Signals Ahead: The Mother of all Network Benchmark Tests Volume 3 - Detailed Performance Analysis (the LTE edition)
Signals Research Group, LLC, January 2012
Over the last several months Signals Research Group has been conducting a network/technology benchmark study of all of the leading next-generation wireless technologies, as exemplified by operator deployments in North America. 'Signals Ahead: The Mother of all Network Benchmark Tests ' is a series of three reports:
Volume 1
Network and Technology Performance
Volume 2
Quantifying the User Experience
Volume 3 and Volume 4
Detailed Performance Analysis
The four reports 'Signals Ahead: The Mother of all Network Benchmark Tests ' are included with any paid corporate subscription to Signals Ahead, or can be purchased separately.
Volume 4 is free if you purchase Volume 3 or the complete set of reports.
ABOUT VOLUME 3 - DETAILED PERFORMANCE ANALYSIS
This reports provides a deep-dive analysis of how LTE networks perform today as exemplified by operator deployments in North America, and to a lesser extent Europe. Specifically, the focus is almost entirely on AT&T and Verizon Wireless, as well as Alcatel Lucent (ALU) and Ericsson (ERICY). It also includes at least some analysis of Nokia Siemens Networks (NSN) and Huawei from previous testing that we have done – the latter based solely on Clearwire's demonstration LTE network. Finally, the “Chips and Salsa” analysis from a previous reports is extended by comparing and contrasting the performance of different dongles and LTE chipsets.
A companion report intended to be publish in early February will provide similar analysis for the HSPA+/DC-HSDPA networks, thus directing the attention on AT&T, T-Mobile, Ericsson, Nokia Siemens Networks, and Alcatel Lucent.
Key takeaways and observations discussed in far more detail in this report and/or in our forthcoming HSPA+/DC-HSDPA report include the following:
- The differences in downlink throughput that we observed on the AT&T and Verizon Wireless LTE networks had absolutely nothing to do with the underlying RF conditions or with network loading. Instead, we attribute the differences entirely to …
-The differences in uplink throughput that we observed in our LTE network testing were vendor specific and they represent a critical differentiator that exists today. Based on the underlying test data, the throughput differences were due to …
- Infrastructure suppliers have taken different approaches regarding which form of MIMO (e.g., OL-SM or CL-SM) they want to implement first. More importantly, they have different philosophies regarding how aggressively they want to use MIMO.
- Frequency selective scheduling (FSS) is an important means of driving network efficiency and maximizing the SNR to drive higher individual data rates. Unfortunately, not all vendors currently support the full capability and ICIC (Inter-cell Interference Coordination) is just around the corner.
- 64QAM and spatial multiplexing are two key enablers of higher data rates. We provide test results that span frequency bands (e.g., 700MHz and 2600MHz), continents and vendors to quantify how frequently MIMO is being used at 700MHz versus 2600MHz and how frequently 64QAM is being used with LTE versus HSPA+. We conclude that …
- Throughput during a cell handover is important for delivering a consistent user experience, but it requires a degree of network optimization. We demonstrate material differences between two infrastructure [HSPA+] suppliers, as well as between two [LTE] dongles/chipsets.
- In addition to performance during a cell handover, we demonstrate that there were material performance differences between different LTE dongles, as well as between different LTE chip-sets – a nice continuation from our last “Chips and Salsa” report.
- DC-HSDPA leverages two separate radio channels but for one infrastructure supplier it had a very difficult time maintaining both radio channels, especially during a cell handover, thus relegating DC-HSDPA to a semi-portable technology.
- There were noticeable throughput differences between the three HSPA+ suppliers once we normalized the results for a given channel condition and network loading. One infrastructure supplier, in particular, did much better than its peers.
- 64QAM and 14 HS-PDSCH codes are required, among other things, to achieve the peak data rates associated with HSPA+. We found that one infrastructure vendor's solution was limited to only 13 HS-PDSCH codes [26 HS-PDSCH codes with DC-HSDPA].
REASONS TO BUY
This report is critical for operators trying to understand how to market their broadband wireless service offering as well as how they should prioritize their network optimization activities in order to achieve the best possible user experience for their subscribers. In addition to mobile operators, this report provides invaluable insight to application developers and content providers who require a greater appreciation for how network performance characteristics impact the user experience.
SIGNALS AHEAD SUBSCRIPTION INFORMATION
Signals Ahead is a research-focused product that is published on a periodic basis. Its clientele include all facets of the wireless ecosystem, including some of the largest mobile operators, the top handset suppliers, the major infrastructure vendors, subsystem suppliers, semiconductor companies and financial institutions, including Wall Street, Private Equity and Venture Capitalists, spread across five continents.
