R1 EP5G channel measurement

R1 EP5G channel measurement This dataset consists of measurements conducted in the reactor hall at KTH using the EP5G (Ericsson Private 5G) network. The measurements were conducted statically, i.e. the robot was kept stationary during each measurement and each measurement was done over a period of 5-7 seconds. Over 9000 measurements were collected, these included the robot's position, the network throughput, RTT; MCS and data layers or Rank Indicator since the EP5G utilizes MIMO technology. The measurements also included lower-level KPI such as SINR and RSSI, but throughput was of main interest during the campaign. Dataset Overview Environment: Reactor hall, KTH Royal Institute of Technology. Network: Ericsson Private 5G (EP5G). Reciever: Quectel RM500Q-GL 5G modem mounted on a TurtleBot equipped with a LiDAR sensor. Network Tests: Channel info, such as MCS, RI, and SINR using AT commands. Ping measurements to the edge server. iPerf3 throughput measurements. File details The recordings.zip file contains the raw data gathered during measurement. This includes maps obtained by using ROS and navigating with a TurtleBot3. The archive contains files of data obtained through iPerf3 and ping. As well as ROS bags which contain the robots position and channel info obtained through AT commands. The ROS bags are included in their raw form as .db3 files but the data also exists in the more readable .csv format. scripts.zip contains the scripts used to generate the results and curate the data. It also contains the scripts used to read and convert the .db3 bag files. The rest of the files, of which aggregated_data represents the highest level, contain the following: synced_data.csv contains lines where the AT, ping and iPerf measurements are synchronized based on time. filtered_data.csv contains the same data as synced_data but after filtering. (removing outliers and invalid data points) aggregated_data.csv averages out the results for each measurement position to obtain a more readable summary of the measurements. coverage.txt contains data from the rosbags in the desired format. This includes x and y positions, throughput, mcs etc. It is only included for the curious and is not necessary to look at. 0,0 is the bottom left corner of the hall. Known Limitations Due to the sparsity of the measurement environment and the range of the LiDAR, the position estimates were subject to drift and fluctuations and are not to be regarded as 100% accurate. Countermeasures, such as adding extra landmarks and manually moving the robot to areas where it's easier for it to localize before resuming measurements, were employed to mitigate these effects.