RF LTE Planning and Dimensioning
RF LTE Planning and Dimensioning
RF LTE Planning and Dimensioning
Dimensioning
Simulation, Lab
1 and Field Findings
Performance
Simulation
Simulation, Lab
1 and Field Findings
Performance
Simulation
X2
MME S10
S7 Rx+
PCRF
S11
S1-U S5/S8 SGi
PDN
LTE-Uu Serving PDN
Evolved Node B
Gateway Gateway
LTE-UE (eNB)
SAE
Gateway
peakrate DL/UL 10/5 Mbps 50/25 Mbps 100/50 Mbps 150/50 Mbps 299/75 Mbps 301/50 Mbps 301/102 Mbps 3000/1500 Mbps
MIMO DL Optional 2x2 2x2 2x2 4x4 2x2 or 4x4 2x2 or 4x4 8x8
- At this phase, the operator provides VoIP over LTE access and IMS is used as
enabling SIP session control machinery for VoIP. However, as shown in it is assumed
that LTE coverage is not yet complete and thus interworking with underlying legacy access
technology is required. From the voice traffic perspective this implies handing over LTE VoIP call to
CS voice call provided by the legacy access technology. The handover functionality from VoIP to
CS domain is referred to as Single Radio Voice Call Continuity (SRVCC).
Simulation, Lab
1 and Field Findings
Performance
Simulation
Key aspects in
3GPP Rel.10 Carrier Aggregation
Carrier Bandwidth extension by carrier aggregation
.. Downlink: Up to 100 MHz bandwidth with 2 Release 8 carriers from
Carrier1 Carrier2 Carrier n different frequency bands
8x MIMO 4x Uplink: Only single band carrier aggregation
New codebook for downlink (DL) 8TX MIMO
Feedback enhancements for DL 2TX/4TX Multiuser MIMO
Coordinated Multipoint
2TX/4TX Uplink Single/Multiuser MIMO
Coordinated multipoint transmission (CoMP), also
known as cooperative system
Receiving transmission from multiple sectors
(not necessary visible for UE)
Single Relay Node architecture based on self-backhauling eNB
Simple intercell interference coordination in time domain
Enhancements for office Femto handovers
1.5Gbps
8x MIMO 4x
Coordinated Multipoint
Cooperation of antennas of multiple
sectors / sites
Interference free
by coordinated
transmission /
reception
Highest
performance
potential Service Area
Simulation, Lab
1 and Field Findings
Performance
Simulation
f f f f
t t t t
f f f f
OFDM is the state-of-the-art and most efficient and robust air interface
User 1 User 2 User 3 User 4
21 2017 Nokia Solutions and Networks. All rights reserved.
Air Interface: OFDM Basics
Transmits hundreds or even thousands of separately modulated radio
signals using orthogonal subcarriers spread across a wideband channel
OFDMA
allocation offered by OFDM.
SC-FDMA
It can reduce the PAPR between 69dB compared to
OFDMA
TS36.201 and TS36.211 provide the mathematical
description of the time domain representation of an SC-
FDMA symbol.
- Reduced PAPR means lower RF hardware requirements
(power amplifier)
Downlink Uplink
12 subcarriers
In both the DL & UL direction, data is
.. ..
allocated to users in terms of
Frequency resource blocks (RBs).
Resource 1 ms subframe a RB consists of 12 consecutive
block or TTI subcarriers in the frequency domain,
reserved for the duration of 0.5 ms
0.5 ms slot slot.
Time The smallest resource unit a
During each TTI,
scheduler can assign to a user is a
resource blocks for
scheduling block which consists of
different UEs are
two consecutive resource blocks
scheduled in the
eNodeB
0 1 2 3 4 5 6 0 1 2 3 4 5 6
180
FDD Frame structure ( also called Type 1 Frame) is common to both UL & DL
Divided into 20 x 0.5ms slots
Structure has been designed to facilitate short round trip time
- Frame length = 10 ms
0.5 ms slot
- FDD: 10 sub-frames of 1 ms for UL & DL
sy sy sy sy sy sy sy
0 1 2 3 4 5 6 - 1 Frame = 20 slots of 0.5ms each
10 ms frame - 1 slot = 7 (normal CP) or 6 OFDM
symbols (extended CP)
s s s s s s s s .. s18 s19
0 1 2 3 4 5 6 7
0.5 ms slot
SF: SubFrame
SF0 SF1 SF2 SF3 .. SF9
s: slot
1 ms sub-frame Sy: symbol
In FDD, there is a time offset between uplink and downlink transmission.
SF SF SF SF
DwPTS
SF SF SF
UpPTS
UpPTS
GP
GP
UL/DL
carrier
#0 #2 #3 #4 #5 #7 #8 #9
SUBFRAME 1
34 2017 Nokia Solutions and Networks. All rights reserved.
