Intelligent connectivity and data usage management for mobile devices in a converged network
Various embodiments comprise systems, methods, architectures, mechanisms and apparatus for managing user equipment (UE) communications in a converged network by causing such UE to automatically select a “best” wireless network (cellular RAN/WiFi) without any user notification/interaction to the user of the UE, such as via the device user interface (UI). The best wireless network may be determined using contemporaneous QoS related measurements, or predicted QoS measurements such as provided by a RAN node or channel QoS model generated via machine learning techniques, which may also be used to determine a type of QoS needed for each application or service commonly invoked at the UE so as to automatically provide the best possible user experience without any user intervention.
1 . A method for generating quality of service (QoS) predictive data, the method comprising:
collecting QoS-related data;
executing a time series prediction algorithm to process at least the collected QoS data to generate QoS-related predictions or network element models, the generated QoS-related predictions or network element models enabling determination of a plurality of expected nominal QoS respectively associated with a plurality of computerized apparatus of respective different technology types, the plurality of computerized apparatus of respective different technology types comprising mobile network node apparatus and Wi-Fi access point (WAP) apparatus; and
utilizing the plurality of expected nominal QoS to select, for use by one or more dual radio access network (RAN)-capable user equipment (UE) within a converged network, one of at least one of the mobile network node apparatus or the Wi-Fi access point (WAP) apparatus for provision of one or more network services.
2 . The method of claim 1 , wherein the executing of the time series prediction algorithm to process at least the collected QoS data to generate the QoS-related predictions or network element models comprises executing of the time series prediction algorithm to process at least the collected QoS data to generate expected over-utilization or under-utilization of one or more router interface queuing structures.
3 . The method of claim 1 , wherein the executing of the time series prediction algorithm to process at least the collected QoS data to generate the QoS-related predictions or network element models comprises executing of the time series prediction algorithm to process at least the collected QoS data to generate expected outages at various system, channel, service, link or other portions of the converged network.
4 . The method of claim 1 , wherein the executing of the time series prediction algorithm to process at least the collected QoS data to generate the QoS-related predictions or network element models comprises executing of the time series prediction algorithm to process at least the collected QoS data to generate expected conditions which indicate a possibility of negatively impacting QoS levels of services provided to customers via a specific one of the plurality of mobile network node apparatus or WAP apparatus.
5 . The method of claim 1 , further comprising collecting performance data associated with one or more computer program applications of the one or more dual RAN-capable UE;
wherein the executing of the time series prediction algorithm to process at least the collected QoS data to generate the QoS-related predictions or network element models comprises executing of the time series prediction algorithm to process least the collected QoS data and the performance data to generate the QoS-related predictions or network element models.
6 . The method of claim 5 , wherein the collecting of the performance data comprises collecting data indicative of historical QoS associated with minimum or sufficient performance of the one or more computer program applications.
7 . The method of claim 1 , wherein the executing of the time series prediction algorithm to process at least the collected QoS data to generate the QoS-related predictions or network element models comprises executing of the time series prediction algorithm to process at least the collected QoS data and network topology data to generate the QoS-related predictions or network element models.
8 . The method of claim 1 , further comprising:
updating the QoS-related data; and
executing the time series prediction algorithm to process at least the updated QoS-related data to generate updated QoS-related predictions or network element models.
9 . The method of claim 1 , wherein the execution of the time series prediction algorithm comprises execution of the time series prediction algorithm to generate expected outages at one or more system, channel, service, link or other portions of the converged network.
