updated model back to the server. This enables continuous, privacy-preserving model improvement with low communication overhead. Additionally, methods are provided for sequential task execution at the edge: a smart contract provides a pointer to a covariance matrix, which is loaded into a neural network. The edge node processes sensor data to generate a task output, then repeats the process for subsequent tasks using new matrices and updated models as needed.

Abstract: This disclosure describes methods for adaptive machine learning in distributed edge computing. An edge node collects local data, selects a suitable large language model (LLM) or small learning model (SLM), trains it, and shares updates with a federated server or peer nodes. Another method matches AI functions with appropriate models, uses datasets with confidence values, and applies a Kalman Filter to assign weights and update covariance matrix confidence. In collaborative training, edge nodes store trained models with per-layer covariance values, transmit them to a control node, and update models based on aggregated inputs. Blockchain may be used for secure model storage and distribution, with smart contracts managing access and updates. These approaches support efficient, privacy-preserving learning by adapting models using statistical confidence and decentralized coordination.

Abstract: Systems and methods for adaptive edge computing using artificial intelligence (AI) include monitoring real-time accuracy of a neural network by using a feedback loop configured to detect changes in inference accuracy and dynamically adjusting the structure of the neural network by adding or removing hidden layers based on monitored error rates and predetermined computational constraints. A Kalman gain computation determines neural network weight adjustments based on monitored error rates. Weight matrices undergo incremental updates derived from these adjustments. Incremental weight adjustments remain stored in memory to enable low-bandwidth model updates. The neural network stores inference results and refined weights in an inference result database. Pre-trained models periodically receive incremental updates based on stored adjustments. Predictive holistic inference logic (PHIL) applied to stored inference results improves the accuracy of the inference results.

Abstract: Authenticating a mobile user device with a mobile network operator in advance of an MNO authentication request from the mobile user device can be done by a satellite network sending an authentication request to the MNO system, obtaining a set of authentication vectors related to the mobile user device, storing them into a proxy home location register, receiving the MNO authentication request from the mobile user device, generating an authentication request response based on the authentication vectors, sending the authentication request response to the mobile user device, receiving an authentication response including a received signed response, comparing the received signed response with the stored signed response, and if the received signed response and stored signed response match, deem that to be a successful authentication, add an MNO location update message to a request queue and forward the MNO location update message to the MNO system over the channel when available.

Abstract: A method for processing a plurality of orders of shares of exchange traded funds for retirement plans includes providing a pool of at least one share of each of the exchange traded funds to be traded by the retirement plans, receiving an order from a retirement plan that specifies a first amount other than a whole number of shares of an exchange traded fund, identifying a second amount that is a whole number of shares nearest to the first amount, adding, from the pool, a number of shares to the first amount such that a sum of the number of shares and the first amount equals the second amount, directing an exchange to execute a trade for the second amount, and allocating the first amount to the retirement plan and a difference between the first amount and the second amount to the pool.

Abstract: A method and edge computing system configured to generate or process extended reality (XR) data. Processors in the edge computing system receive, process, and analyze a sensory feed from an optical device to generate analysis results that include a relative position of the optical device from surrounding objects. The processors generate mapper output results (that include virtual coordinates) based on the analysis results, request and receive information (e.g., salient points of interest, etc.), and compare the generated mapper output results to the received information to identify a correlation between a feature included in the processed sensory feed and a feature included in the received information. The processors generate and send augmented information to a renderer and/or send the processed sensory feed to a cloud object recognizer for further processing.

Abstract: Authenticating a mobile user device with a mobile network operator in advance of an MNO authentication request from the mobile user device can be done by a satellite network sending an authentication request to the MNO system, obtaining a set of authentication vectors related to the mobile user device, storing them into a proxy home location register, receiving the MNO authentication request from the mobile user device, generating an authentication request response based on the authentication vectors, sending the authentication request response to the mobile user device, receiving an authentication response including a received signed response, comparing the received signed response with the stored signed response, and if the received signed response and stored signed response match, deem that to be a successful authentication, add an MNO location update message to a request queue and forward the MNO location update message to the MNO system over the channel when available.

