Abstract: Dynamic bandwidth expansion (DBE) with channel switching may be provided. Managing DBE with channel switching includes operating, by an access point (AP), at a baseline operating bandwidth. The AP expands from the baseline operating bandwidth to an expanded operating bandwidth for a client device associated with the AP. When the AP determines to switch from a first channel to a second channel, the AP reverts to the baseline operating bandwidth and switches from the first channel to the second channel.

Abstract: In one implementation, a device that executes a first agent obtains a conversation example for use by the first agent. The first agent initiates, based on the conversation example, a conversation with one or more artificial intelligence agents. The first agent records the conversation. The device provides the conversation as synthetic data to train or test an artificial intelligence model.

Abstract: Interoperable Transmit Power Envelop (TPE) signaling with Automated Frequency Coordination (AFC) frequency response may be provided. First, AFC information may be received. Next a mask may be determined for a punctured channel indicated in the AFC information. Then a first amount may be determined that the mask needs to be altered to reach an AFC response for the punctured channel indicated in the AFC information. A Transmit Power Envelop (TPE) value may then be reported for the punctured channel comprising the first amount plus a second amount.

Abstract: In one implementation, a device distills raw data regarding a past experience of an agentic system completing a first task into distilled experience data, the agentic system comprising a plurality of artificial intelligence-based agents that cooperated to complete the first task. The device receives a request for the agentic system to perform a second task. The device retrieves the distilled experience data and the raw data regarding the past experience in part by performing a matching of the second task to the distilled experience data. The device provides the distilled experience data and the raw data regarding the past experience to the agentic system for use to complete the second task.

Abstract: In some implementations, a device assigns sub-tasks to a plurality of artificial intelligence-based agents in a multi-agent system to complete a task, based on information indicative of their capabilities stored in an agent directory. The device makes a determination that a particular agent from among the plurality of artificial intelligence-based agents is unable to perform its assigned sub-task based on telemetry data regarding performance of the sub-tasks by the plurality of artificial intelligence-based agents. The device re-assigning, based on the determination, a sub-task from the particular agent to another agent in the multi-agent system during completion of the task. The device updates the information in the agent directory to reflect the determination.

Abstract: Optimal Power Spectral Density (PSD) for unpunctured bandwidth may be determined. First, a plurality of (un)punctured modes that can be signaled in a maximum bandwidth of interest may be identified. Next, a plurality of grades may be determined where each of the plurality of grades respectively corresponds to each of the plurality of (un)punctured modes. Then a one of the plurality of (un)punctured modes that is associated with a best of the plurality of grades may be selected.

Abstract: In one implementation, a device may determine that an operations supply chain for a quantum noise source is a compromised supply chain. The device may discontinue generation of a quantum random number utilizing the compromised supply chain. The device may initiate sourcing of trusted quantum noise from a trusted quantum noise source. The device may cause the trusted quantum noise to be utilized in the generation of a quantum random number.

Abstract: In one implementation, a device that executes a metrics agent obtains a description of a use case for an agentic system that comprises one or more artificial intelligence-based agents. The device retrieves example rubrics based on the description of the use case. The device generates a set of rubrics and performance metrics by inputting the description of the use case and the example rubrics to the metrics agent. The device provides the set of rubrics and performance metrics to test how well the agentic system performs for the use case.

Abstract: The present disclosure describes a technique for access points to set guard intervals. An apparatus includes a first radio, one or more memories, and one or more processors communicatively coupled to the one or more memories. The first radio receives a reflection comprising a signal reflected by a structure. A combination of the one or more processors determines, based on the reflection, a position of the structure, sets a first guard interval based on the position of the structure, and transmits a first message according to the first guard interval.

Abstract: This disclosure describes techniques and mechanisms for providing initialization vector (IV) uniqueness and extending rekeying windows for network devices that perform secure association sharing in multi-source and multi-destination environments. The techniques may apply to existing hardware of the network devices. The techniques may enable the network devices to execute in a non-XPN mode and establish secure tunnels corresponding to secure association sessions. The techniques may utilize software to partition extra bits included in packet headers. The network devices may perform a process to update a loop count value utilizing a portion of the extra bits, thereby exponentially extending the rekeying windows. Further, by utilizing a portion of the extra bits, the system may ensure IV uniqueness for the secure association session between network devices in the multi-source and multi-destination environment.

Abstract: Techniques are described for providing users of a data intake and query system with pre-trained ML models capable of identifying malicious threats (e.g., malware, botnets, ransomware, etc.) in users' computing environments based on an analysis of Domain Name System (DNS) log data collected from DNS servers in users' environments. DNS log data is ingested by a data intake and query system and processed to obtain searchable timestamped event data. This event data can then be used as input to ML models provided by a security ML application described herein to detect potential occurrences of malicious activity within users' computing environments.

