Estimation of network load and downlink throughput using RF scanner data for LTE networks

This paper provides a methodology of estimating downlink throughput using the drive test data collected by an RF scanner device rather than obtaining the downlink throughput via FTP downloading from UE. The throughput test using the UE's FTP application to perform the throughput test has a merit of presenting the packet data speed from the end user perspective however, it also has significant shortcomings such as multiple RF layers cannot be tested simultaneously as UE can have only a single radio connection to eNB and no analytic performance benchmarking between different LTE networks is possible. The proposed algorithm utilizes the simple measurement quantities collected by an RF scanner in order to estimate the LTE network's traffic load and the available PRBs without relying on network parameters or site information. Furthermore, the proposed algorithm calculates the potential downlink throughput at each location at which the RF scanner measured RSRP, RSRQ and SINR assuming that the available PRBs of the LTE cell are fully utilized at the location. Since the RF scanner is capable of measuring the multiple RF layers simultaneously, the proposed algorithm enables whole LTE networks to be benchmarked comprehensively. In addition, the proposed algorithm helps in analytically determining the root causes of the superiority or inferiority of a LTE network whether it is due to traffic load, system bandwidth, or spectrum efficiency.