Method and cloud server for cloud based client steering
Methods, systems, and apparatus' for optimizing network performance are described herein. A Cloud-Server may optimize one or more mesh networks of a plurality of 802.11 access points (APs) and stations (STA) connected to the APs. The Cloud-Server may work with a client steering daemon (CSD) miming on each AP. The Cloud-Server may operate in a location that is remote to the AP(s). The Cloud-Server may collect, store, and process network and client related data from one or more CSDs. The processing may include measuring network (e.g., premises) and client (e.g., STA) performance as well as analyzing (e.g., machine learning, etc.) to determine network parameters that will optimize network performance The Cloud-Server may then apply these network parameters to the relevant CSDs in order to improve per client and per network performance.
1. A method for mesh network performance optimization performed by a Cloud-Server, the method comprising:
receiving data from a plurality of client steering daemons (CSDs) each operating on an 802.11 access point (AP);
performing processing on the received data, wherein the processing includes performing at least machine learning on the received data to determine a pair of client steering profiles for each CSD of the plurality of CSDs, wherein the pair of client steering profiles include a client steering decision profile (SDP) and a client steering mechanism profile (SMP), wherein the SDP is a set of parameters used for computing a cost of steering a client STA, and wherein the SMP is a set of actions for steering the client STA based on the corresponding SDP of a pair of client steering profiles;
generating a pool of client steering profiles comprising a set of pairs of SDPs and SMPs based on the processing; and
sending a subset of the generated pool of client steering profiles to each of the plurality of CSDs to enable client steering of the client STA based on a pair of client steering profiles from the subset of client steering profiles.
2. The method of claim 1 , wherein the received data includes one or more of the following: station (STA) capabilities, STA RSSI/TX rate samples, STA traffic demand, interference measurements from access points (APs) and STAs, STA behavioral events, and AP metrics.
3. The method of claim 1 , wherein the processing further includes consideration of cost metrics.
4. The method of claim 3 , wherein the cost metrics include or more or more of the following: end-to-end throughput, end-to-end traffic aware throughput, load balancing, STA technology type, STA capability.
5. The method of claim 4 , further comprising sending the cost metrics to the plurality of CSDs.
6. The method of claim 1 , further comprising receiving a plurality of queries from the plurality of CSDs prior to receiving data, and in response, sending a request for data to the plurality of CSDs.
7. The method of claim 1 , further comprising determining the subset of client steering profiles based on the received data from the plurality of CSDs such that the subset of client steering profiles is relevant to the receiving CSD.
8. A Cloud-Server for mesh network performance optimization, the Cloud-Server comprising:
a processor; and
a transceiver for sending and receiving signals coupled to the processor, wherein the processor and transceiver are configured to:
receive data from a plurality of client steering daemons (CSDs) each operating on an 802.11 access point (AP);
perform processing on the received data, wherein the processing includes performing at least machine learning on the received data to determine a pair of client steering profiles for each CSD of the plurality of CSDs, wherein the pair of client steering profiles include a client steering decision profile (SDP) and a client steering mechanism profile (SMP), wherein the SDP is a set of parameters used for computing a cost of steering a client STA, and wherein the SMP is a set of actions for steering the client STA based on the corresponding SDP of a pair of client steering profiles;
generate a pool of client steering profiles comprising a set of pairs of SDPs and SMPs based on the processing; and
send a subset of the generated pool of client steering profiles to each of the plurality of CSDs to enable client steering of the client STA based on a pair of client steering profiles from the subset of client steering profiles.
9. The Cloud-Server of claim 8 , wherein the received data include one or more of the following: station (STA) capabilities, STA RSSI/TX rate samples, STA traffic demand, interference measurements from access points (APs) and STAs, STA behavioral events, and AP metrics.
10. The Cloud-Server of claim 8 , wherein each CSD of the plurality of CSDs operates on an AP.
11. The Cloud-Server of claim 8 , wherein the processing further includes consideration of cost metrics.
12. The Cloud-Server of claim 11 , wherein the cost metrics include one or more of the following: end-to-end throughput, end-to-end traffic aware throughput, load balancing, STA technology type, STA capability.
13. The Cloud-Server of claim 12 , wherein the processor and transceiver are further configured to send cost metrics to the plurality of CSDs.
14. The Cloud-Server of claim 8 , wherein the processor and transceiver, prior to receiving data, are further configured to: receive a plurality of queries from the plurality of CSDs, and in response, send a request for data to the plurality of CSDs.
15. The Cloud-Server of claim 8 , wherein the processor and transceiver are further configured to determine the subset of client steering profiles based on the received data from the plurality of CSDs such that the subset of client steering profiles is relevant to the receiving CSD.