Systems and methods for automatic application-level network slicing over private cellular networks

1. A method comprising:

detecting, by a device, that a communication session in a private cellular wireless network is associated with an application;

determining, by the device, that data units associated with the communication session match a network slicing rule associated with the private cellular wireless network;

classifying, by the device, the communication session to a network slice based on the network slicing rule; and

assigning, by the device, the data units associated with the communication session to the network slice.

2. The method of claim 1 , wherein the network slicing rule assigns different applications of a same application type to different network slices.

3. The method of claim 1 , further comprising:

receiving the network slicing rule from an orchestration device, wherein the network slicing rule assigns a particular application to a particular network slice; and

storing the network slicing rule in a database associated with a traffic classifier.

4. The method of claim 1 , wherein the network slicing rule associates a virtual local area network (VLAN) identifier with the application and wherein detecting that the communication session in the private cellular wireless network is associated with the application includes:

detecting that data units associated with the communication session include the VLAN identifier.

5. The method of claim 1 , wherein the network slicing rule associates an Internet Protocol (IP) address and a port number with the application and wherein detecting that the communication session in the private cellular wireless network is associated with the application includes:

detecting that data units associated with the communication session are associated with the IP address and the port number.

6. The method of claim 1 , wherein the network slicing rule associates a device group identifier with the application and wherein detecting that the communication session in the private cellular wireless network is associated with the application includes:

detecting that data units associated with the communication session include the device group identifier.

7. The method of claim 1 , wherein detecting that the communication session in the private cellular wireless network is associated with the application includes:

determining that the data units associated with the communication session satisfy a data unit parameter criterion.

8. The method of claim 7 , wherein determining that the data units associated with the communication session satisfy the data unit parameter criterion includes at least one of:

determining that the data units associated with the communication session are associated with a pattern of one uplink data unit to multiple downlink data units,

determining that a payload associated with uplink data units is greater than an uplink payload threshold, or

determining that a cumulative payload associated with downlink data units is greater than a downlink payload threshold.

9. The method of claim 1 , wherein detecting that the communication session in the private cellular wireless network is associated with the application includes:

determining that the data units associated with the communication session correspond to video data units.

10. The method of claim 9 , wherein determining that the data units associated with the communication session correspond to video data units includes at least one of:

determining that throughput associated with the data units is greater than a throughput threshold,

determining that an average downlink payload associated with the data units is greater than a payload threshold, or

determining that a data unit count associated with the data units is greater than a data unit count threshold.

11. The method of claim 1 , wherein assigning the data units associated with the communication session to the network slice includes:

assigning the data units associated with the communication session to a particular class of service.

12. The method of claim 1 , wherein the device includes:

a packet data network gateway (PGW),

an eNodeB,

a User Plane Function (UPF), or

a gNodeB.

13. A device comprising:

a processor configured to:

detect that a communication session in a private cellular wireless network is associated with an application;

determine that data units associated with the communication session match a network slicing rule associated with the private cellular wireless network;

classify the communication session to a network slice based on the network slicing rule; and

assign the data units associated with the communication session to the network slice.

14. The device of claim 13 , wherein the network slicing rule assigns different applications of a same application type to different network slices.

15. The device of claim 13 , wherein the processor is further configured to:

receive the network slicing rule from an orchestration device, wherein the network slicing rule assigns a particular application to a particular network slice; and

store the network slicing rule in a database associated with a traffic classifier.

16. The device of claim 13 , wherein, when detecting that the communication session in the private cellular wireless network is associated with the application, the processor is further configured to at least one of:

detect that data units associated with the communication session include a particular virtual local area network (VLAN) identifier;

detect that data units associated with the communication session are associated with a particular Internet Protocol (IP) address and a particular port number; or

detect that data units associated with the communication session include a particular device group identifier.

17. The device of claim 13 , wherein, when detecting that the communication session in the private cellular wireless network is associated with the application, the processor is further configured to at least one of:

determine that the data units associated with the communication session are associated with a pattern of one uplink data unit to multiple downlink data units,

determine that a payload associated with uplink data units is greater than an uplink payload threshold, or

determine that a cumulative payload associated with downlink data units is greater than a downlink payload threshold.

18. The device of claim 13 , wherein, when detecting that the communication session in the private cellular wireless network is associated with the application, the processor is further configured to at least one of:

determine that the data units associated with the communication session correspond to video data units.

19. The device of claim 13 , wherein, when assigning the data units associated with the communication session to the network slice, the processor is further configured to:

assign the data units associated with the communication session to a particular class of service.

20. A non-transitory computer-readable memory device storing instructions executable one a processor, the non-transitory computer-readable memory device comprising:

one or more instructions to detect that a communication session in a private cellular wireless network is associated with an application;

one or more instructions to determine that data units associated with the communication session match a network slicing rule associated with the private cellular wireless network;

one or more instructions to classify the communication session to a network slice based on the network slicing rule; and

one or more instructions to assign the data units associated with the communication session to the network slice.