LTE - RF Drivetest and Coverage Analysis
Different Between TD-LTE & FD-LTE Item
LTE-TDD
LTE-FDD
Duplex mode
TDD
FDD
Frame structure
Type 2
Type 1
UL and DL Ratio
7 types of UL and DL ratio, flexible
All subframes can be allocated only for the uplink or downlink.
RRU Noise Figure
A T/R converter is required. The T/R converter will bring about the insertion loss of 2~2.5 dB .
Beamforming
A duplexer is required and the duplexer bri ngs about the insertion loss of 1 dB.
Supported (exchangeability based on Not supported (no exchangeability based on uplink uplink and downlink channel) and downlink channels)
MIMO Mode
Modes 1 –8 are supported.
Mode 1 –6 are supported.
Network Interference
Strict synchronization is required in the whole network.
Synchronization requirement is not strict. strict.
Different Between TD-LTE & FD-LTE Item
LTE-TDD
LTE-FDD
Duplex mode
TDD
FDD
Frame structure
Type 2
Type 1
UL and DL Ratio
7 types of UL and DL ratio, flexible
All subframes can be allocated only for the uplink or downlink.
RRU Noise Figure
A T/R converter is required. The T/R converter will bring about the insertion loss of 2~2.5 dB .
Beamforming
A duplexer is required and the duplexer bri ngs about the insertion loss of 1 dB.
Supported (exchangeability based on Not supported (no exchangeability based on uplink uplink and downlink channel) and downlink channels)
MIMO Mode
Modes 1 –8 are supported.
Mode 1 –6 are supported.
Network Interference
Strict synchronization is required in the whole network.
Synchronization requirement is not strict. strict.
Drive Test Peripheral
LTE Dongle D ongle
Notebook
GPS
Reference Signal Received Power Pow er (RSRP) (RSRP)
LTE RS Power Allocation •
How to calculate RS Power ? RRU 3252 run at 4T4R configuration, have total power 80W (Max 20W/Port). RSRE Power = Psingle port -10*log(12*Nrb)+10*log(1+Pb) Where ; PSingle Port = PRRU - 10*log(Nport) ρB/ ρA Pb is Power Bosting
Psingle port = 49-10*log(4) = 43 dBm = 20Watt
Single Antenna Port
2 or 4 Antenna Port
0
1
5/4
1
4/5
1
2
3/5
¾
3
2/5
½
PB
If operator have 20 MHz for the first carrier and 10 Mhz for second carrier each carrier can use 10W for first carrier and 5W for second carrier to maintain the
LTE RS Power Allocation •
RS Power for 20 MHz @ 10W/port RS = 40dBm – 10*log(12*100) + 10*log(1+1) = 12.2 dBm
•
RS Power for 10 MHz @ 5W/port RS = 37dBm – 10*log(12*50) + 10*log(1+1) = 12.2 dBm
RS Power for 10 MHz @ 10W/port RS = 40dBm – 10*log(12*50) + 10*log(1+1) = 15.2 dBm With the same total power, coverage LTE 10 Mhz is larger than with LTE 20 MHz
Impact on Radio Network Performance: A larger value of Pb results in a larger increase in ReferenceSignalPwr, better channel estimation performance, and better PDSCH demodulation performance, but it also leads to lower transmit power of the PDSCH (type B) and thus increases
RS Power for 10 MHz @ 10W/port RS = 40dBm – 10*log(12*50) + 10*log(1+1) = 15.2 dBm With the same total power, coverage LTE 10 Mhz is larger than with LTE 20 MHz
LTE RS Power Allocation •
Power Boosting for RS
P =1 by default B
•
RS Power for 20MHz
Bandwidth 10M 15M 20M
PB 1 1 1
PRS ( dBm)
18.2 16.4 15.2
RS Power Overhead Comparison with CPICH
Type B Symbol: with RS REs
• • • •
Type A Symbol: without RS REs
RS power per RE is 15.2dBm (0.033W) for 20MHz Total RS power in 20MHz for Type B Symbol is 0.033*2 (RS REs/ RB) * 100 RBs = 6.6W Total RS power in 20MHz for Type A Symbol is 0 Only two symbols carry RS within 0.5ms and hence the RS power overhead is about 6.6/20 * 2/7 = 9.4% over 1 timeslot
LTE RS power overhead is about 9.4% which is similar to 10% CPICH power overhead
RxLev, RSRP and RSCP Comparison Items
GSM
UMTS
LTE
43
43
43
Bandwidth (MHz)
0.2
5
20
Number of RB
N/A
N/A
100
BCCH Power/ CPICH power /RS power per RE (dBm)
43
33
15.2
CL (dB)
120
120
120
-77
-87
-104.8
(e)NodeB power per Tx (dBm)
Rx Lev/RSCP/RSRP (dBm) Received RS signal strength over whole bandwidth
-81.8
RSRP is the received signal strength over 15KHz bandwidth while bandwidth of RSCP is 5MHz
Only 1/6 REs is used for RS transmission within one RB and hence the total received RS power is 10*log10(100*12*1/6) = 23dB higher than RSRP
Factors Influencing LTE Coverage Frequency
LTE Specific
Band
ICIC
Data Rate
TX Power Factors Affecting LTE Link Budget Cell Load
Interference Margin
RB Number
MCS
MIMO
LTE Standard
Radio Condition
LTE Specific LTE Specific
Receiver Sensitivity
Weak Coverage and Coverage Holes The signal quality in cells is poorer than the optimization baseline in an area.