10 . A quality of service (QoS) manager apparatus configured for management of network connectivity of one or more dual radio access network (RAN)-capable user equipment (UE) within a converged network, the QoS manager apparatus comprising:
communications interface apparatus;
processor apparatus in data communication with the communications interface apparatus; and
storage apparatus in data communication with the processor apparatus and comprising a storage medium, the storage medium comprising at least one computer program configured to, when executed by the processor apparatus, cause the QoS manager apparatus to:
collect QoS-related data;
process at least the collected QoS data to determine a plurality of expected nominal QoS levels respectively associated with a plurality of computerized apparatuses of respective different technology types, the processing of at least the collected QoS data to determine the plurality of nominal QoS comprising execution of a time series prediction algorithm to generate one or more QoS-related predictions or network element models, the one or more QoS-related predictions or network element models enabling the determination of the plurality of expected nominal QoS levels; and
utilize at least the plurality of expected nominal QoS levels to select, for use by the one or more dual RAN-capable UE, one of the plurality of computerized apparatuses of a certain technology type of the respective different technology types for provision of one or more network services.
11 . The QoS manager apparatus of claim 10 , wherein the plurality of computerized apparatuses of the respective different technology types comprises at least one mobile network node apparatus and at least one Wi-Fi access point (WAP) apparatus.
12 . The QoS manager apparatus of claim 11 , wherein the determination of the plurality of nominal QoS respectively associated with the plurality of computerized apparatuses of the respective different technology types comprises determination of the plurality of nominal QoS using contemporaneously measured QoS indicators of each of a plurality of mobile network node and WAP channels.
13 . The QoS manager apparatus of claim 10 , wherein the one or more QoS-related predictions or network element models comprise models generated by a machine learning module based at least on historical QoS data.
14 . The QoS manager apparatus of claim 13 , wherein the machine learning module is configured to execute the time series prediction algorithm to provide thereby a continually updated QoS model of the plurality of computerized apparatuses of the respective different technology types.
15 . The QoS manager apparatus of claim 13 , wherein the historical QoS data comprises data indicative of sufficiency of QoS for application performance.
16 . Computer readable apparatus comprising a non-transitory storage medium, the non-transitory storage medium comprising at least one computer program having a plurality of instructions, the plurality of instructions configured to, when executed on a processing apparatus, cause a quality of service (QoS) manager apparatus to:
collect QoS-related data;
process at least the collected QoS data to determine a plurality of expected nominal QoS-related values respectively associated with computerized apparatus of respective different technology types, the computerized apparatus of respective different technology types comprising (i) one or more mobile network node apparatus and (ii) one or more Wi-Fi access point (WAP) apparatus, the processing of at least the collected QoS data to determine the plurality of expected nominal QoS-related values comprising execution of a time series prediction algorithm to generate one or more QoS-related predictions or network element models, the one or more QoS-related predictions or network element models enabling the determination of the plurality of expected nominal QoS-related values; and
utilize the plurality of expected nominal QoS-related values to select, for use by one or more dual radio access network (RAN)-capable user equipment (UE) within a converged network, one of (i) the one or more mobile network node apparatus or (ii) the one or more Wi-Fi access point (WAP) apparatus for provision of one or more network services to at least one computerized client device.
17 . The computer readable apparatus of claim 16 , wherein the execution of the time series prediction algorithm comprises execution of the time series prediction algorithm to generate expected over-utilization or under-utilization of one or more router interface queuing structures.
18 . The computer readable apparatus of claim 16 , wherein the execution of the time series prediction algorithm comprises execution of the time series prediction algorithm to generate expected conditions which indicate a possibility of negatively impacting QoS levels of services provided to customers via a specific one of the one or more mobile network node apparatus or the one or more WAP apparatus.
19 . The computer readable apparatus of claim 16 , wherein:
the plurality of instructions are further configured to, when executed on the processing apparatus, cause the QoS manager apparatus to:
collect performance data associated with one or more computer program applications of the one or more dual RAN-capable UE, the performance data comprising data indicative of historical QoS associated with minimum or sufficient performance of the one or more computer program applications; and
the execution of the time series prediction algorithm comprises execution of the time series prediction algorithm to process least the collected QoS data and the performance data to generate the QoS-related predictions or network element models.
20 . The computer readable apparatus of claim 16 , wherein the execution of the time series prediction algorithm comprises execution of the time series prediction algorithm to process at least the collected QoS data and network topology data to generate the QoS-related predictions or network element models.