Quality of service management for 5G networks

1 . A computing device configured to manage Quality of Service (Qos), comprising:

processing circuitry; and

a memory device including instructions embodied thereon, wherein the instructions, which when executed by the processing circuitry, configure the processing circuitry to cause operations that:

identify QoS characteristics for data flows of a user equipment (UE), the data flows performed via multiple access networks that include a 3rd Generation Partnership Project (3GPP) access network;

map the QoS characteristics to network functions of at least one of the multiple access networks;

control the network functions of the at least one of the multiple access networks, based on the QoS characteristics, wherein the network functions are implemented at respective resources located within at least one of the multiple access networks;

register a network exposure function of the 3GPP access network to a network service provided to the UE;

register the QoS characteristics to the network service;

configure the network exposure function to meet QoS requirements for the network service based on the QoS characteristics; and

monitor traffic associated with the network service in the 3GPP access network and update a configuration of the network exposure function to maintain the QoS requirements based on the monitored traffic and the QoS characteristics.

2 . The computing device of claim 1 , wherein the multiple access networks include a non-3GPP access network, and wherein the network functions are controlled using Access Traffic Steering, Switching and Splitting (ATSSS) functionality in the 3GPP access network.

3 . The computing device of claim 2 , wherein the instructions further configure the processing circuitry to cause operations that:

distribute data using both of the 3GPP access network and the non-3GPP access network; and

in response to a fault in one of the 3GPP access network and the non-3GPP access network or QoS requirements for at least one data flow of the data flows not being met, switch traffic from the one of the 3GPP access network and the non-3GPP access network to an other of the 3GPP access network and the non-3GPP access network, while maintaining the QoS requirements on the other of the 3GPP access network and the non-3GPP access network.

4 . The computing device of claim 2 , wherein the instructions further configure the processing circuitry to cause operations that:

control routing of data between the 3GPP access network and the non-3GPP access network based on radio access network measurements or a QoS tag used in a downlink flow;

negotiate QoS requirements and traffic priority in the non-3GPP access network; and

adjust the routing of data between the 3GPP access network and the non-3GPP access network, based on the negotiated QoS requirements and traffic priority in the non-3GPP access network.

5 . The computing device of claim 4 , wherein the data is routed to the non-3GPP access network in response to an acceptance of the negotiated QoS requirements in the non-3GPP access network, and wherein the data is routed to the 3GPP access network in response to rejection of the negotiated QoS requirements in the non-3GPP access network.

6 . The computing device of claim 1 , wherein the instructions configure the processing circuitry to cause operations that:

control an edge processing unit associated with the computing device to meet the Qos requirements, wherein the control includes reconfiguration of a dedicated hardware packet queue of the edge processing unit based on the QoS characteristics, and wherein the edge processing unit comprises a network interface controller (NIC), smart NIC, infrastructure processing unit (IPU), or data processing unit (DPU).

7 . The computing device of claim 6 , wherein the dedicated hardware packet queue is an application device queue, wherein the instructions further configure the processing circuitry to cause operations that apply an ingress application data queue policy and an egress application data queue policy, and wherein the application device queue is mapped to data flows to meet the QoS requirements.

8 . The computing device of claim 7 , wherein the QoS requirements relate to at least one of queue size of the application device queue, dequeue rate of the application device queue, or traffic limits applied to the data flows via the application device queue.

9 . The computing device of claim 6 , wherein the instructions further configure the processing circuitry to cause operations that:

correlate properties of the dedicated hardware packet queue with platform resources of a network node at which the edge processing unit operates, the platform resources comprising at least one of a central processing unit resource, a cache resource, a memory resource, or an input/output resource; and

determine, using one or more resource learning agents operating at least one model, recommended configurations for the dedicated hardware packet queue and the platform resources to meet the QoS requirements.

10 . The computing device of claim 1 , wherein the instructions further configure the processing circuitry to cause operations that:

identify, using resource learning agents, changes to a configuration of the network exposure function to maintain the QoS requirements, and wherein the resource learning agents operate at least one model to identify characteristics of the respective resources, the respective resources including at least one of: a software application, an operating system, a firmware configuration, or a hardware configuration.