Air Interface: Cell-Specific Reference Signals Mapping
R0 R0
R0 R0
For Channel estimation
R0 R0
Measurements (e.g. RSRP)
R0
l =0
R0
l =6 l =0 l =6
Synchronisation in frequency & time
Cell Id (Physical Layer Cell Identity)
R0 R0 R1
R0 R0 R1 R1 R2 R3
R0 R0 R1 R1 R2 R3
R0 R0 R1 R1 R2 R3
R0 R0 R1 R1 R2 R3
l =0 l =6 l =0 l =6 l =0 l =6 l =0 l =6 l =0 l =6 l =0 l =6 l =0 l =6 l =0 l =6
Reference Signals
20 slots = 10 subframes = 10 ms = 10 TTI
Synchronisation Signals
72 subcarriers
Check Online
http://dhagle.in/LTE
PDCCH
72 subcarriers
PHICH
PDSCH
Reference Signals
Simulation, Lab
1 and Field Findings
Performance
Aspects
Assumption:
All Technologies using the same power 20Watt
The same Value for PL(gsm)=PL(umts)=PL(lte)=140db*
Assumption
Operating Band: 1800Mhz (FDD) UE Power Class : Class 3 (23dBm, h=1.5 meter)
Channel Bandwidth: 5 Mhz Cell Load Factor : 50%
Flexi RF Unit: 20Watt (43 dBm) Propagation: Cost 231 TWO slope
Feeder Loss: 0.5 dB (Feederless) Other parameters: using
Antenna Configuration: 2Tx-2Rx MIMO Channel Mode: Enhanced Pedestrian 5Hz(EPA05)
Antenna Gain and Heigh: 18 dBi/30 meter (Dense valid for low speed mobile in General
Urban)
* 8db Std Dev, 93% Coverage Probability, and 20 db Building Penetration Loss for Dense
Urban
44 2017 Nokia Solutions and Networks. All rights reserved.
Uplink Limited
Air Interface Dimensioning: Impact of cell-edge throughput
* 8db Std Dev, 93% Coverage Probability, and 20 db Building Penetration Loss for Dense
Urban
* 8db Std Dev, 93% Coverage Probability, and 20 db Building Penetration Loss for Dense
Urban
Special Sub-frame 5
* 8db Std Dev, 93% Coverage Probability, and 20 db Building Penetration Loss for Dense
Urban
50 2017 Nokia Solutions and Networks. All rights reserved.
Air Interface Dimensioning : LTE-FDD1800 vs GSM1800 Coverage
* 8db Std Dev, 93% Coverage Probability, and 20 db Building Penetration Loss for Dense
Urban
TDD2300 with different DL:UL Ratio Configuration TDD2600 with different Channel Bandwidth
* *
* *
* 8db Std Dev, 93% Coverage Probability, and 20 db Building Penetration Loss for Dense
Urban
52 2017 Nokia Solutions and Networks. All rights reserved.
Table of contents
Simulation, Lab
1 and Field Findings
Performance
Simulation
customer confidential
55 2017 Nokia Solutions and Networks. All rights reserved.
Baseband Capacity Dimensioning: Cell Throughput
Peak throughput
Maximum FSMD / FSME / FSMF / Lite HW Capacity with 2 TX MIMO and
2 RX
Peak L1 UL Throughput
12 25 38 50
per cell [Mbit/s] *
customer confidential
57 2017 Nokia Solutions and Networks. All rights reserved.
Table of contents
Simulation, Lab
1 and Field Findings
Performance
Simulation
The E-UTRAN Cell Global Identifier (ECGI) is specified within 3GPP TS 36.300. It is used to identify
cells globally, and is constructed from the MCC, MNC and E-UTRAN Cell Identifier (ECI). The ECI is
used to identify cells within the PLMN. The MCC, MNC and ECI are broadcast within SIB 1. The ECI
has a fixed length of 28 bits and contains the eNodeB Identifier (eNB-ID).
ECGI
eNB-ID Cell-ID
Id Range
eNB id 0 to1048575
Cell id 0 to 255
Simulation, Lab
1 and Field Findings
Performance
Simulation
100
L TE 2 0 M H z
90 L TE 1 0 M H z
L TE 5 M H z
80 HSDPA 5 M Hz
70
60
Median (50%) data rate over
Mbps
10
100
90
80
70
60
UE
dB/Mbps
PCTel S-SCH
50 PCTel RS
R&S S-SCH
Throughput
40
30
Big variance on SINR
20
measurements,
depending on:
10
Measurement method
0
1 15 29 43 57 71 85 99 113 127 141 155 169 183 197 211 225 239 253 267 281 295 309 323 337 351 365 379 Measurement
sec
equipment
Both cell in idle state Download started in Download started in Download stopped in Neighbor cell
(no traffic) the serving cell the neighbor cell the neighbor cell shut down
Original PCID
settings
PCID 31 -> 33
PCID 32 cell shut down
Interference margin IM
Interference Margin can be defined as a relation between signals received with & without interference
S/N
IM =
S /( I own + I other + N )
100% orthogonality could be assumed in UL & DL due to OFDM & SC-FDMA so that the Intra-cell
interference is close to zero
The only interference which counts is the Inter-cell interference
DL Interference Margin could be derived analytically
IM as a function of Neighbour Cell Load for different MCS and cell Edge User
Throughputs