Link performance prediction technologies

Abstract

The present disclosure is related to Link Performance Predictions (LPPs), which are used in connection with management of radio communication links. The LPPs are predictions of future network behaviors/metrics (e.g., bandwidth, latency, capacity, coverage holes, and/or the like). The LPPs are communicated to network nodes, which allows the network nodes to make operational decisions for improved signaling/link resource utilization. The link performance analysis is divided into multiple layers that determine their own link performance metrics, which are then fused together to make an LPP. Each layer runs different algorithms and/or machine learning models, and provides respective results to an LPP layer/engine that fuses the results together to obtain the LPP. Other embodiments are described and/or claimed.

1 . An apparatus employed as a link performance prediction (LPP) service provider, the apparatus comprising:
 network interface circuitry arranged to communicatively couple the apparatus with one or more network access nodes (NANs), wherein individual NANs of the one or more NANs are arranged to provide network connectivity to one or more user equipment (UEs) via respective links between the individual NANs and the one or more UEs; and   processor circuitry connected to the network interface circuitry, wherein the processor circuitry is to: obtain an LPP service request from an LPP service consumer via the network interface circuitry, wherein the LPP service request is a request for LPP services from the LPP service provider, determine an LPP for a link based on a set of predicted performance metrics obtained from a set of performance prediction layers (PPLs), and send an LPP notification to the LPP service consumer via the network interface circuitry, wherein the LPP notification includes the determined LPP.   
     
     
         2 . The apparatus of  claim 1 , wherein the processor circuitry is to determine the LLP based on a combination of the one or more predicted performance metrics. 
     
     
         3 . The apparatus of  claim 1 , wherein the LPP service request is a request message, the LPP notification is a reply message, and the processor circuitry is to: determine the LPP based at least in part on information contained in the request message; and generate the LPP notification in response to receipt of the request message. 
     
     
         4 . The apparatus of  claim 1 , wherein the LPP service request is a subscribe message, the LPP notification is a notify message, and the processor circuitry is to: add an identifier associated with the LPP service consumer and information contained in the subscribe message to a subscriber list; detect a trigger to determine the LPP for the link; and in response to detecting the trigger, generate the LPP notification and send the notify message to the LPP service consumer. 
     
     
         5 . The apparatus of  claim 1 , wherein the LPP notification is to cause the LPP service consumer to perform one or more operational decisions for link resource utilization based on the LPP notification. 
     
     
         6 . The apparatus of  claim 1 , wherein the LPP notification includes a time field to indicate a timestamp of the corresponding LPP, a type field to indicate an LPP type of the corresponding LPP, a value field to indicate a value of the LPP, and a probability field to indicate a likelihood that the corresponding LPP comes true or an estimated standard deviation of the corresponding LPP, wherein the LPP type is one of bandwidth, latency, jitter, round trip time (RTT), number of interrupts, out-of-order delivery of data packets, transmission power, bit error rate, bit error ratio (BER), packet loss rate, a packet reception rate (PRR), a signal-to-noise ratio (SNR), a signal-to-noise and interference ratio (SINR), a signal-plus-noise-plus-distortion to noise-plus-distortion (SINAD) ratio, a peak-to-average power ratio (PAPR), a Block Error Rate (BLER), a Reference Signal Received Power (RSRP), a Reference Signal Received Quality (RSRQ), a Received Signal Strength Indicator (RSSI), a channel interference measurement, a thermal noise power measurement, a received interference power measurement, network or cell load, a recommended transmission power. 
     
     
         7 . The apparatus of  claim 1 , wherein, to determine the LPP, the processor circuitry is to: combine the one or more predicted performance metrics from individual PPLs of the set of PPLs with one or more other predicted performance metrics from at least one other prediction layer of the set of PPLs. 
     
     
         8 . The apparatus of  claim 1 , wherein a subset of PPLs of the set of PPLs is arranged to use a corresponding machine learning (ML) algorithm to generate an ML model for corresponding NANs of the one or more NANs, and each PPL in the subset of PPLs is arranged to use the corresponding ML model for the corresponding NANs to generate respective predicted performance metrics. 
     
     
         9 . The apparatus of  claim 8 , wherein at least one PPL of the subset of PPLs is arranged to use an ML algorithm that is different than ML algorithms used by other PPLs of the subset of PPLs. 
     
     
         10 . The apparatus of  claim 9 , wherein the processor circuitry is arranged to operate individual prediction layers in respective virtual machines (VMs) or virtualization containers. 
     
     
         11 . The apparatus of  claim 10 , wherein the apparatus is a System-On-Chip (SoC) or a Multi-Chip Package (MCP) disposed in a server computing system, the server computing system is to operate as an application server, an edge compute node, a web server, and one or more cloud compute nodes of a cloud computing service. 
     
     
         12 . The apparatus of  claim 1 , wherein the individual prediction layers are to be operated by respective edge compute nodes, and the network interface circuitry is arranged to receive the predicted performance metrics from the respective edge compute nodes.