Abstract: A plurality of datasets that correspond to ones of a plurality of digitized voice signal recordings are generated, each digitized voice signal recording corresponding to a support call between a technician and a support representative providing support to the technician, and each dataset including one or more textual documents that include words spoken during the corresponding support call between the technician and the support representative. An unsupervised learning model operative is trained using the datasets to identify clusters of the datasets based on similarity to a corresponding topic of a plurality of topics, each topic including a set of topic terms. Information that corresponds to at least one topic is output.

Abstract: An access point (AP) communicates to one or more stations (STAs) information indicating a set of channels and/or resource units (RUs), corresponding to a wideband basis service set (BSS) for an authorized unlicensed radio local area network (RLAN) using 6 GHz spectrum, which are subject to a lower power spectral density (PSD) level than other portions of the BSS frequency range or are not to be used for transmission. In some, but not necessarily all, embodiments, the AP and the STAs do not encode data on the identified channels or RUs, but do encode data on other channels and/or other RUs, which are not subject to the lower PDS constraint.

Abstract: Various embodiments include methods and devices for identity graphing of network genomes. Embodiments may include processing a node data set having unstructured node identifiers (IDs), determining potential node IDs from the node data set, determining node counts based on the potential node IDs, and classifying probabilities of accuracy of the node counts via a node count machine learning classifier model. Embodiments may further include classifying probabilities of accuracy of node IDs via a node ID machine learning classifier model. Embodiments may further include generating a data structure associating at least a first node ID of the classified node IDs with a network equipment ID. Embodiments may further include associating a classified probability accuracy of the first node ID with the first node ID.

Abstract: A method for Internet of Things (IoT) network security includes collecting information for each network device (device), determining a minimum viable resource allocation for each device based on the information, which defines the minimum resources needed by each device to engage the IoT network and handle data, and for each device, distributing minimum viable resource allocations and rules, determining monitoring sets, monitoring using the monitoring set, collecting updated information based partially on the monitoring set, analyzing the updated information to determine trends and insights relative to the devices and the IoT network, updating the monitoring set, minimum viable resource allocation, and rules based on the analyzed updated information, checking compliance with a current minimum viable resource allocation and rules, identifying devices having violations, and performing same on a continuous as it and automatic basis.

Abstract: A base station, e.g., a CBSD, selects or more user equipment (UE) devices, e.g., cell phones, to operate in monitoring mode. In monitoring mode the UE device, in addition to its normal communications with the base station, performs received power measurements in a frequency band, e.g., a CBRS band. The UE reports to the base station information indicating a UE determined measured power level attributed to incumbent user or measurement information used to derive a power level attributed to incumbent user. One or more base stations report received power measurement values attributed to incumbent user to a system management device, e.g., a SAS. The system management device processes the received information, e.g. combining information, and comparing a result to a threshold level. If the result exceeds the threshold the system management device sends a command to one or more base station to shut down DL traffic in the CBRS band.

Abstract: Apparatus and methods for providing interference management and load balancing in a wireless network. In one embodiment, the method and apparatus utilize quasi-licensed CBRS (Citizens Broadband Radio Service) wireless spectrum in conjunction with enhanced SAS (Spectrum Access System) and base station (e.g., CBSD) components to enable creation and management of virtual clusters of bases stations connected to the network, so as to enable inter-cluster interference mitigation, while also supporting load balancing between the base stations using coverage area overlap. In one implementation, the SAS reduces the coverage area (transmit power) of one or more base stations on a cluster edge to mitigate inter-cluster interference, and increases or adjusts the coverage of one or more base stations inside of the cluster to enable load balancing.

Abstract: In response to receiving instructions from a Spectrum Access System (SAS) to modify (e.g., reduce and/or cease) network transmissions in one or more Citizens Broadband Radio Service (CBRS) sub-bands following the SAS's detection of incumbent activity, the network's Operation Support System (OSS) predicts or measures received signal strength (RSS) levels based on the SAS's original instructions, estimates RSS levels based on modified versions of the SAS's instructions that prioritize higher-NPV (net present value) CBRS base stations (CBSDs) over lower-NPV CBSDs, selects a modified version whose RSS levels are approximately the same as or less than the RSS levels of the SAS's original instructions, and transmits to the SAS requests for the selected modified version. In this way, the OSS can achieves more-efficient allocation of the CBRS spectrum and therefore more-desirable network communications without interfering with the incumbent's communications.

