6G-SENSES 5G NR Indoor Measurements at 18 dBm and 24 dBm RU Transmit Power at OTE

This dataset contains indoor 5G New Radio performance and coverage measurements collected during the 6G-SENSES Proof of Concept 3 activities at the OTE laboratory testbed in Athens, Greece. The experimental platform comprised an Accelleran O-RAN gNB connected to the ATHONET 5G Core Network deployed in a Proxmox-based virtualised environment. The measurements were collected to characterise network performance and radio coverage under different radio-unit transmit-power configurations and to support the development, training and evaluation of Network Digital Twin models within the 6G-SENSES project. The deposited files correspond to two radio-unit transmit-power settings: DataSet6G-SENSES-PoC3_14dBm.json: measurements collected with the RU transmit power configured to 14 dBm. DataSet6G-SENSES-PoC3_24dBm.json: measurements collected with the RU transmit power configured to 24 dBm. A third experimental configuration at 19 dBm is documented in the 6G-SENSES D5.3 deliverable but is not included in the present dataset release. The 5G NR radio configuration used an operating frequency around 3.5 GHz and a channel bandwidth of 100 MHz for both downlink and uplink. The gNB configuration included NR-ARFCN 636666, SSB frequency 634080 and a configured physical cell identifier of 3. Measurements were collected using the OTE-developed COSMOTools performance and coverage monitoring platform. The tool recorded network and user-equipment measurements as a function of time and geographic location. The available parameters include: GPS latitude and longitude; UDP downlink and uplink throughput in Mbit/s; latency in milliseconds, where available; packet jitter; lost and transmitted packet counts; Reference Signal Received Power; Reference Signal Received Quality; reported signal-to-noise ratio; network, cell and mobility-related identifiers. The throughput tests were generated using iPerf3 with an offered UDP load of 1 Gbit/s. The 14 dBm campaign used 10-second iPerf3 tests, while the 24 dBm campaign used 5-second tests. The nominal maximum data rate of the experimental system was approximately 500 Mbit/s. Throughput values may therefore be normalised for comparative analysis using: normalised throughput = measured throughput / 500 Mbit/s. Because the offered UDP traffic load exceeded the nominal system data rate, packet loss is an expected result of the saturation tests and should not, by itself, be interpreted as a data-quality defect. The files contain per-interval iPerf3 observations and aggregate observations. Aggregate records are identified using the isSummary field. Users should avoid combining interval and summary records in the same statistical aggregation unless this distinction is explicitly accounted for. The 14 dBm dataset contains latency, downlink throughput and uplink throughput measurements. The 24 dBm dataset contains downlink and uplink throughput measurements but does not contain separate latency observations. Consequently, direct comparisons between the two campaigns should be restricted to common variables and should account for the different iPerf3 test durations. The radio unit was installed at a height of approximately 2 m. Measurements were collected within an indoor area of approximately 20 m × 10 m. The first measurement in each campaign represents a local reference position defined as (0,0), located next to the radio unit. Relative measurement locations can be estimated from the GPS coordinates included in the files. This dataset supports research on radio coverage modelling, network-performance prediction, energy-aware RAN configuration and Network Digital Twin training and evaluation. The work was performed within 6G-SENSES — “SEamless integratioN of efficient 6G wireleSs tEchnologies for communication and Sensing,” funded by the European Union’s Smart Networks and Services Joint Undertaking under Horizon Europe, Grant Agreement No. 101139282.