1.0 Executive Summary
2.0 key conclusions and observations
3.0 lte and hspa+ comparative analysis
4.0 detailed test results and analysis – lte downlink throughput
4.1 Generalized Findings and Observations
4.2 Detailed Test Results and Analysis – LTE Downlink Throughput
4.2.1 Test Scenario 137 – Verizon Wireless (ALU) and AT&T (ERICY) LTE Network Analysis
4.2.2 Supporting Information from other Drive Test Scenarios
5.0 detailed test results and analysis – a tale of different chipsets and Infrastructure suppliers
5.1 Stationary Test Results on the Verizon Wireless Network (Alcatel Lucent) with Two Dongles/Chipsets –
Test Scenario 161
5.2 Drive Test Results on the Verizon Wireless Network (Alcatel Lucent) with Two Dongles/Chipsets –
Test Scenario 152
5.3 Drive Test Results on the Verizon Wireless Network (Ericsson) – Test Scenario 115
6.0 detailed test results and analysis – lte uplink throughput
7.0 test methodology
8.0 conclusions
INDEX OF TABLES
table 1. LTE Handover Analysis – by operator, vendor, market and frequency band
table 2. LTE Application Layer Handover Interruption Time
INDEX OF FIGURES
figure 1. AT&T LTE and HSPA+ Networks – Drive Route (Test Scenarios 138-140)
figure 2. AT&T LTE and HSPA+ Networks – Key Performance Indicators (Test Scenarios 138-140)
figure 3. Modulation Scheme Distribution for Numerous LTE Networks – by operator, vendor, market and frequency band
figure 4. Rank Indicator Distribution for Numerous LTE Networks – by operator, vendor, market and
frequency band
figure 5. Drive Route – Test Scenario 137
figure 6. Number of Assigned Resource Blocks versus Throughput – AT&T (ERICY) LTE Network (Test Scenario 137)
figure 7. Number of Assigned Resource Blocks versus Throughput – Verizon Wireless (ALU) LTE Network (Test Scenario 137)
figure 8. Cell/Sector ID versus Throughput – AT&T (ERICY) LTE Network (Test Scenario 137)
figure 9. Cell/Sector ID versus Throughput – Verizon Wireless (ALU) LTE Network (Test Scenario 137)
figure 10. Reported MCS Codeword #0 and Codeword #1 versus Throughput - AT&T (ERICY) LTE Network
(Test Scenario 137)
figure 11. Reported MCS Codeword #0 and Codeword #1 versus Throughput Scatter Pot - AT&T (ERICY)
LTE Network (Test Scenario 137)
figure 12. Reported MCS Codeword #0 and Codeword #1 versus Throughput – Verizon Wireless (ALU)
LTE Network (Test Scenario 137)
figure 13. Reported MCS Codeword #0 and Codeword #1 versus Throughput Scatter Plot –
Verizon Wireless (ALU) LTE Network (Test Scenario 137)
figure 14. Distribution of Transmission Modes - AT&T (ERICY) LTE Network (Test Scenario 137)
figure 15. Distribution of Transmission Modes – Verizon Wireless (ALU) LTE Network (Test Scenario 137)
figure 16. Distribution of Modulation Schemes – AT&T (ERICY) LTE Network (Test Scenario 137)
figure 17. Distribution of Modulation Schemes – Verizon Wireless (ALU) LTE Network (Test Scenario 137)
figure 18. Unadjusted RSRQ versus Throughput Scatter Plot – AT&T (ERICY) LTE Network (Test Scenario 137)
figure 19. Unadjusted RSRQ versus Throughput Scatter Plot – Verizon Wireless (ALU) LTE Network
(Test Scenario 137)
figure 20. Unadjusted RSRQ versus Rank Indicator Scatter Plot – AT&T (ERICY) LTE Network (Test Scenario 137)
figure 21. PUSCH Transmit Power versus Throughput Timeline Series – AT&T (ERICY) LTE Network
(Test Scenario 137)
figure 22. PUSCH Transmit Power versus Throughput Scatter Plot – AT&T (ERICY) LTE Network (Test Scenario 137)
figure 23. PUSCH Transmit Power versus Throughput Timeline Series – Verizon Wireless (ALU) LTE Network
(Test Scenario 137)
figure 24. PUSCH Transmit Power versus Throughput Scatter Plot – Verizon Wireless (ALU) LTE Network
(Test Scenario 137)
figure 25. PUSCH Transmit Power versus Serving Cell/Sector ID – AT&T (ERICY) LTE Network (Test Scenario 137)
figure 26. PUSCH Transmit Power versus Serving Cell/Sector ID – Verizon Wireless (ALU) LTE Network