Abstract: An embodiment uses entanglement and quantum teleportation to build a quantum backbone network. A network interface interconnects packetized quantum networks with entanglement-based quantum backbone networks.

Abstract: In one implementation, a device identifies a sequence of related prompts for input to a generative model. The device makes individual maliciousness assessments of those prompts in the sequence of related prompts. The device makes a collective maliciousness assessment of the sequence of related prompts. The device prevents at least a portion of the sequence of related prompts from being input to the generative model, when any of the individual maliciousness assessments or the collective maliciousness assessment indicates a malicious request to the generative model.

Abstract: Techniques and mechanisms for enabling an end-to-end controller of a network to offload decisions about placement of network slices to the transport controller. By doing so, the transport controller can intelligently determine optimal placement based on intent (e.g., external intent of the end-to-end controller and internal intent of the transport controller), internal transport network analytics functions, SLO/SLE constraints, real-time telemetry data, and more to provide dynamic and optimal placement of network slices. That is the described techniques dynamically utilize information from within the transport domain to define the placement of network slices. Thus, placements are more accurate, thereby reducing latency and improving functioning of the network, as well as providing an improved user experience by meeting SLO/SLE constraints.

Abstract: Techniques and architecture are described that leverage a trusted cryptographic channel by a network controller to authenticate an unfamiliar host key found on a network device when the network controller is attempting to establish a secure connection over a secure channel. The network controller may authenticate an unfamiliar host key found on a network device by using the trusted cryptographic channel to retrieve the network device's host keys directly from the network device when the network controller encounters the unfamiliar host key. Additionally, the network controller may correlate all the audit events available to the network controller with the appearance of the unfamiliar host key and accepting the unfamiliar host key only if the unfamiliar host key is confirmed to be present on the network device and if the audit events indicate that the unfamiliar host key was recently properly installed or generated on the network device.

Abstract: In one implementation, a device identifies a first set of one or more Common Vulnerabilities and Exposures (CVEs) by searching a CVE database based on a request sent via a network towards a service. The device also identifies a second set of one or more CVEs by querying a large language model (LLM) based on the request. The device determines that the request is associated with a particular CVE based on the first set of one or more CVEs and the second set of one or more CVE. The device initiates a corrective measure with respect to the request in the network.

Abstract: Techniques for improved pattern and receive diversity in wireless systems are provided. Orientation data for a first access point (AP) of a plurality of APs in an ultra-wideband (UWB) deployment is accessed. An antenna configuration for the first AP is selected based at least in part on the orientation data. A first clock synchronization frame received, from a second AP and based on the antenna configuration, at a first timestamp, the first clock synchronization frame comprising a second timestamp. A clock offset is generated based on the first and second timestamps, and ranging is performed with one or more client devices, in conjunction with the second AP, based at least in part on the clock offset.

Abstract: The present disclosure describes a network that uses dual NAVs. An apparatus includes one or more memories and one or more processors communicatively coupled to the one or more memories. A combination of the one or more processors generates a message that includes (i) a first NAV indicating a first time period and (ii) a second NAV indicating a second time period and transmits the message. The first NAV indicates to a first access point assigned to an overlapping basic service set to defer transmissions during the first time period. The second NAV indicates to a second access point assigned to a MAPC group that transmissions are allowed during the second time period.

Abstract: According to one or more embodiments of the disclosure, an example process herein may comprise: determining, by a process, one or more solution bundles of particular services in an extensibility platform, each solution bundle of the one or more solution bundles submitted by a respective solution provider, the one or more solution bundles having workflow definitions and corresponding container images, the workflow definitions being custom-defined configurations of one or more triggers, one or more inputs, one or more outputs, and one or more sequential steps; listening, by the process, to published observability data of the extensibility platform; determining, by the process and from the published observability data, occurrence of a trigger condition based on the one or more triggers of the workflow definitions of the one or more solution bundles; and initiating, by the process, one or more particular workflows corresponding to the trigger condition in response to occurrence of the trigger condition.

Abstract: Systems and methods are disclosed that are directed to improving the prioritization and viewing of system alerts through the use of machine learning to group the alerts and prioritize the groupings. Additionally, a graphical user interface (GUI) is generated that illustrates the prioritized listing of the plurality of groupings. Thus, a system administrator or other user receives an improved experience as the number of notifications provided to the system administrator are reduced due to the grouping of individual alerts into related groupings and further due to the prioritization of the groupings. Additionally, user feedback may be obtained through user interaction with GUI, where the user feedback comprises either explicit feedback or implicit feedback. The machine learning model may then be retrained based on the user feedback.