Method and system for assigning one or more optimal wireless channels to a Wi-Fi access point using a cloud-based software defined network (SDN)

1. A method for assigning at least one optimal wireless channel to a Wi-Fi access point of a plurality of Wi-Fi access points using a cloud-based software defined network (SDN), the method comprising:

collecting, by one or more processors, radio frequency (RF) measurements from a plurality of client devices in a cloud computing network, wherein the RF measurements include a received signal strength indicator (RSSI) measured by the plurality of client devices and a transmission channel from connected or nearby Wi-Fi access points;

deriving, by one or more processors, the at least one optimal wireless channel for the Wi-Fi-access point by solving an optimization formulation by employing an artificial intelligence (AI) module integrated with a global continuous optimization algorithm, wherein the AI module utilizes the collected RF measurements and a plurality of measured variables;

wherein the AI module uses a machine learning algorithm to evaluate the optimization formulation in combinations of wireless channels to provide relaxation of discrete wireless channels considering an interference coefficient, wherein each interference coefficient comprises an extent of interference from a Wi-Fi access point transmitting on a respective channel; and

training, by one or more processors, the AI module based on interference coefficient values to relate the channels of the Wi-Fi access points to a corresponding coefficient for different combinations of the Wi-Fi access points.

2. The method according to claim 1 , wherein the global continuous optimization algorithm is at least one of a genetic algorithm and a simulated annealing algorithm.

3. The method according to claim 1 , wherein the plurality of measured variables include one of RSSI at the Wi-Fi access point, RSSI at a client device of the plurality of client devices, a number of Wi-Fi access points whose signals reach the client device, and a number of Wi-Fi access points whose signals reach the Wi-Fi access point.

4. The method according to claim 1 , wherein a rectified linear unit (ReLU) activation function in the machine learning algorithm provides an approximation of actual values in addition to providing a smooth function for the global continuous optimization algorithm.

5. The method according to claim 4 further comprising comprises, transforming, by one or more processors, the optimization formulation to change the at least one optimal wireless channel based on measurements, wherein the optimization formulation includes the trained AI module, wherein an AI function of the AI module can delivers the interference coefficient for decimal channels.

6. A system for assigning at least one optimal wireless channel to a Wi-Fi access point of a plurality of Wi-Fi access points using a cloud-based software defined network (SDN), the system comprising:

a memory;

a processor communicatively coupled to the memory, wherein the processor is configured to:

collect radio frequency (RF) measurements from a plurality of client devices in cloud, wherein the RF measurements include a received signal strength indicator (RSSI) measured by the plurality of client devices and a transmission channel from connected or nearby Wi-Fi access points;

derive the at least one optimal wireless channel for the Wi-Fi-access point by solving an optimization formulation by employing an artificial intelligence (AI) module integrated with a global continuous optimization algorithm, wherein the AI module utilizes the collected RF measurements and a plurality of measured variables;

wherein the AI module uses a machine learning algorithm to evaluate the optimization formulation in combinations of wireless channels to provide relaxation of discrete wireless channels considering an interference coefficient, wherein each interference coefficient comprises an extent of interference from a Wi-Fi access point transmitting on a respective channel; and

wherein the processor is configured to train the AI module based on interference coefficient values to relate the channels of the Wi-Fi access points to a corresponding coefficient for different combinations of the Wi-Fi access points.

7. The system according to claim 6 , wherein the global continuous optimization algorithm comprises a genetic algorithm or a simulated annealing algorithm.

8. The system according to claim 6 , wherein the plurality of measured variables include one of the RSSI at the Wi-Fi access point, the RSSI at a client device of the plurality of client devices, a number of Wi-Fi access points with signals reaching the client device, and a number of Wi-Fi access points with signals reaching the Wi-Fi access point.

9. The system according to claim 6 , wherein a rectified linear unit (ReLU) activation function in the machine learning algorithm provides an approximation of actual values in addition to providing a smooth function for the global continuous optimization algorithm.

10. The system according to claim 9 , wherein the processor is further configured to transform the optimization formulation, to change the at least one optimal wireless channel based on measurements, wherein the optimization formulation includes the trained AI module, wherein an AI function of the AI module delivers the interference coefficient for decimal channels.

11. A method for assigning at least one optimal wireless channel to a Wi-Fi access point of a plurality of Wi-Fi access points using a cloud-based software defined network (SDN), the method comprising:

collecting, by one or more processors, radio frequency (RF) measurements from a plurality of client devices in a cloud computing network, wherein the RF measurements include a received signal strength indicator (RSSI) measured by the plurality of client devices and a transmission channel from connected or nearby Wi-Fi access points;

deriving, by one or more processors, the at least one optimal wireless channel for the Wi-Fi-access point by solving an optimization formulation by employing an artificial intelligence (AI) module integrated with a global continuous optimization algorithm, wherein the AI module utilizes the collected RF measurements and a plurality of measured variables;

wherein the AI module uses a machine learning algorithm to evaluate the optimization formulation in combinations of wireless channels to provide relaxation of discrete wireless channels considering an interference coefficient, wherein each interference coefficient comprises an extent of interference from a Wi-Fi access point transmitting on a respective channel; and

wherein a rectified linear unit (ReLU) activation function in the machine learning algorithm provides an approximation of actual values in addition to providing a smooth function for the global continuous optimization algorithm.

12. The method according to claim 11 , wherein the global continuous optimization algorithm comprises a genetic algorithm or a simulated annealing algorithm.

13. The method according to claim 11 , wherein the plurality of measured variables include one of the RSSI at the Wi-Fi access point, the RSSI at a client device of the plurality of client devices, a number of Wi-Fi access points with respective signals reaching the client device, and a number of Wi-Fi access points with signals reaching the Wi-Fi access point.

14. The method of claim 11 , further comprising training, by one or more processors, the AI module based on interference coefficient values to relate the channels of the Wi-Fi access points to a corresponding coefficient for different combinations of Wi-Fi access points.

15. The method according to claim 14 , further comprising transforming, by one or more processors, the optimization formulation to change the at least one optimal wireless channel based on measurements, wherein the optimization formulation includes the trained AI module, wherein an AI function of the AI module delivers the interference coefficient for decimal channels.