Predictive artificial intelligence (AI)-based wireless station load balancing based on access point uplink utilization
AssigneeFortinet, Inc.
InventorsKarthikeyan Balasubramanian, Venkatesh Kannan
Life of this patent
Estimating the term…
Abstract
An uplink utilization is monitored for each station connected to an access point over a wireless network, including jitter, latency, and dropped packets. Uplink utilization is monitored for access points that are neighbors to the access point, as determined from neighbor reports. An AI model is generated from monitoring data. When an uplink threshold of the access point has been exceeded at the access point. A new access point is selected from the AI model for at least one of the stations based at least on a least used uplink in addition to RF parameters.
Claims20
A computer-implemented method in a Wi-Fi network device for Artificial Intelligence (AI)-based wireless station load balancing based on access point uplink utilization, the method comprising the steps:
- monitoring uplink utilization for each station connected to an access point of a plurality of access points over a wireless network, including jitter, latency, and dropped packets;
- monitoring uplink utilization for access points from the plurality of access points that are neighbors to the access point, as determined from neighbor reports;
- generating an AI model of uplink utilization for the plurality of access points from monitoring data;
- detecting that an uplink threshold of the access point has been exceeded at the access point;
- and
- selecting, using the AI model, a new access point from the plurality of access points for at least one of the stations based at least on a least used uplink in addition to Radio Frequency (RF) parameters.
A non-transitory computer-readable medium storing computer-readable instructions in a Wi-Fi network device on a data communication network, that when executed by a processor, perform a method for Artificial Intelligence (AI)-based wireless station load balancing based on access point uplink utilization, the method comprising:
- monitoring uplink utilization for each station connected to an access point over a wireless network, including jitter, latency, and dropped packets;
- monitoring uplink utilization for access points that are neighbors to the access point, as determined from neighbor reports;
- generating an AI model from monitoring data;
- detecting that an uplink threshold of the access point has been exceeded at the access point;
- and
- selecting, from the AI model, a new access point for at least one of the stations based at least on a least used uplink in addition to RF parameters.
A Wi-Fi network device for Artificial Intelligence (AI)-based wireless station load balancing based on access point uplink utilization, the Wi-Fi network device comprising:
- a processor;
- a network communication interface, communicatively coupled to the processor and to the data communication network;
- and
- a memory, communicatively coupled to the processor and storing source code executed by the processor, comprising:
- an uplink utilization module to monitor uplink utilization for each station connected to an access point over a wireless network, including jitter, latency, and dropped packets,
- wherein the uplink utilization module monitors uplink utilization for access points that are neighbors to the access point, as determined from neighbor reports;
- an AI model module to generate an AI model from monitoring data;
- a threshold module to detect that an uplink threshold of the access point has been exceeded at the access point;
- and
- a selection module to select, from the AI model, a new access point for at least one of the stations based at least on a least used uplink in addition to RF parameters.
Dependencies are inferred from the claim text.