Weak coverage
As a result, UEs cannot be registered with the network or accessed services cannot meet QoS requirements.
If there is no network coverage or coverage levels are excessively low in an area, the area is called a weak coverage area. The receive level of a UE is less than its minimum access level (RXLEV_ACCESS_MIN) because
Coverage holes
downlink receive levels in a weak coverage area are unstable. In this situation, the UE is disconnected from the network. After entering a weak coverage area, UEs in connected mode cannot be handed over to a high-level cell, and even service drops occur because of low levels and signal quality.
Resolving Weak Coverage Problems Analyze
geographical environments and
Deploy
new eNodeBs if coverage hole
Use
RRUs, indoor distribution systems,
check the receive levels of adjacent
problems cannot be resolved by
leaky feeders, and directional antennas to
eNodeBs.
adjusting antennas.
resolve the problem with blind spots in
Increase
elevator shafts, tunnels, underground
Analyze
the EIRP of each sector based on
coverage by adjacent eNodeBs
parameter configurations and ensure
to achieve large coverage overlapping
garages or basements, and high
EIRPs can reach maximum values if
between two eNodeBs and ensure a
buildings.
possible.
moderate handover area.
Analyze
Note: Increasing coverage may lead to
terrains on coverage.
Increase Adjust
pilot power.
antenna azimuths and tilts,
increase antenna height, and use high-gain antennas.
co-channel and adjacent-channel interference.
the impact of scenarios and
Case: Searching for a Weak Coverage Area by Using a Scanner or Performing Drive Tests on UEs
Perform drive tests in zeroload environments to obtain the distribution of signals on test routes. Then, find a weak coverage area based on the distribution, as shown in the figure. Adjust RF parameters of the eNodeB covering the area.
Weak coverage area
Lack of a Dominant Cell
In an area without a do minant cell, the receive level of the serving cell is similar to the receive levels of its neighboring cells and the receive levels of d ownlink signals between different cells are close to cell reselection thresholds. Receive levels in an area without a dominant cell are also unsatisfactory. The SINR of
Lack of a dominant cell
the serving cell becomes unstable because of frequency reuse, and even receive quality b ecomes unsatisfactory. In this situation, a dominant cell is frequently reselected and changed in idle mode. As a result, frequent handovers or service drops o ccur on UEs in connected mode b ecause of poor signal quality. An area without a dominant cell can also be regarded as a weak coverage area.
Resolving Problems with Lack of a Dominant Cell Determine
cells covering an
Adjust engineering
area without a dominant cell
parameters of a cell that can
during network planning, and
optimally cover the area as
adjust antenna tilts and
required.
azimuths to increase coverage by a cell with strong signals and decrease coverage of other cells with weak signals.
…
Case: Searching for an Area Without a Dominant Cell Symptom
UEs frequently perform cell reselections or handovers between identical cells. Analysis Analysis can be based on signaling procedures and PCI distribution. According to PCI distribution shown in the figure, PCIs alternate in two or more colors if there is no dominant cell. Solution According to the coverage plan, cell 337 is a dominant cell covering the area and cell 49 also has strong signals. To ensure handovers between cells 337 and 49 at crossroads, increase tilts in cell 49. Lack of a dominant cell
Cross Coverage
Cross coverage means that the coverage scope of an eNodeB exceeds the planned one and generates discontinuous dominant areas in the coverage scope of other eNodeBs. For example, if the height of a site is much higher than the average height of surrounding buildings, its transmit signals propagate far along hills or roads and form dominant coverage in the coverage scope of other eNodeBs. This is an “island” phenomenon.
Cross coverage
If a call is connected to an island that is far away from an eNodeB but is still served by the eNodeB, and cells around the island are not configured as neighboring cells of the current cell when cell handover parameters are configured, call drops may occur i mmediately once UEs leave the island. If neighboring cells are configured but the island is excessively small, call drops may also occur because UEs are not promptly handed over. In addition, cross coverage occurs on two sides of a bay because a short distance between the two sides. Therefore, eNodeBs on two sides of a bay must be specifically designed.
