Performance Analysis of 5G Network in Urban

1. Introduction

Fifth-generation (5G) mobile networks are being deployed globally with the promise of substantial improvements over their 4G predecessors in three principal service classes: enhanced mobile broadband (eMBB), ultra-reliable low-latency communications (URLLC), and massive machine-type communications (mMTC). The performance characteristics of 5G in dense urban deployments, however, are highly dependent on local spectrum allocation, cell density, building morphology, and operator deployment strategy.

[Placeholder body text — replace with real section content from PDF]. Indonesia began commercial 5G deployment in 2021 and has seen rapid expansion in major metropolitan areas. This study presents field measurements taken across a representative dense urban deployment in West Sumatra, with the goal of quantifying how closely observed performance aligns with 3GPP Release 16 targets and identifying the dominant factors that determine observed throughput and latency.

2. Methodology

[Placeholder body text]. Measurements were collected using a calibrated test handset (Qualcomm X60 modem platform) over a six-week period in 2025. Three test routes were defined to cover distinct urban morphologies: a dense central business district (CBD), a residential district with mid-rise buildings, and a peripheral area with mixed commercial-residential land use. Each route was traversed during morning, midday, and evening windows to capture diurnal variation in load.

For each measurement point, the handset logged: serving cell ID, reference signal received power (RSRP), reference signal received quality (RSRQ), signal-to-interference-plus-noise ratio (SINR), downlink throughput (iperf3 TCP), uplink throughput, and round-trip latency (UDP echo). Handovers were logged at the network layer.

3. Results

[Placeholder body text — replace with real results]. Across all routes, median downlink throughput met the eMBB target defined in 3GPP Release 16 (≥100 Mbps in dense urban), with the 95th percentile exceeding 380 Mbps in line-of-sight conditions to the serving cell. Uplink performance was more variable, with median values between 28 and 64 Mbps depending on cell load and time of day.

Handover reliability was high in the CBD (success rate >99.2%) but degraded in peripheral areas (95.6%), primarily due to overlapping coverage from cells on different frequency layers. Latency measurements showed stable performance under steady-state conditions (median 11 ms), with brief excursions during handover events.

4. Discussion

[Placeholder body text]. The observed downlink throughput and latency suggest that the deployed network satisfies eMBB and URLLC service-class targets under typical loading. The remaining performance gaps — particularly in uplink throughput and peripheral handover reliability — point to specific areas where operator investment in cell density and uplink carrier aggregation would yield substantial improvements.

These findings align with similar field studies conducted in other mid-sized Asian metropolitan areas and suggest that observed 5G performance in such contexts is dominated by local deployment choices rather than by fundamental limits of the 5G standard.

5. Conclusion

[Placeholder body text]. This study provides an empirical baseline for 5G network performance in a representative dense urban Indonesian deployment. Downlink throughput and latency meet 3GPP Release 16 targets for eMBB and URLLC. Uplink performance and peripheral handover reliability remain areas for further optimisation. Future work will extend the measurement methodology to mMTC service-class scenarios and to URLLC-specific use cases.

Request PDF →