11 . The computing device of claim 1 , wherein the instructions further configure the processing circuitry to cause operations that:

monitor real-time link status of each of the multiple access networks; and

dynamically adjust a traffic splitting weight for distributing the data flows across the multiple access networks based on the monitored real-time link status, wherein the traffic splitting weight is adjusted without waiting for end-to-end measurement results.

12 . The computing device of claim 1 , wherein the instructions further configure the processing circuitry to cause operations that:

tag individual packets of the data flows with at least one QoS indicator prior to transmission over at least one of the multiple access networks, wherein the at least one QoS indicator comprises at least one of a Differentiated Services Code Point (DSCP) value, a 5G QoS Indicator value, or a wireless local area network (WLAN) user priority value.

13 . The computing device of claim 1 , wherein the instructions further configure the processing circuitry to cause operations that:

monitor bandwidth used by the network service provided to the UE; and

based on the monitored bandwidth, initiate a bandwidth-based update to the configuration of the network exposure function via an application programming interface to the network exposure function, wherein the bandwidth-based update adaptively configures a fifth generation (5G) network bandwidth allocation for the network service based on the monitored bandwidth.

14 . A method of managing Quality of Service (QoS), comprising:

identifying QoS characteristics for data flows of a user equipment (UE), the data flows performed via multiple access network that include a 3rd Generation Partnership Project (3GPP) access network and a non-3GPP access networks;

mapping the QoS characteristics to network functions of at least one of the multiple access networks;

controlling the network functions of the at least one of the multiple access networks using Access Traffic Steering, Switching and Splitting (ATSSS) functionality in the 3GPP access network and, based on the QoS characteristics, wherein the network functions are implemented at respective resources located within at least one of the multiple access networks;

controlling routing of data between the 3GPP access network and the non-3GPP access network based on radio access network measurements or a QoS tag used in a downlink flow;

negotiating QoS requirements and traffic priority in the non-3GPP access network; and

adjusting the routing of data between the 3GPP access network and the non-3GPP access network, based on the negotiated QoS requirements and traffic priority in the non-3GPP access network.

15 . The method of claim 14 , further comprising:

distributing data using both of the 3GPP access network and the non-3GPP access network; and

in response to a fault in one of the 3GPP access network and the non-3GPP access network or QoS requirements for at least one data flow of the data flows not being met, switching traffic from the one of the 3GPP access network and the non-3GPP access network to an other of the 3GPP access network and the non-3GPP access network, while maintaining the QoS requirements on the other of the 3GPP access network and the non-3GPP access network.

16 . The method of claim 14 , wherein the data is routed to the non-3GPP access network in response to an acceptance of the negotiated QoS requirements in the non-3GPP access network, and wherein the data is routed to the 3GPP access network in response to rejection of the negotiated QoS requirements in the non-3GPP access network.

17 . The method of claim 14 , wherein the method further comprises:

registering a network exposure function of the 3GPP access network to a network service provided to the UE;

registering the QoS characteristics to the network service;

configuring the network exposure function to meet QoS service requirements for the network service based on the QoS characteristics; and

monitoring traffic associated with the network service in the 3GPP access network and updating a configuration of the network exposure function to maintain the QoS service requirements based on the monitored traffic and the QoS characteristics.

18 . The method of claim 17 , wherein the method further comprises identifying, using resource learning agents, changes to a configuration of the network exposure function to maintain the QoS service requirements, and wherein the resource learning agents operate at least one model to identify characteristics of the respective resources, the respective resources including at least one of: a software application, an operating system, a firmware configuration, or a hardware configuration.

19 . The method of claim 14 , further comprising:

controlling an edge processing unit to meet the QoS requirements, wherein the control includes reconfiguration of a dedicated hardware packet queue of the edge processing unit based on the QoS characteristics, and wherein the edge processing unit comprises a network interface controller (NIC), smart NIC, infrastructure processing unit (IPU), or data processing unit (DPU).

20 . The method of claim 19 , wherein the dedicated hardware packet queue is an application device queue, wherein the method further comprises applying an ingress application data queue policy and an egress application data queue policy, and wherein the application device queue is mapped to data flows to meet the QoS requirements.

21 . The method of claim 20 , wherein the QoS requirements relate to at least one of queue size, dequeue rate, or traffic limits.