Abstract: To avoid being overwhelmed by repeated authentication requests from faulty user equipment, in response to rejecting at least one authentication request from a UE, a wireless network (i) transmits a timeout message to the UE to instruct the UE to refrain from transmitting subsequent authentication requests for a timeout period corresponding to a specified timeout duration and/or (ii) establishes a local timeout period corresponding to the specified timeout duration during which the local network drops subsequent authentication requests received from the UE.

Abstract: Apparatus and methods for managing content delivery in a packetized network. In one embodiment, the network provide content to a plurality of clients via a plurality of nodes and origin points, and resources are discreetly represented (e.g., with IP addresses, such as those afforded under the IPv6 protocol) to allows for direct advertisement of resources. Exemplary solutions described herein further advantageously leverage extant architectures and protocols (such as BGP), and make use of a common control plane, which can be utilized for example by different content delivery network (CDN) operators and different delivery components to advertise resources. Internally within a given CDN, increased granularity of resource addressing and advertisement may provide benefits including: (i) resource affinity; (ii) resource-level balancing; (iii) dynamic resource scoping; and (iv) “zero-touch” provisioning and resource relocation.

Abstract: Methods and systems for wireless coverage adjustment in a system with Citizens Broadband Radio Service (CBRS). A method includes receiving, by a coverage adjustment engine from a spectrum access system (SAS), coverage reduction messages for one or more CBRS devices (CBSDs) in a wireless network, identifying, by the coverage adjustment engine, one or more coverage expansion base stations in the wireless network to mitigate an impact of wireless coverage area gaps resulting from the coverage reduction messages, and sending, by the coverage adjustment engine to each of the one or more coverage expansion base stations, expansion parameters to adjust an associated wireless coverage area to mitigate the wireless coverage area gaps, where adjustments to one or more associated wireless coverage areas are done prior to execution of the coverage reduction messages at the one or more CBSDs.

Abstract: Apparatus and methods for improving coverage and/or power-saving in a wireless network. In one embodiment, the apparatus and methods provide mechanisms for wireless user device (such as e.g., a 3GPP-compliant UE) to utilize downlink transmissions related to a first access technology (e.g., LTE) to enhance the detection and reception of downlink transmissions from a second access technology (e.g., 5G NR). In one variant, transmissions relating both access technologies are multiplexed on a shared frequency channel; a UE trying to detect 5G NR transmissions from a gNB leverages the easier detection of certain LTE signals such as PSS to enhance detection of the 5G NR signals. Moreover, the location of the LTE PSS (or SSS) can be used to indirectly assist a UE with detection of the NR SSB, such as via reduction of the number of noise-only samples passed to the NR PSS/SSS autocorrelation detector.

Abstract: Communication management hardware receives a first preliminary wireless pathloss model indicating expected pathloss between a first wireless base station and different locations in a first geographical region. Based on testing, the communication management hardware also receives feedback generated by multiple wireless stations. The feedback indicates actual pathloss associated with the wireless stations in the first geographical region receiving wireless signals from the first wireless base station. The communication management hardware applies the feedback to the first preliminary pathloss model to produce an updated first pathloss model associated with the first geographical region. The communication management hardware repeats this process of producing a respective pathloss model for each of multiple different types of geographical regions. The pathloss models are then subsequently used to determine pathloss in environments similar to the tested geographical regions.

Abstract: Various embodiments comprise systems, methods, architectures, mechanisms and apparatus for managing user equipment (UE) communications in a converged network by causing such UE to automatically select a “best” wireless network (cellular RAN/WiFi) without any user notification/interaction to the user of the UE, such as via the device user interface (UI). The best wireless network may be determined using contemporaneous QoS related measurements, or predicted QoS measurements such as provided by a RAN node or channel QoS model generated via machine learning techniques, which may also be used to determine a type of QoS needed for each application or service commonly invoked at the UE so as to automatically provide the best possible user experience without any user intervention.