(Test Scenario 137)
figure 27. Dallas Early Morning Drive Route – Test Scenario 1 and Test Scenario 2
figure 28. San Francisco Drive Route – Test Scenario 98
figure 29. Unadjusted RSRQ versus Throughput Scatter Plot – Verizon Wireless (ALU) LTE Network (Test Scenario 1)
figure 30. Unadjusted RSRQ versus Throughput Scatter Plot – Verizon Wireless (ALU) LTE Network (Test Scenario 2)
figure 31. Unadjusted RSRQ versus Throughput Scatter Plot – Verizon Wireless (ERICY) LTE Network (Test Scenario 98)
figure 32. Unadjusted RSRQ versus Throughput Scatter Plot – Verizon Wireless (ALU) LTE Network (Test Scenario 59)
figure 33. Reported MCS Codeword #0 and Codeword #1 versus Throughput Scatter Plot – Verizon Wireless (ALU) LTE Network (Test Scenario 59)
figure 34. Number of Assigned Resource Blocks versus Throughput in an Alcatel Lucent Network – Pantech Dongle/Qualcomm Chipset (Test Scenario 161)
figure 35. Number of Assigned Resource Blocks versus Throughput in an Alcatel Lucent Network – LG Dongle/LG Chipset (Test Scenario 161)
figure 36. Reported MCS Codeword #0 and Codeword #1 versus Throughput in an Alcatel Lucent Network Scatter Plot – Pantech Dongle/Qualcomm Chipset (Test Scenario 161)
figure 37. Reported MCS Codeword #0 and Codeword #1 versus Throughput in an Alcatel Lucent Network Scatter Plot – LG Dongle/LG Chipset (Test Scenario 161)
figure 38. Distribution of Rank Indicator and Modulation Schemes in an Alcatel Lucent Network – Pantech Dongle/Qualcomm Chipset and LG Dongle/LG Chipset (Test Scenario 161)
figure 39. Distribution of Modulation Schemes in an Alcatel Lucent Network – Pantech Dongle/Qualcomm Chipset (Test Scenario 161)
figure 40. Distribution of Modulation Schemes and ACK/NACK/DTX in an Alcatel Lucent Network – Pantech Dongle/Qualcomm Chipset (Test Scenario 161)
figure 41. Houston Drive Route – Test Scenario 152
figure 42. Reported MCS Codeword #0 and Codeword #1 versus Throughput in an Alcatel Lucent Network – Pantech Dongle/Qualcomm Chipset (Test Scenario 152)
figure 43. Reported MCS Codeword #0 and Codeword #1 versus Throughput in an Alcatel Lucent Network Scatter Plot – Pantech Dongle/Qualcomm Chipset (Test Scenario 152)
figure 44. Reported MCS Codeword #0 and Codeword #1 versus Throughput in an Alcatel Lucent Network – LG Dongle/LG Chipset (Test Scenario 152)
figure 45. Reported MCS Codeword #0 and Codeword #1 versus Throughput in an Alcatel Lucent Network Scatter Plot – LG Dongle/LG Chipset (Test Scenario 152)
figure 46. Distribution of Rank Indicator and Modulation Schemes on an Alcatel Lucent Network – Pantech Dongle/Qualcomm Chipset (Test Scenario 152)
figure 47. Distribution of Rank Indicator and Modulation Schemes, including ACK/NACK/DTX, on an Alcatel Lucent Network – LG Dongle/LG Chipset (Test Scenario 152
figure 48. Number of Assigned Resource Blocks versus Throughput on an Alcatel Lucent Network – Pantech Dongle/Qualcomm Chipset (Test Scenario 152)
figure 49. Number of Assigned Resource Blocks versus Throughput on an Alcatel Lucent Network – LG Dongle/LG Chipset (Test Scenario 152)
figure 50. Cell/Sector ID versus Throughput on an Alcatel Lucent Network – Pantech Dongle/Qualcomm Chipset (Test Scenario 152)
figure 51. Cell/Sector ID versus Throughput on an Alcatel Lucent Network – LG Dongle/LG Chipset (Test Scenario 152)
figure 53. Oakland Drive Route – Test Scenario 115
figure 54. Number of Assigned Resource Blocks versus Throughput on an Ericsson Network – Pantech Dongle/Qualcomm Chipset (Test Scenario 115) figure 55. Number of Assigned Resource Blocks versus Throughput on an Ericsson Network – LG Dongle/LG Chipset (Test Scenario 115)
figure 56. Cell/Sector ID versus Throughput on an Ericsson Network – Pantech Dongle/Qualcomm Chipset (Test Scenario 115)
figure 57. Cell/Sector ID versus Throughput on an Ericsson Network – LG Dongle/LG Chipset (Test Scenario 115)
