Optimization of distributed Wi-Fi networks
Systems and methods for optimization of access points in a Wi-Fi system by a cloud controller include receiving inputs related to operation of the Wi-Fi system; performing an optimization based on the inputs to maximize an objective function which maximizes excess capacity for a load ratio considering a load desired by each Wi-Fi client device; and providing outputs comprising operational parameters for the Wi-Fi system based on the optimization. The optimization chooses which access point each Wi-Fi client device connects to in the Wi-Fi system.
1. A method for optimization of access points in a Wi-Fi system by a cloud controller, the method comprising:
receiving inputs related to operation of the Wi-Fi system, wherein the Wi-Fi system comprises a plurality of access points communicatively coupled to one another in a same Wi-Fi network with at least one access point communicatively coupled to or in a gateway providing external communication for the Wi-Fi system and with one or more remaining access points of the plurality of access points connected to one another and to the gateway via wireless backhaul links;
performing an optimization based on the inputs to maximize an objective function which maximizes a goal for the Wi-Fi system and for the plurality of access points operating together, wherein the objective function maximizes excess capacity for a load ratio in the plurality of access points considering a load desired by each Wi-Fi client in the Wi-Fi system with the load one or more of estimated based on recent history, estimated based on long term history, provided by an access point or the Wi-Fi client, and the load set to a reserved minimum; and
providing outputs comprising operational parameters for the Wi-Fi system and for the plurality of access points based on the optimization and causing the Wi-Fi system and the plurality of access points to implement the operational parameters, wherein the operational parameters comprise channels and topology for the wireless backhaul links.
2. The method of claim 1 , wherein the inputs comprise a plurality of traffic load required by each Wi-Fi client device, signal strength for each possible link, data rate for each possible link, packet error rates on each link, strength and load on in network interferers, and strength and load on out of network interferers; and
wherein the outputs comprise a plurality of channel and bandwidth (BW) selection, routes and topology, Request to Send/Clear to Send (RTS/CTS) settings, Transmitter (TX) power, clear channel assessment, client association steering, band steering, Arbitration inter-frame spacing (AIFS), and Wi-Fi contention windows.
3. The method of claim 1 , wherein the load desired by each Wi-Fi client is an input to the optimization, and the load is determined by one or more of measured by the access points, estimated based on previous measurements, or unknown and set to an assumed value.
4. The method of claim 3 , wherein the load desired by each Wi-Fi client is set at a minimum reservation capacity.
5. The method of claim 1 , wherein the optimization is performed for the Wi-Fi system and one or more additional Wi-Fi systems which are clustered.
6. The method of claim 1 , wherein the operational parameters are set such that one or more of the following are true: not all of the access points are used, Wi-Fi client devices do not necessarily associate with a closest access point, and backhaul links utilize different channels.
7. The method of claim 1 , wherein the outputs define a topology of the access points in the Wi-Fi system in a tree structure.
8. The method of claim 1 , wherein the outputs define a topology in which at least one node has two or more parents and multi-path Transmission Control Protocol (TCP) is utilized for communication between the two or more parents.
9. The method of claim 1 , wherein the optimization function incorporates a cost for making changes to the operational parameters for the Wi-Fi system.
10. The method of claim 1 , further comprising:
applying a hysteresis threshold to the outputs and performing the providing based on the hysteresis threshold.
11. A cloud controller for a Wi-Fi system configured to provide optimization, the cloud controller comprising:
a network interface communicatively coupled to the Wi-Fi system;
one or more processors; and
memory storing instructions that, when executed, cause the one or more processors to:
receive inputs related to operation of the Wi-Fi system, wherein the Wi-Fi system comprises a plurality of access points communicatively coupled to one another in a same Wi-Fi network with at least one access point communicatively coupled to or in a gateway providing external communication for the Wi-Fi system and with one or more remaining access points of the plurality of access points connected to one another and to the gateway via wireless backhaul links;
perform an optimization based on the inputs to maximize an objective function which maximizes a goal for the Wi-Fi system and for the plurality of access points operating together, wherein the objective function maximizes excess capacity for a load ratio in the plurality of access points considering a load desired by each Wi-Fi client in the Wi-Fi system with the load one or more of estimated based on recent history, estimated based on long term history, provided by an access point or the Wi-Fi client, and the load set to a reserved minimum; and
provide outputs comprising operational parameters for the Wi-Fi system and for the plurality of access points based on the optimization and cause the Wi-Fi system and the plurality of access points to implement the operational parameters, wherein the operational parameters comprise channels and topology for the wireless backhaul links.
12. The cloud controller of claim 11 , wherein the inputs comprise a plurality of traffic load required by each Wi-Fi client device, signal strength for each possible link, data rate for each possible link, packet error rates on each link, strength and load on in network interferers, and strength and load on out of network interferers; and
wherein the outputs comprise a plurality of channel and bandwidth (BW) selection, routes and topology, Request to Send/Clear to Send (RTS/CTS) settings, Transmitter (TX) power, clear channel assessment, client association steering, band steering, Arbitration inter-frame spacing (AIFS), and Wi-Fi contention windows.
13. The cloud controller of claim 11 , wherein the load desired by each Wi-Fi client is an input to the optimization, and the load is determined by one or more of measured by the access points, estimated based on previous measurements, or unknown and set to an assumed value.
14. The cloud controller of claim 11 , wherein the optimization is performed for the Wi-Fi system and one or more additional Wi-Fi systems which are clustered.
15. The cloud controller of claim 11 , wherein the operational parameters are set such that one or more of the following are true: not all of the access points are used, Wi-Fi client devices do not necessarily associate with a closest access point, and backhaul links utilize different channels.
16. The cloud controller of claim 11 , wherein the outputs define a topology of the access points in the Wi-Fi system in a tree structure.
17. The cloud controller of claim 11 , wherein the outputs define a topology in which at least one node has two or more parents and multi-path Transmission Control Protocol (TCP) is utilized for communication between the two or more parents.
18. A Wi-Fi system configured for optimization by a cloud controller, the Wi-Fi system comprising:
a plurality of access points communicatively coupled to one another and at least one access point communicatively coupled to or in a gateway providing external communication for the Wi-Fi system and with one or more remaining access points of the plurality of access points connected to one another and to the gateway via wireless backhaul links; and
a cloud-based system configured to
receive inputs related to operation of the Wi-Fi system;
perform an optimization based on the inputs to maximize an objective function which maximizes a goal for the Wi-Fi system and for the plurality of access points operating together, wherein the objective function maximizes excess capacity for a load ratio in the plurality of access points considering a load desired by each Wi-Fi client in the Wi-Fi system with the load one or more of estimated based on recent history, estimated based on long term history, provided by an access point or a Wi-Fi client, and the load set to a reserved minimum; and
provide outputs comprising operational parameters for the Wi-Fi system and for the plurality of access points based on the optimization and cause the Wi-Fi system and the plurality of access points to implement the operational parameters, wherein the operational parameters comprise channels and topology for the wireless backhaul links.