The paper is organized into eight broad sections, Introduction being the first one. The aim of this whitepaper is twofold.

Firstly, it highlights the specifications of the 5G technology, identifies the potential use cases targeting to different industries, deliberates those areas that will require likely investment for 5G networks. Also, it provide you with an overview on 5G NR and describes how 5G solutions will use radio waves to wirelessly transmit data faster and with less latency than previous radio access technology specifications.

Secondly it presents in detail, the use of Machine Learning (ML) technique to predict the performance of 5G access networks. In this context the AI/ML based solution pertaining to spatial variations of network quality based on power and throughput information experienced by different users at different geo locations is presented. As a rule of thumb for ML algorithms, they need data for training, which includes features and labels. In this case, a production 5G dataset along with captured metrics composed of client-side key performance indicators is explored and illustrates the variance in power & throughput.

It summarized the progress and key findings to measure the service performance and user perceived quality over the 5G experimental networks, in order to prove and validate that cognitive network measurement is likely to pave the way as a key enabler of 5G performance expectations. Download whitepaper to continue reading.