Abstract: An edge computing gateway (ECG) includes a processor configured to establish a distributed container network that includes a plurality of interconnected ECGs, each of which includes an orchestrator and containers. The processor may determine available resources on one or more additional ECGs in the distributed container network, select one of the additional ECGs in the distributed container network to run a container based on the available resources determined by the orchestrator of the ECG, and allocate resources from the selected ECG to execute the container. In response to changes in resource demands, the processor may dynamically reallocate resources between the ECG and the selected ECG to adjust container performance across the distributed container network.

Abstract: A method for establishing connections and forming groups in an edge computing system includes detecting and identifying devices attempting to connect to the network using a processor. The method involves authenticating detected devices with a common pre-shared key (PSK) stored in memory, forming groups of connected devices based on predefined criteria, and sharing the PSK within each group via a secured channel. It also includes creating a subnetwork or private LAN for each subscriber using network configuration data, assigning virtual pre-shared keys (vPSKs) to devices based on service requirements, determining device capabilities by analyzing received device-specific information, and identifying supported applications based on device capabilities and application compatibility data stored in memory.

Abstract: A method for automated device provisioning and management in an edge computing system includes retrieving self-provisioning parameters and device credentials from a database using a processor. The method involves adding a new device, determining if the allowed device limit for a group is exceeded, removing devices if necessary, and adding devices through a top-off process if the limit is not exceeded. It forms trusted and self-service device groups, monitors and manages devices based on policies, changes services or applications, dynamically changes device group associations, and assigns virtual pre-shared keys (vPSKs) to private LANs. The method also includes creating private LANs and virtual private clouds (vPCs), sharing device credentials with upstream services, and generating and using virtual networks (vNets) for device provisioning and management.

Abstract: A method for dynamically managing network segmentation, slicing, and scaling operations in an edge computing system includes retrieving system parameters and device credentials from a secure database using a processing system. The method involves segmenting the network by analyzing device types, capabilities, and roles, categorizing devices into logical groups, assigning devices to these groups, selecting appropriate policies, and storing group membership and policy information in a central database. It configures network slices by determining the need for dedicated slices for new service requests, initiating network discovery to map topology, selecting and assigning network elements, configuring and activating the slice, updating routing tables, enforcing access controls, and dynamically allocating resources in real-time based on each slice's requirements.

Abstract: An edge computing system configured to dynamically offload tasks from a user device to an edge device. The edge computing system may receive a processed sensory feed from the user device, analyze the received processed sensory feed, and generate mapper output results. The edge computing system may compare the generated mapper output results to information received from a datastore, identify a correlation between a feature included in the received processed sensory feed and a feature included in the received information, and determine whether a confidence value associated with the identified correlation exceeds a threshold value. The edge computing system may further process the received processed sensory feed locally in the edge computing system or send the received processed sensory feed to a cloud component for further processing based on whether the confidence value exceeds the threshold value.

Abstract: A multiple layer printed circuit board including a plurality of layers, vertical interconnect accesses (VIAs), and a vertical interconnect access (VIA) bridge. The layers may include signal layers, prepreg substrate layers disposed between the signal layers, ground plane layers, wherein each of the ground plane layers abuts one of the prepreg substrate layers, inner signal layers, wherein each of the inner signal layers abuts one of the prepreg substrate layers, and a core substrate layer disposed between the signal layers, wherein two of the inner signal layers abut opposed sides of the core substrate layer. The VIAs extend through at least some of the layers, wherein each of the VIAs is formed by aligned apertures through adjoining ones of the prepreg substrate layers, ground plane layers, and inner signal layers. The VIA bridge is coupled to the VIAs to convey heat to a heat sink.

Abstract: An edge device may be configured to generate a secure container to perform a software application on the edge device. A security daemon operating on a processor of the edge device may receive a configure host request message from a container manager. In response, the security daemon may determine integrity of metadata, extract licenses from the metadata, determine image permissions, create a user or group account, and update one or more system service access-control lists (ACLs). The security daemon may generate and send a configure host response message to the container manager, which may create and/or start the container.