Resolving Cross Coverage Problems Adjust
antenna azimuths properly
Adjust
antenna tilts or replace
Decrease
the antenna height for
so that the direction of the main
antennas with large-tilt antennas
a high site.
lobe slightly obliques from the
while ensuring proper antenna
Decrease
direction of a street. This reduces
azimuths. Tilt adjustment is the
carriers when cell performance is
excessively far coverage by electric
most effective approach to control
not affected.
waves because of reflection from
coverage. Tilts are classified into
buildings on two sides of the street.
electrical tilts and mechanical tilts.
…
Electrical tilts are preferentially adjusted if possible.
transmit power of
Case: Cross Coverage Caused by Improper Tilt Settings
Symptom As shown in the upper right figure, cross coverage occurs in a cell whose PCI is 288. Therefore, the cell interferes with other cells, which increases the probability of service drops.
Analysis The most possible cause for cross coverage is excessively antenna height or improper tilt settings. According to a check on the current engineering parameter settings, the tilt is set to an excessively small value. Therefore, it is recommended that the tilt be increased.
Solution Adjust the tilt of cell 288 from 3 to 6. As shown in the lower right figure, cross coverage of cell 288 is significantly reduced after the tilt is a djusted.
Case: Inverse Connections Involved in the Antenna System
Symptom The RSRPs of cells 0 and 2 at the Expo Village site are low and high respectively in the red area shown in the figure. The signal quality of cells 0 and 2 is satisf actory in the areas covered by cells 2 and 0 res pectively. Analysis After installation and commissioning are complete, the RSRP in the direction of the main lobe in cell 0 is low. After cell 0 is disabled and cell 2 is enabled, the RSRP in cell 2 is normal and the SINR is higher than that tested in cell 0. Therefore, this problem may occur because the antenna systems in the two cells are connected inversely. Test results are as expected after optical fibers on the baseband board are swapped. Solution Swap optical fibers on t he baseband board or adjust feeders and antennas properly. It is recommended that optical fibers on the baseband board be swapped because this operation can be performed in the equipment room. Suggestions Network planning personnel must participate in installation. Alternatively, customer service personnel have detailed network planning materials and strictly supervise project constructors for installation. After installation is complete, labels must be attached and installation materials must be filed.
Imbalance Between Uplink and Downlink When UE transmit power is less than eNodeB transmit power, UEs in idle mode may receive eNodeB signals and successfully register in cells. However, the eNodeB cannot receive uplink signals because of limited power when UEs perform random access or upload data. In this situation, the uplink coverage distance is less than
Imbalance between uplink and downlink
the downlink coverage distance. Imbalance between uplink and downlink involves limited uplink or downlink coverage. In limited uplink coverage, UE transmit power reaches its maximum but still cannot meet the requirement for uplink BLERs. In limited downlink coverage, the downlink DCH transmit code power reaches its maximum but still cannot meet the requirement for the downlink BLER. Imbalance between uplink and downlink leads to service drops. The most common cause is limited uplink coverage.
Uplink coverage area
Downlink coverage area
Resolving Problems with Imbalance Between Uplink and Downlink
If
no performance data is available for RF
If
uplink interference leads to imbalance between
optimization, trace a single user in the OMC
uplink and downlink, monitor eNodeB alarms to
equipment room to obtain uplink measurement
check for interference.
reports on the Uu interface, and then analyze the
Check
measurement reports and drive test files.
whether alarms are generated if imbalance between
If
uplink and downlink is caused by other factors, for
performance data is available, check each
carrier in each cell for imbalance between uplink
whether equipment works properly and
and downlink based on uplink and downlink
example, uplink and downlink gains of repeaters and trunk amplifiers… are set incorrectly, the antenna
balance measurements.
system for receive diversity is faulty when reception and transmission are separated, or power amplifiers are faulty. If equipment works properly o r alarms are generated, take measures such as replacement, isolation, and adjustment.
Signal to Noise & Interference Ratio (SINR)
Traditional Frequency Planning 1*3*3 Frequency Planning
Advantage Lower interference and larger coverage radius Disadvantage Lower spectrum efficiency Suitable Scenario Abundant frequency resource or inconsecutive spectrum scenarios large coverage scenarios. 1*3*1 Frequency Planning Advantage Higher spectrum eff iciency Disadvantage Lower cell edge throughput due to serious interference Suitable Scenario Lacking frequency resource
1*3*3
Interference and Capacity Comparison 1*3*3 Vs 1*3*1 1*3*3 10MHz channel (30MHz) compare with 1*3*1 10MHz channel (10MHz) 1 0.8 F D C
0.6 0.4 0.2 0 -10
0
10
SINR
1×3×1
20
30
40
1×3×3
SINR distribution comparison 1*3*3 with low interference because of more frequency resource. 1*3*3 with high sector capacity because of low interference.
