Abstract: Systems and methods are provided for optimizing resource consumption by bringing intelligence to the key allocation process for fast roaming. Specifically, embodiments of the disclosed technology use machine learning to predict which AP a wireless client device will migrate to next. In some embodiments, machine learning may also be used to select a subset of top neighbors from a neighborhood list. Thus, instead of allocating keys for each of the APs on the neighborhood list, key allocation may be limited to the predicted next AP, and the subset of top neighbors. In some embodiments, a reinforcement learning model may be used to dynamically adjust the size of the subset in order to optimize resources while satisfying variable client demand.

Abstract: Systems and methods are provided for defining a split multi-link system. For example, the split multi-link system can define a multi-link device (MLD), where the MLD operates a first radio from a first AP of a plurality of APs and a second radio from a second AP of the plurality of APs through a Media Access Control (MAC) Service Access Point (SAP) of the split multi-link system. The first radio from the first AP and the second radio from the second AP may operate at different frequencies. The split multi-link system may also affiliate the first radio from the first AP and the second radio from the second AP with the MLD, where, using the MLD, the split multi-link system can transmit or receive one or more data frames from the MLD to a non-MLD or MLD client device using the first radio or the second radio.

Abstract: Examples described herein provide automatic location of access points by a computing device. Examples may include receiving, by the computing device from each AP in a subset of a plurality of APs, a Global Navigation Satellite System (GNSS) signal measurement, and based on each received GNSS signal measurement, constraining, by the computing device, the map of relative AP locations by at least one translational degree of freedom or one rotational degree of freedom. Examples may include resolving, by the computing device, locations of the plurality of APs in the map of relative AP locations.

Abstract: Systems and methods for determining a physical location of access points in a wireless network that include a plurality of access points, the plurality of access points in the wireless network including anchor access points with respective known locations and unanchored access points without respective known locations, may include: a first plurality of unanchored access points neighboring the anchor access points receiving known-location information from the anchor access points, performing range measurements to the anchor access points to determine their respective locations in the wireless network, thereby becoming pseudo-anchor access points; a second plurality of unanchored access points in communicative contact with a plurality of the pseudo-anchor points receiving the determined-location information from the pseudo-anchor access points, performing range measurements to the pseudo-anchor access points to determine their respective locations in the wireless network, thereby becoming pseudo-anchor access p

Abstract: Systems and methods are provided for defining a split multi-link system. For example, the split multi-link system can define a multi-link device (MLD), where the MLD operates a first radio from a first AP of a plurality of APs and a second radio from a second AP of the plurality of APs through a Media Access Control (MAC) Service Access Point (SAP) of the split multi-link system. The first radio from the first AP and the second radio from the second AP may operate at different frequencies. The split multi-link system may also affiliate the first radio from the first AP and the second radio from the second AP with the MLD, where, using the MLD, the split multi-link system can transmit or receive one or more data frames from the MLD to a non-MLD or MLD client device using the first radio or the second radio.

Abstract: Systems and methods are provided for defining basic service set (BSS) colors and spatial reuse groups (SRGs) for multiple basic service set identifier (MBSSID) sets. The systems and methods can set a first BSS color for a first MBSSID set and a second BSS color for a second MBSSID set, the MBSSID sets pertaining to a first access point. For a second access point, the system can set a third BSS color for a third MBSSID set and a fourth BSS color for a fourth MBSSID set. A SRG can be defined for the first MBSSID set, wherein the first spatial reuse group advertises the first BSS color and any BSS colors associated with the second access point.

Abstract: Systems and methods for dynamic client segregation in 802.11ax networks includes determining, for each access point of a plurality of access points in a network, a ratio comprising a number of fully uplink multi-user capable client devices to a total number of client devices connected to the access point; and steering, based on at least the ratio, client devices of the network to particular access points of the plurality of access points.

Abstract: Systems and methods are provided for seamless roaming in a network. First, a client device is authenticated at a first access point of the network. Next, a processor selectively determines, among remaining access points in the network, second access points at which respective precursor keys, such as Pairwise Master Keys R1 (PMK-R1s) are to be computed. The second access points are determined based on any of respective path losses from the first access point to the second access points and respective historical frequencies at which the client device associates at the respective remaining access points. For the second access points, the respective PMK-R1s are computed and transmitted to the second access points to be cached. Next, following a request from the client device to reassociate to a second access point of the second access points, the client device is authenticated at the second access point based on a corresponding PMK-R1.

Abstract: Examples described herein provide channel assignments for ranging between network devices in a network during beacon intervals. Examples described herein may assign a plurality of channels to a plurality of network devices based on a plurality of channel assignment permutations, and initiate, during beacon intervals, ranging measurements between the plurality of network devices on the plurality of channels based on each of the channel assignment permutations to generate ranging results. Examples described herein may determine whether total ranging measurements are performed for a threshold percentage of available links between the plurality of network devices on the plurality of channels, wherein the total ranging measurements include the ranging measurements based on each of the channel assignment permutations.

Abstract: Examples described herein provide coordinated ranging between APs in a network. Examples described herein may receive, by a computing device, neighbor adjacency information for a plurality of access points (APs) in a network, and based on the neighbor adjacency information, assign, by the computing device, subsets of the plurality of APs to non-overlapping zones. Examples described herein may initiate, by the computing device for each of the non-overlapping zones, ranging measurements between the subset of APs in the zone to generate a ranging result, wherein the ranging measurements between the subsets of APs in the non-overlapping zones are performed in parallel. Examples described herein may, based on the ranging results, resolve, by the computing device, locations of the plurality of APs.

