• +353-1-416-8900(GMT OFFICE HOURS)
  • 1-800-526-8630(US/CAN TOLL FREE)
  • 1-917-300-0470(EST OFFICE HOURS)
Signals Ahead! LTE Advanced Network Drive Test – Gangnam Style! (As the Carrier Aggregation World Turns) - Product Image

Signals Ahead! LTE Advanced Network Drive Test – Gangnam Style! (As the Carrier Aggregation World Turns)

  • Published: October 2013
  • Region: Global
  • 62 Pages
  • Signals Research Group, LLC

This report provides results from a network performance benchmark study of LTE Advanced Carrier Aggregation, based on testing done in Seoul, South Korea. All drive test data was collected and analyzed with the Accuver suite of drive test solutions.

This report is specifically tailored to those operators and other entities who are interested in understanding the performance characteristics of carrier aggregation when it comes to downlink throughput versus a wide assortment of other network parameters and how these characteristics impact the user experience while performing various mobile data applications, including web browsing, video telephony, VoLTE, Skype Voice/Video, Google Play and 1080p video.

This report is included as part of a subscription to Signals Ahead or it can be purchased separately.

KEY QUESTIONS ANSWERED

1) What is the overall downlink performance of carrier aggregation and how do the primary and secondary carriers individually contribute to the total performance?

2) How does the simultaneous use of a high-band and a low-band carrier translate into actual performance, including other KPIs, such as resource block allocations, SINR READ MORE >

Contents
1.0 Executive Summary

2.0 Key Observations and Conclusions

3.0 Downlink Drive Tests - Detailed Analysis and Commentary
3.1 Early Wednesday AM Drive Tests
3.2 1815 Hours Drive Test
3.3 0906 Hours Drive Test
3.4 Simultaneous LTE Advanced Carrier Aggregation and LTE Release 8 Drive Test

4.0 Uplink Drive Tests - Detailed Analysis and Commentary

5.0 User Experience Tests - Detailed Analysis and Commentary
5.1 Web Browsing
5.2 FTP Server versus Speedtest.net versus Google Play
5.3 Voice and Video Telephony Applications