Average sector capacity comparison
SINR The SINR is not specifically defined in 3GPP specifications. A common formula is as follows: SINR = S/(I + N) S: indicates the power of measured usable signals. Reference signals (RS) and physical downlink shared
channels (PDSCHs) are mainly involved. I: indicates the power of measured signals or channel interference signals from other cells in the current
system and from inter-RAT cells. N: indicates background noise, which is related to measurement bandwidths and receiver noise
coefficients. Empirical SINR at the edge of a cell:
The SINR is greater than -3 dB in 99% areas in Norway. The SINR is greater than -3 dB in 99.25% areas in the Huayang field in Chengdu.
Signal Quality (SINR is mainly involved)
Site selection Cell layout Frequency plan
Antenna height
Antenna azimuths Antenna tilts
Resolving Signal Quality Problems Caused by Improper Parameter Settings Optimizing frequencies
Adjusting the antenna system
Adding dominant
coverage
Change and optimize frequencies based on drive test and performance measurement data.
Adjust antenna azimuths and tilts to change the distribution of signals in an interfered area by increasing the level of a dominant sector and decreasing levels of other sectors.
Increase power of a cell and decrease power of other cells to form a dominant cell.
Decrease RS power to reduce coverage if the antenna pattern is distorted because
Case: Adjusting Antenna Azimuths and Tilts to Reduce Interference
Symptom Cross coverage occurs at sites 1, 2, 3, 7, 8, 9, 10, 11, and 12, and co-channel interference occurs in many areas. Analysis According to the analysis of engineering parameters and drive test data, cell density is large in coverage areas. Coverage by each cell can be reduced by adjusting antenna azimuths and tilts. Solution Change the tilt in cell 28 from 2 degrees to 4 degrees so that the direction points to a demonstration route. Change the tilt in cell 33 f rom 3 degrees to 6 degrees so that the direction points to the Wanke Pavilion. Change the tilt in cells 50 and 51 from 3 degrees to 6 degrees so that the direction points to the Communication Pavilion. Decrease the transmit power in cell 33 by 3 dB to reduce its interference to overhead footpaths near China Pavilion.
Poor signal quality before
Case: Changing PCIs of Intra-frequency Cells to Reduce Interference
Symptom Near Japan Pavilion, UEs access a cell whose PCI is 3 and SINRs are low. UEs are about 200 m away from the eNodeB. This problem may be caused by co-channel interference.
Analysis This problem is not caused by co-channel interference because no neighboring cell has the same frequency as the current cell. Cell 6 interferes with cell 3. SINRs increase after cell 6 is disabled. In theory, staggered PCIs can reduce interference.
Solution Change PCI 6 to PCI 8. Test results show that SINRs increase by about 10 dB.
Case: Handover Failure Caused by Severe Interference
Symptom During a test, handovers from PCI 281 to PCI 279 fail.
Analysis Cell 281 is a source cell and is interfered by cells 279 and 178. Delivered handover commands always fail and cannot be received correctly by UEs. Cell 279 is a target cell for handover, and its coverage is not adjusted preferentially because the signal strength in the handover area can ensure signal quality after handovers. Therefore, cell 178 must be adjusted to reduce its interf erence to cell 281.
Solution Adjust antenna tilts to decrease coverage by cell 178.
SINR Improvement INITIAL PLAN
AFTER ACP
In the inner city of Jakarta where ZTE antenna configuration taken into the initial planning show there are so much SINR around 0~5 (dB). After do the ACP Optimization the SINR much improve with
Initial Plan
After ACP
Radio Parameter @ GENEX Probe PCI (Physical Cell Identifier) Value range : 0 – 839, cross-check any cross feeder problem when conducting moving test. RSRP (Reference Signal Receive Power) → Good -70 dBm to -90 dBm → Normal -91 dBm to -110 dBm -110 dBm to -130 dBm → Bad SINR (Signal to Interference+Noise Ratio)
16 dB to 30 dB
→ Good
1 dB to 15 dB
→ Normal
-10 dB to 0 dB
→ Bad
Radio Parameter @ GENEX Probe…cont Modulation Coding Scheme
64 QAM → Good 16 QAM → Normal QPSK
→ Bad
Neighboring cell
On-Site Hardware
MIMO Antenna
BBU : Baseband Unit
Signal quality overview plot (Serving PCI) RNO-1
Signal quality overview plot (RSRP)
Signal quality overview plot (SINR)
Signal quality overview plot (DL Throughput)
Signal quality overview plot (UL Throughput)
