Rostand A. K. Fezeu,Jason Carpenter,Claudio Fiandrino,Eman Ramadan,Wei Ye,Joerg Widmer,Feng Qian,Zhi-Li Zhang
Abstract:Fifth Generation (5G) mobile networks mark a significant shift from previous generations of networks. By introducing a flexible design, 5G networks support highly diverse application requirements. Currently, the landscape of previous measurement studies does not shed light on 5G network configuration and the inherent implications to application performance. In this paper, we precisely fill this gap and report our in-depth multi-country measurement study on 5G deployed at mid-bands. This is the common playground for U.S. and European carriers. Our findings reveal key aspects on how carriers configure their network, including spectrum utilization, frame configuration, resource allocation and their implication on the application performance.
What problem does this paper attempt to address?
### Problems the Paper Aims to Solve
This paper aims to fill the research gap regarding the configuration of 5G mid-band networks and their impact on application performance. Specifically, through large-scale, multi-country field measurement studies conducted in Europe and the United States, the paper explores the following issues:
1. **5G Mid-Band Network Configuration**:
- How carriers configure their networks, including spectrum utilization, frame structure configuration, resource allocation, etc.
- Differences in 5G mid-band configurations between different countries and regions.
2. **Impact on Application Performance**:
- How network configuration parameters (such as channel bandwidth, frame structure, number of MIMO layers, etc.) affect physical layer (PHY) performance, such as downlink and uplink throughput, user plane latency, etc.
- The impact of these configuration parameters on actual application performance (such as file download time, video streaming quality, etc.).
### Main Objectives
1. **Comparative Analysis**:
- Provide a comparative analysis of 5G mid-band configurations in the United States and Europe, revealing the characteristics of 5G mid-band deployments in various countries.
2. **Performance Evaluation**:
- Explore the impact of different deployment configurations on 5G network and application performance, especially the impact of frame structure (TDD configuration) on physical layer performance and application experience.
### Challenges and Methods
1. **Challenges**:
- How to select countries and operators for comprehensive investigation.
- How to coordinate cross-continental data collection to obtain high-quality data.
- How to accurately analyze the relationship between network configuration parameters and 5G network and application performance.
2. **Methods**:
- Use multiple 5G smartphones and multiple SIM cards to collect data from different operators.
- Use professional 5G measurement tools like Accuver XCAL to collect detailed 5G NR protocol stack information.
- Select a variety of benchmark applications, such as iPerf, ping, file download, video streaming, etc.
- Choose servers located in the operator's core network or major cloud service providers for testing.
### Key Contributions and Findings
1. **Large-Scale Measurement Study**:
- Conducted the first large-scale, comprehensive 5G mid-band measurement study in selected cities in four European countries (Spain, France, Germany, Italy) and the United States.
- Revealed the 5G mid-band configurations used by operators, such as channels/bands, bandwidth, frame structure, number of MIMO layers, etc.
2. **Performance Impact Factors**:
- Identified key factors affecting 5G mid-band performance, including physical layer downlink and uplink throughput, user plane latency, etc.
- Found that MIMO and carrier aggregation (CA) can significantly enhance the physical layer downlink throughput of 5G mid-band.
3. **Impact of Configuration Parameters**:
- Observed no direct correlation between channel bandwidth and physical layer downlink throughput.
- Physical layer uplink throughput is significantly lower than downlink throughput and is far below 100 Mbps under "good" channel conditions.
- Frame structure (TDD configuration) has a significant impact on 5G mid-band performance, especially on physical layer user plane latency.
4. **Application Performance**:
- Faster physical layer downlink channels significantly reduce file download time only when the file size is large enough.
- For video streaming, 5G mid-band configuration, stability of the channel environment, and server location all have a significant impact on the overall application experience.
### Conclusion
This study provides valuable insights into the role of 5G mid-band network configuration in actual performance, helping operators better configure and optimize 5G mid-band services. It also provides guidance for users and application developers in choosing operators, application tuning, and server placement. Additionally, the study has generated the largest cross-continental, cross-national 5G mid-band measurement dataset to date and plans to make it publicly available.