6.0 Test Methodology

7.0 Final Thoughts

Index of Figures

Figure 1. LTE Advanced Carrier Aggregation Downlink Throughput
Figure 2. Early Wednesday AM Drive Test Routes
Figure 3. Downlink Throughput by Primary and Secondary Carrier (Early AM) - Probability Distribution and Pie Charts
Figure 4. RSRP by Primary and Secondary Carrier (Early AM) - Probability Distribution
Figure 5. SINR by Primary and Secondary Carrier (Early AM) - Probability Distribution
Figure 6. Resource Block Allocation by Primary and Secondary Carrier (Early AM) - Probability Distribution
Figure 7. Detailed Analysis of MIMO, Secondary Carrier and Category 4 Device Utilization (Early AM)
Figure 8. Downlink Throughput by Primary and Secondary Carrier (Early AM) - Time Series
Figure 9. RSRP by Primary and Secondary Carrier versus Serving Cell PCI (Early AM) - Time Series
Figure 10. SINR by Primary and Secondary Carrier versus Serving Cell PCI (Early AM) - Time Series
Figure 11. Secondary Carrier Cell PCI Values - Full and Enhanced Views
Figure 12. Secondary Carrier SINR Values - Full and Enhanced Views
Figure 13. Resource Block Allocation by Primary and Secondary Carrier (Early AM) - Time Series
Figure 14. SINR versus Downlink Throughput - primary and secondary carriers (Early AM) - Scatter Plots
Figure 15. 1815 Hours AM Drive Test Routes
Figure 16. Downlink Throughput by Primary and Secondary Carrier (1815) - Probability Distribution and Pie Charts
Figure 17. RSRP by Primary and Secondary Carrier (1815 Hours) - Probability Distribution
Figure 18. SINR by Primary and Secondary Carrier (1815 Hours) - Probability Distribution
Figure 19. Resource Block Allocation by Primary and Secondary Carrier (1815 Hours) - Probability Distribution
Figure 20. Detailed Analysis of MIMO, Secondary Carrier and Category 4 Device Utilization (1815 Hours)
Figure 21. Downlink Throughput by Primary and Secondary Carrier (1815 Hours) - Time Series
Figure 22. Downlink Throughput by Primary Carrier with Individual Data Stream Contributions (1815 Hours) - Time Series
Figure 23. Downlink Throughput by Secondary Carrier with Individual Data Stream Contributions (1815 Hours) - Time Series
Figure 24. SINR and Resource Block Allocation by Primary and Secondary Carrier (1815 Hours) - Time Series
Figure 25. SINR versus Downlink Throughput - primary and secondary carriers (1815 Hours) - Scatter Plots
Figure 27. Downlink Throughput by Primary and Secondary Carrier (0906) - Probability Distribution
Figure 26. 0906 Hours AM Drive Test Routes
Figure 28. RSRP by Primary and Secondary Carrier (0906 Hours) - Probability Distribution
Figure 29. SINR by Primary and Secondary Carrier (0906 Hours) - Probability Distribution
Figure 30. Detailed Analysis of MIMO, Secondary Carrier and Category 4 Device Utilization (0906 Hours)
Figure 31. Simultaneous LTE Advanced and LTE Release 8 Drive Test Routes
Figure 32. Downlink Throughput by LTE Advanced (Total, Primary and Secondary) and Release 8 Carriers - Time Series
Figure 33. Downlink Throughput by LTE Advanced (Total, Primary and Secondary) and Release 8 Carriers with Serving Cell PCI Values and Carrier Frequency Assignments - Time Series
Figure 34. Combined Downlink Throughput for LTE Advanced and Release 8 - Probability Distribution
Figure 35. Uplink Drive Test Route
Figure 36. Uplink Throughput - Probability Distribution and Pie Charts
Figure 37. PUSCH Transmit Power (Uplink Drive Test) - Probability Distribution
Figure 38. Uplink Resource Block Allocation (Uplink Drive Test) - Probability Distribution
Figure 39. Detailed Analysis of Uplink Physical Layer Throughput versus Transmit Power versus Resource Block Allocation (South Korea)
Figure 40. Detailed Analysis of Uplink Physical Layer Throughput versus Transmit Power versus Resource Block Allocation (LTE TDD - Japan)
Figure 41. Detailed Analysis of Uplink Physical Layer Throughput versus Transmit Power versus Resource Block Allocation (LTE FDD 10 MHz - Japan)
Figure 42. Detailed Analysis of Uplink Physical Layer Throughput versus Transmit Power versus Resource Block Allocation (LTE FDD 5 MHz - Japan)
Figure 43. Web Page Load Times - by access network technology
Figure 44. Web Page Browsing with a Notebook Computer and its Impact on Resource Allocation - per TTI subframe Time Series
Figure 45. Carrier and MIMO Rank Indicator 2 Utilization while Web Browsing with a Notebook Computer
Figure 46. Carrier Aggregation Utilization while Web Browsing with a Notebook Computer - as a function of TBS
Figure 47. Carrier and MIMO Rank Indicator 2 Utilization while Web Browsing with a Smartphone Carrier Aggregation Utilization while Web Browsing with a Smartphone
Figure 48. Carrier Aggregation Utilization while Web Browsing with a Smartphone - as a function of TBS
Figure 49. A Comparison of TBS Allocations - notebook computer versus smartphone
Figure 50. Carrier Aggregation and MIMO Rank Indicator 2 Utilization while Web Browsing with a Smartphone - as a function of TBS
Figure 51. Maximum Achievable Throughput - by Test Methodology and Application
Figure 53. Key Network Application Parameters - by Application
Figure 54. You Tube 1080p Video Playback - Time Series
Figure 55. SINR by Primary and Secondary Carrier (YouTube 1080p playback) - Time Series
Figure 56. RSRP by Primary and Secondary Carrier (YouTube 1080p playback) - Time Series
Figure 57. XCAL in Action