Abstract: Methods, devices, and systems for providing a centralized access control system. A method includes receiving, at a control policy database, control policies for devices registered with the system. A machine learning engine updates the control policies to include websites identified by the machine learning engine as malicious. The system receives a request for access by a registered device using one of the multiple access platforms, determines a control policy for the registered device based on a variety of parameters including an access platform type of the multiple access platforms, and sends an access decision based on the determined control policy to an access device associated with the access platform type.

Abstract: In certain embodiments, an access point (AP) of a wireless, e.g., WiFi, network actively assists user equipment (UE) in the onboarding of an IoT device onto the network. The AP determines whether the UE is using AP mode or EZ mode onboarding. If EZ mode, then the AP determines whether bandsteering is needed and, if so, the AP actively tries to get the UE to switch to the 2.4 GHz band. If switching is not successful, then the AP may relay incoming IoT onboarding messages received from the UE in a non-2.4 GHz band as outgoing IoT onboarding messages in the 2.4 GHz band to attempt to associate the IoT device with the AP. Network-assisted onboarding can address problems that would otherwise result in failed onboarding such as when the user forgets to switch the UE to the 2.4 GHz band for EZ mode onboarding.

Abstract: Data is collected at access points about mobile communications systems, e.g., vehicles, use of access points and the quality of the connections that are provided. The collected information is provided to a machine learning access point (MLAP) model training device that generates one or more MLAP prediction models with each model corresponding to a geographic area in which access points are located. Vehicles provide location, acceleration and/or other information, such as a user entered destination. A vehicle request for an AP recommendation includes this information which allows the AP prediction engine to predict a path of travel. The MLAP prediction engine uses the stored models and/or information about AP loading, to predict what AP a vehicle requesting an AP recommendation should attach to next. The predicted AP is returned as a handoff recommendation along with the location and/or time the vehicle should initiate the handoff to the recommended AP.

Abstract: A communication management resource distributes first access configuration settings to a first wireless access point. The first access configuration settings are operative to control selection, by the first wireless access point, of at least one backhaul amongst multiple backhauls available for use by the first wireless access point. In other words, the first wireless access point uses the first access configuration settings as a basis in which to select amongst the multiple backhauls to communicate with a remote. The first wireless access point generates feedback based on the first wireless access point providing multiple mobile communication devices access to the remote network via the first access configuration settings. The first wireless access point communicates the feedback to the communication management resource.

Abstract: A plurality of training inputs are obtained, each training input including information associated with a digitized voice signal recording including speech signals of a corresponding support call between a technician and a support representative providing support to the technician. A machine-learned support call classification model is trained using the training inputs to generate classification outputs indicative of a particular support call classification for each respective training input based on a loss function that evaluates a difference between the particular support call classification and a ground-truth support call classification associated with each respective training input.

Abstract: It is determined that a request has been received to make a change in a radiation pattern of an antenna coupled to a Citizens Broadband Radio Service Device (CBSD) from a first radiation pattern to a second radiation pattern. A registration object associated with the CBSD maintained by a spectrum access system (SAS) is obtained. A new radiation pattern characteristic of the second radiation pattern that is different from an existing radiation pattern characteristic of the first radiation pattern is determined. A new registration object that includes the new radiation pattern characteristic and omits the existing radiation pattern characteristic is generated. The new registration object is caused to be stored by the SAS in lieu of the registration object.

Abstract: A wireless system as discussed herein includes a communication management resource configured to control flows of data through a first wireless access point to a remote network. To control the flows of data, the communication management resource receives a schedule indicating a pattern of multiple mobile communication devices associated with a subscriber domain. The schedule may indicate a respective pattern associated with the multiple mobile communication devices. For example, the respective pattern may indicate times when the multiple mobile communication devices are present in the subscriber domain and/or times when the communication devices communicate through the first wireless access point to one or more destination in a remote network. The communication management resource utilizes the schedule to control priority of conveying communications from the multiple mobile communication devices through the first wireless access point to the remote network.