Streaming Media Quality of Experience Prediction for Network Slice Selection in 5G Networks

Abstract

Systems, methods and computer software are disclosed for predicting streaming media Quality of Experience (QoE) for network slice selection in 5G networks. In one embodiment, a method includes developing a mathematical model of QoE parameters at base station transmitters; providing feedback from the models to a Quality of Service (QoS) manager; and selecting, by the QoS manager based on the feedback from the models, a 5G network slice that meets QoS requirements for allocation to a user.

1 . A method for predicting streaming media Quality of Experience (QoE) for network slice selection in 5G networks, the method comprising:
 developing a mathematical model of QoE parameters at base station transmitters;   providing feedback from the models to a Quality of Service (QoS) manager; and   selecting, by the QoS manager based on the feedback from the models, a 5G network slice that meets QoS requirements for allocation to a user.   
     
     
         2 . The method of  claim 1  wherein providing feedback from the models to the QoS manager comprises communicating buffer underflow/overflow probability values. 
     
     
         3 . The method of  claim 1  wherein the selecting of the slice includes updating a bit rate of the content delivery. 
     
     
         4 . The method of  claim 1  further comprising evaluating a buffer underflow probability or a buffer overflow probability. 
     
     
         5 . The method of  claim 4  wherein evaluating a buffer underflow probability or a buffer overflow probability includes checking a change in queue length in a time slot allocated to a user. 
     
     
         6 . The method of  claim 4  wherein a buffer overflow event occurs when a maximum number of slots are occupied and an enqueue of a frame occurs. 
     
     
         7 . The method of  claim 4  wherein a buffer underflow event occurs when no slots in the buffer are occupied and a dequeue of a frame occurs. 
     
     
         8 . A non-transitory computer-readable medium containing instructions for predicting streaming media Quality of Experience (QoE) for network slice selection in 5G networks, which, when executed, cause a system to perform steps comprising:
 developing a mathematical model of QoE parameters at base station transmitters;   providing feedback from the models to a Quality of Service (QoS) manager; and   selecting, by the QoS manager based on the feedback from the models, a 5G network slice that meets QoS requirements for allocation to a user.   
     
     
         9 . The non-transitory computer-readable medium of  claim 8  further including instructions for providing feedback from the models to the QoS manager comprises communicating buffer underflow/overflow probability values. 
     
     
         10 . The non-transitory computer-readable medium of  claim 8  further including instructions wherein the selecting of the slice includes updating a bit rate of the content delivery. 
     
     
         11 . The non-transitory computer-readable medium of  claim 8  further including instructions for evaluating a buffer underflow probability or a buffer overflow probability. 
     
     
         12 . The non-transitory computer-readable medium of  claim 8  further including instructions wherein evaluating a buffer underflow probability or a buffer overflow probability includes checking a change in queue length in a time slot allocated to a user. 
     
     
         13 . The non-transitory computer-readable medium of  claim 11  further including instructions wherein a buffer overflow event occurs when a maximum number of slots are occupied and an enqueue of a frame occurs. 
     
     
         14 . The non-transitory computer-readable medium of  claim 11  further including instructions wherein a buffer underflow event occurs when no slots in the buffer are occupied and a dequeue of a frame occurs. 
     
     
         15 . A system for predicting streaming media Quality of Experience (QoE) for network slice selection in 5G networks, comprising:
 a plurality of base station transmitters; and   a Quality of Service (QoS) manager;   wherein a mathematical model of QoE parameters are developed at the base station transmitters, wherein the models provide feedback to the QoS manager, and wherein the QoS manager selects a 5G network slice that meets QoS requirements for allocation to a use based on the feedback from the models.   
     
     
         16 . The system of  claim 15  wherein the feedback provided from the models to the QoS manager comprises buffer underflow/overflow probability values. 
     
     
         17 . The system of  claim 15  wherein a buffer underflow probability or a buffer overflow probability is evaluated. 
     
     
         18 . The system of  claim 17  wherein a buffer underflow probability or a buffer overflow probability being evaluated includes checking a change in queue length in a time slot allocated to a user. 
     
     
         19 . The system of  claim 17  wherein a buffer overflow event occurs when a maximum number of slots are occupied and an enqueue of a frame occurs. 
     
     
         20 . The system of  claim 17  wherein a buffer underflow event occurs when no slots in the buffer are occupied and a dequeue of a frame occurs.