Abstract: Examples described herein provide ranging by a network device during a beacon interval. Examples described herein may discover, by a network device during a first beacon interval, a plurality of APs in a network that are capable of ranging and a received signal measurement of each of the plurality of APs. Examples described herein may select, by the network device, an AP with a strongest received signal measurement among the plurality of APs, determine, by the network device, whether the selected AP is available for ranging, and based on a determination that the selected AP is available for ranging, initiate, by the network device during a second beacon interval, ranging measurements with the selected AP to generate a ranging result. Examples described herein may, based on the ranging result, resolve locations of the plurality of APs.

Abstract: Examples described herein mitigate adjacent channel interference for dual radios of a network device. Examples described herein may associate, by a network device, a first client device with a first radio of the network device, associate, by the network device, a second client device with a second radio of the network device, determine that the first and second client devices are within a steering threshold, and based on the determination that the first and second client devices are within the steering threshold, steer, by the network device, the second client device from the second radio to the first radio. Examples described herein may communicate, using the first radio of the network device, with the first and second client devices.

Abstract: Systems and techniques are described that are directed to intelligent scheduling of Wi-Fi services for applications, including enhanced dynamic prioritization. A device, such as an access point (AP), can receive data packets from multiple connected devices to dynamically identify an application flow for each data packet, and dynamically identify a user associated with the application flow for each data packet. The AP can generate prioritized candidate lists for selected data packets in queues corresponding to an access category (AC). In response to determining that the identified user associated with the application flow corresponds with a critical user, the AP can select data packets for the prioritized candidate lists based at least in part on priority policies for each of a plurality of applications and based at least in part on dynamic prioritization of applications for each of a plurality of applications; and schedule data packets from the prioritized candidate lists.

Abstract: Systems and methods are provided for optimizing resource consumption by bringing intelligence to the key allocation process for fast roaming. Specifically, embodiments of the disclosed technology use machine learning to predict which AP a wireless client device will migrate to next. In some embodiments, machine learning may also be used to select a subset of top neighbors from a neighborhood list. Thus, instead of allocating keys for each of the APs on the neighborhood list, key allocation may be limited to the predicted next AP, and the subset of top neighbors. In some embodiments, a reinforcement learning model may be used to dynamically adjust the size of the subset in order to optimize resources while satisfying variable client demand.

Abstract: Examples of performing selective Fine Timing Measurement (FTM) are described. For an FTM scan cycle, a first AP (i.e., an initiator) may determine scanning parameter values of a plurality of second APs (i.e., potential responders) based on previously performed FTM scans. The first AP may determine weights of the plurality of second APs based on the scanning parameter values and select a set of target APs based on the weights. The first AP may then scan the set of target APs for the FTM scan so that the rest of the plurality of second APs are relieved from participating in the FTM scan thereby reducing the performance impact on the rest of the plurality of second APs. After the FTM scan cycle is completed, the first AP may update the weights and select another set of target APs based on the updated weights to perform another FTM scan cycle.

Abstract: Systems and methods for providing enhanced Quality of Service (QoS) network transmissions can be based on an application sub-class or a user class. Systems and methods can include inspecting the information packet having a network level QoS field having a first network level QoS portion and a second network level QoS portion, determining an application sub-class or user class associated with the information packet, tagging the first network level QoS portion of the information packet according to a first network level QoS value, tagging the second network level QoS portion of the information packet according to a traffic priority indication and to a determined application sub-class or user class, and queuing the information packet for transmission from a network element based on the tagged first network level QoS portion and the second network level QoS portion.

Abstract: Examples described herein mitigate adjacent channel interference for dual radios of a network device. Examples described herein may associate, by a network device, a first client device with a first radio of the network device, associate, by the network device, a second client device with a second radio of the network device, determine that the first and second client devices are within a steering threshold, and based on the determination that the first and second client devices are within the steering threshold, steer, by the network device, the second client device from the second radio to the first radio. Examples described herein may communicate, using the first radio of the network device, with the first and second client devices.

Abstract: Embodiments are directed to automatic location of access points in a network. An embodiment of one or more non-transitory computer-readable storage mediums includes instructions for transmitting a request from a computing device to multiple access points in a network to determine a distance between each pair of access points of the multiple access points; receiving at the computing device the determined distances between each pair of access points; generating a proximity matrix containing the determined distances between each pair of access points; solving the proximity matrix to automatically generate a set of locations for the multiple access points; and orienting the generated set of locations for the multiple access points based on known locations of one or more anchor points in a subset of the access points.

Abstract: Systems and methods are provided for optimizing resource consumption by bringing intelligence to the key allocation process for fast roaming. Specifically, embodiments of the disclosed technology use machine learning to predict which AP a wireless client device will migrate to next. In some embodiments, machine learning may also be used to select a subset of top neighbors from a neighborhood list. Thus, instead of allocating keys for each of the APs on the neighborhood list, key allocation may be limited to the predicted next AP, and the subset of top neighbors. In some embodiments, a reinforcement learning model may be used to dynamically adjust the size of the subset in order to optimize resources while satisfying variable client demand.

Abstract: An example network information aggregator is disclosed. The network information aggregator includes a network interface, a memory, and processing circuitry. The processing circuitry is to receive a machine learning model trained using initialization data from a model creation device. The processing circuitry is also to generate a device steering rule to steer a client device from a first radio to a second radio using the machine learning model. The processing circuitry is also to send a steer command to the first radio.