Method and apparatus for load control of an enterprise network on a campus based upon observations of busy times and service type

A method of monitoring, managing, and controlling an enterprise wireless communication network, the enterprise network including a Radio Access Network (RAN) that includes a plurality of Base Stations/Access Points (BS/APs) located at a campus location communicating with a plurality of User Equipment devices (UEs), the method comprising:

• monitoring RAN operations over a period of time to provide usage and service data for the BS/APs;
• data mining the usage and service data to detect usage and service patterns;
• and
• monitoring network operations, and re-allocating network resources proactively in response to the usage and service patterns to avoid congestion and provide appropriate service to the plurality of UEs,
• wherein:
• the data mining step further comprises identifying congestion hotspots based on the day-of-the-week and time-of-day, and including using machine learning to predict heatmaps for the individual BS/APs;
• and
• using artificial intelligence planning to proactively re-allocate resources and balance the number of users with the level of service supported with the network resources available.

A method of monitoring, managing, and controlling an enterprise wireless communication network, the enterprise network including a Radio Access Network (RAN) that includes a plurality of Base Stations/Access Points (BS/APs) located at a campus location communicating with a plurality of User Equipment devices (UEs), the method comprising:

• monitoring RAN operations over a period of time to provide usage and service data for the BS/APs;
• data mining the usage and service data to detect usage and service patterns;
• and
• monitoring network operations, and re-allocating network resources proactively in response to the usage and service patterns to avoid congestion and provide appropriate service to the plurality of UEs;
• using machine learning to recognize demand patterns for services;
• recognizing VoIP packets using deep learning;
• and
• adjusting the QoS parameters and bitrate resource allocation based on the recognized and predicted demand for a given service.

A method of proactively controlling the wireless communication load in an enterprise network situated in a campus location that includes a plurality of BS/APs wirelessly communicating with a plurality of UEs on the campus, comprising the steps of:

• monitoring the communications between the BS/APs and the UEs to collect channel occupancy data on each of the BS/APs;
• data mining the channel occupancy data to identify congestion hotspots as a function of time;
• responsive to the congestion hotspots and the channel occupancy data, data mining to create heatmaps of BS/AP usage over time and make congestion predictions as a function of time;
• responsive to predicted congestion, use artificial intelligence (AI) planning to determine a plan for the network to reduce the predicted congestion by at least one of re-allocating resources, refusing admission of new UEs, and offloading UEs to other networks;
• and
• proactively controlling the BS/APs responsive to said AI planning, prior to the predicted congestion, to reduce congestion by at least one of re-allocating resources, refusing admission of new UEs, and offloading UEs to other networks.

A method of proactively controlling service load in a wireless communication enterprise network situated in a campus location that includes a plurality of BS/APs wirelessly communicating with a plurality of UEs on the campus, the UEs making service requests, comprising the steps of:

• monitoring service type requests and collecting service type demand data;
• data mining the service type demand data to learn service demand patterns;
• responsive to the service demand patterns, using artificial intelligence planning to develop policies to meet the service demand patterns and build preferences to support the policies and service demand patterns;
• and
• responsive to the service demand patterns and policies, monitor service type requests, predict service demand increases, and take action proactively to de-prioritize certain services.