Table 1. Key Network Utilization Parameters - by Application

This research report is the first in-depth independent analysis of LTE Advanced Carrier Aggregation. This effort would not have been possible without the support of Accuver, who provided us with access to its XCAL data collection tool and its XCAP post-processing software. We have used the solution numerous times over the last several years so we are very accustomed to using it, although we do stumble upon new capabilities and features each time we use it. In our most recent benchmark studies, including LTE TDD in Tokyo and LTE Advanced Carrier Aggregation in Seoul, the solution’s ability to support recently introduced technology features, including Category 4 chipsets and Carrier Aggregation, proved to be invaluable.

For the LTE Advanced testing we also used Spirent Communications’ Datum solution, which it inherited when it acquired Metrico Wireless. We used Datum for some of the user experience tests that we conducted. We’ve used the tool in the past, including for a multi-network benchmark study that was commissioned by a major operator – Datum was the operator’s preferred solution for the study.

After spending five days in Seoul testing LTE Advanced we are forever tainted and our expectations for what we consider to be great network performance have been raised to an unattainable level. To put things into perspective, the average downlink throughput during all of the testing was more than three orders of magnitude higher than what operators advertised a little more than a decade ago. The uplink throughput for a 10 MHz radio carrier was equally impressive, or roughly 33% higher than the best performance that we have observed in the past – AT&T’s pre-commercial LTE network in Houston.

The great performance that we observed will be hard to replicate, largely because the South Korean operators have deployed very dense networks in Seoul. Separate from the published numbers of deployed base stations and remote radio heads in the country, the quality / density of the network can be observed in the stellar RSRP values, the high uplink throughput with modest transmit power levels, and the fact that we observed throughput greater than 100 Mbps during rush hour traffic for sustained periods of time.

Further, not all operators will deploy 10 MHz + 10 MHz implementations of carrier aggregation since at least in the near term some operators lack 10 MHz of contiguous spectrum in two suitable frequency bands. AT&T, for example, only has 5 MHz of 700 MHz spectrum in Chicago and Miami so the best it will be able to do in these markets is 5 MHz + 5 MHz or 5 MHz + 10 MHz. Other operators, including TeliaSonera and Verizon Wireless, may also slow down their rollout of carrier aggregation since with current chipset limitations they will not be able to use a full 2 x 20.

MHz LTE carrier in Band 4 (VZW) or Band 7 (TeliaSonera). We hope to return to South Korea late next year when carrier aggregation with 10 MHz + 20 MHz channels is ready for primetime.

Key questions that we address in this report include the following:

1) What is the overall downlink performance of carrier aggregation and how do the primary and secondary carriers individually contribute to the total performance?

2) How does the simultaneous use of a high-band and a low-band carrier translate into actual performance, including other KPIs, such as resource block allocations, SINR and RSRP, throughout the cell grid?

3) Other than basic throughput, what other carrier aggregation attributes impact the user experience and to what degree and under what conditions?
4) What impact does carrier aggregation have on the user experience for basic mobile data applications, such as web browsing, Skype Voice/Video, VoLTE, Video Telephony and 1080p video?

5) How does the use of two simultaneous carriers vary as a function of the transport block size (TBS)?

6) How does the TBS differ between web browsing on a notebook computer versus a smartphone browser?

7) How does the uplink performance (throughput, transmit power, transmit power per RB) in Seoul compare with the LTE TDD and LTE FDD networks that we recently tested in Japan?

8) How much incremental performance gain does a Category 4 device deliver versus a Category 3 device and how often is this gain realized?

9) What are the network resource requirements for various applications, including VoLTE, Skype Voice/Video, Video Telephony and 1080p video?

10) What is the aggregate throughput of a Release 10 and Release 8 modem running in parallel and what is the difference in between the two modems?

Note: Product cover images may vary from those shown

ALSO AVAILABLE

  • Signals Ahead - Product Thumbnail Image Signals Ahead

    Signals Ahead

    • Language: English
    • Published: August 2014
    FROM

RELATED PRODUCTS

Our Clients

Our clients' logos