figure 58. Reported MCS Codeword #0 and Codeword #1 versus Throughput in an Ericsson Network Scatter Plot –
Pantech Dongle/Qualcomm Chipset (Test Scenario 115)
figure 59. Distribution of Modulation Schemes in an Ericsson Network – Pantech Dongle/Qualcomm Chipset
(Test Scenario 115)
figure 60. Reported MCS Codeword #0 and Codeword #1 versus Throughput in an Ericsson Network Scatter Plot – LG Dongle/LG Chipset (Test Scenario 115)
figure 61. Distribution of Modulation Schemes in an Ericsson Network – LG Dongle/LG Chipset (Test Scenario 115)
figure 62. Throughput by Codeword in an Ericsson Network - LG Dongle/LG Chipset (Test Scenario 115)
figure 63. PUSCH Transmit Power versus Time in an Ericsson Network – both chipsets
figure 64. Drive Route – Test Scenario 145
figure 65. PUSCH Transmit Power versus Uplink Throughput – AT&T (ERICY) Network (Test Scenario 145)
figure 66. PUSCH Transmit Power versus Uplink Throughput – Verizon Wireless (ALU) Network (Test Scenario 145)
figure 67. PUSCH Transmit Power versus Uplink Throughput – Verizon Wireless (ERICY) Network (Test Scenario 119)
figure 68. PUSCH Transmit Power versus Uplink Throughput – Clearwire (Huawei) Network (Test Scenario 57)
figure 69. PUSCH Transmit Power versus Serving Cell/Sector ID – AT&T (ERICY) Network (Test Scenario 145)
figure 70. PUSCH Transmit Power versus Serving Cell/Sector ID –Verizon Wireless (ALU) Network
(Test Scenario 145)
figure 71. PUSCH Transmit Power versus Serving Cell/Sector ID –Verizon Wireless (ERICY) Network
(Test Scenario 119)
figure 72. PUSCH Transmit Power versus Serving Cell/Sector ID –Clearwire (Huawei) Network (Test Scenario 57)
figure 73. Cell/Sector ID versus Uplink Throughput –AT&T (ERICY) Network (Test Scenario 145)
figure 74. Cell/Sector ID versus Uplink Throughput –Verizon Wireless (ERICY) Network (Test Scenario 145)
figure 75. Cell/Sector ID versus Uplink Throughput –Verizon Wireless (ERICY) Network (Test Scenario 119)
figure 76. Cell/Sector ID versus Uplink Throughput – Clearwire (Huawei) Network (Test Scenario 57)
figure 77. Number of Assigned Resource Blocks versus Uplink Throughput – AT&T (ERICY) Network
(Test Scenario 145)
figure 78. Number of Assigned Resource Blocks versus Uplink Throughput – Verizon Wireless (ALU) Network
(Test Scenario 145)
figure 79. Number of Assigned Resource Blocks versus Uplink Throughput – Verizon Wireless (ERICY) Network
(Test Scenario 145)
figure 80. Number of Assigned Resource Blocks versus Uplink Throughput – Clearwire (Huawei) Network
(Test Scenario 57)
figure 81. Number of Assigned PUSCH Resource Blocks versus Uplink Throughput – AT&T (ERICY) Network
(Test Scenario 145)
figure 82. Number of Assigned PUSCH Resource Blocks versus Uplink Throughput Uplink – Verizon Wireless (ALU) Network (Test Scenario 145)
figure 83. Number of Assigned PUSCH Resource Blocks versus Uplink Throughput – Clearwire (Huawei) Network (Test Scenario 57)
figure 84. Distribution of Reported MCS and Modulation Schemes in the Uplink – AT&T (ERICY) Network (Test Scenario 145)
figure 85. Distribution of Reported MCS and Modulation Schemes in the Uplink – Verizon Wireless (ALU) Network (Test Scenario 145)
figure 86. Distribution of Reported MCS and Modulation Schemes in the Uplink – Verizon Wireless (ERICY) Network (Test Scenario 145)
figure 87. Distribution of Reported MCS and Modulation Schemes in the Uplink Throughput – Clearwire (Huawei) Network (Test Scenario 57)
figure 88. XCAL-M Drive Test Tool in Action – DL performance
Product Samples
A sample for this product is available. Please Login/Register to download this sample.
Also available
Customers who bought this item also bought
All rights reserved. © Copyright 2013 Research and Markets WWW4
Terms and Conditions Privacy Policy Publishers Employment Opportunities Site Map Link to us Webmaster Affiliate Network
