UNIVERSITY OF ÇUKUROVA INSTITUTE OF NATURAL AND APPLIED SCIENCE MSC THESIS İskender KARSLI IMPORTANCE OF MIMO TECHNIQUE FOR HSPA AND LTE NETWORKS AND EMPRICAL COMPARISONS FOR A DEFINED ROUTE DEPARTMENT OF ELECTRICAL AND ELECTRONICS ENGINEERING ADANA, 2013
96
Embed
UNIVERSITY OF ÇUKUROVA INSTITUTE OF …library.cu.edu.tr/tezler/8889.pdf · CPICH RSCP and CPICH EcNo Distribution..... 70 Figure 7.21. Physical Layer Served Thr and HS UL EDCH Throughput
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
UNIVERSITY OF ÇUKUROVA INSTITUTE OF NATURAL AND APPLIED SCIENCE
MSC THESIS İskender KARSLI IMPORTANCE OF MIMO TECHNIQUE FOR HSPA AND LTE NETWORKS AND EMPRICAL COMPARISONS FOR A DEFINED ROUTE
DEPARTMENT OF ELECTRICAL AND ELECTRONICS ENGINEERING
ADANA, 2013
ÇUKUROVA ÜNİVERSİTESİ FEN BİLİMLERİ ENSTİTÜSÜ
IMPORTANCE OF MIMO TECHNIQUE FOR HSPA AND LTE
NETWORKS AND EMRICAL COMPARISONS FOR A DEFINED ROUTE
İskender KARSLI
YÜKSEK LİSANS TEZİ
ELEKTRİK – ELEKTRONİK ANABİLİM DALI
Bu Tez …./…./2013 tarihinde Aşağıdaki Jüri Üyeleri Tarafından Oybirliği /
Oyçokluğu ile Kabul Edilmiştir.
……………………….. ………………………. ..…………………… Prof. Dr. Hamit SERBEST Prof. Dr. Turgut İKİZ Assoc. Prof. Dr. Ali AKDAĞLI DANIŞMAN ÜYE ÜYE Bu Tez Enstitümüz Elektrik-Elektronik Anabilim Dalında Hazırlanmıştır.
Kod No:
Prof. Dr. Selahattin SERİN Enstitü Müdürü Not: Bu tezde kullanılan özgün ve başka kaynaktan yapılan bildirişlerin, çizelge, şekil ve fotoğrafların
kaynak gösterilmeden kullanımı, 5846 Sayılı Fikir ve Sanat Eserleri Kanunu’ndaki hükümlere tabidir.
I
ÖZ
YÜKSEK LİSANS TEZİ
Abazar TAJADDODCHELİK
İskender KARSLI
ÇUKUROVA ÜNİVERSİTESİ FEN BİLİMLERİ ENSTİTÜSÜ
ELEKTRİK-ELEKTRONİK ANABİLİM DALI
Danışman : Prof. Dr. Hamit SERBEST Yıl : 2013, Sayfa: 79 Jüri : Prof. Dr. Hamit SERBEST : Prof. Dr. Turgut İKİZ : Assoc. Dr. Ali AKDAĞLI GPRS (General Packet Radio Service) ve EGPRS (Enhanced Data rates for GSM Evolution) data servislerinin ilk formudur. 3GPP (3G Partnership Project) 1998 yılında 2G’nin evrimleşmiş hali olan 3G mobil sistemin standartların hazırlanmasını amacıyla kurulmuştur. Çalışmaların sonucunda Mayıs 2001’de ilk 3G/W-CDMA şebekesi servise verilmiştir. Geliştirmeler yüksek data iletimini sağlayan HSPA (High-Speed Packet Access) standardı ile devam etmiştir. 3GPP sürekli artan data taleplerini karşılamak için, 2004 yılında LTE (Long Term Evolution) adı verilen yeni bir standart geliştirmek üzere bir yol haritası oluşturmuştur. 2009 Aralık ayında ise ilk ticari LTE şebekesi Stokholm ve Oslo’da servise verilmiştir. HSPA ve LTE ile birlikte en önemli özelliklerden biri smart antenlerin kullanımıdır. MIMO (Multiple Input Multiple Output) zaman zaman smart anten de denilir, operatörlerin kapasitelerini arttırabilmesi için yeni bir teknik olarak ortaya çıkmaktadır. MIMO tekniğinde reciever ve transmitter tarafında çoklu anten kullanımı vardır. MIMO’nun 2 modu vardır; Diversity Space Time Coding ve Spatial Multiplexing. İlk modda TX/RX Diversity kullanılarak aynı data 2 antenden farklı kodlarla veri iletilir. TX/RX Diversity kullanımına bağlı olarak hücre kapsaması iyileşir. 2. modda ise 2 antenden farklı datalar gönderilir yada alınır. Bu da belli bir zaman diliminde indirilen data miktarını 2’ye katlar. Bu tezin amacı MIMO tekniğinin kapsama ve throughput üzerine olan etkisini araştırmaktır. Tezin sonunda MIMO tekniğinin LTE sistemler için önemi test edilecek ve başka bir test setup vasıtası ile şu anda en iyi throughput’lardan birini sunan HSPA Dual Carrier teknolojisi ile kıyaslamalar yapılacaktır. Anahtar Kelimeler: MIMO, LTE, 3GPP, HSPA
MIMO TEKNİĞİNİN HSPA VE LTE ŞEBEKELERİ İÇİN ÖNEMİ VE TANIMLI BİR ROTA İÇİN DENEYSEL OLARAK KIYASLANMASI
II
ABSTRACT
MSc THESIS
IMPORTANCE OF MIMO TECHNIQUE FOR HSPA AND LTE NETWORKS AND EMPRICAL COMPARISONS FOR A DEFINED
ROUTE
İskender KARSLI
ÇUKUROVA UNIVERSITY INSTITUTE OF NATURAL AND APPLIED SCIENCES
DEPARTMENT OF ELECTRICAL AND ELECTRONICS ENGINEERING Supervisor : Prof. Dr. Hamit SERBEST Year: 2013, Pages:79 Jury : Prof. Dr. Hamit SERBEST : Prof. Dr. Turgut İKİZ : Assoc. Dr. Ali AKDAĞLI GPRS (General Packet Radio Service) and EGPRS (Enhanced Data rates for GSM Evolution) is the first form of data services. 3GPP (3G Partnership Project) aimed to prepare standards for a 3G Mobile system based on evolved GSM when it was established in 1998. At last the first 3G/W-CDMA network was launched in May 2001. Developments continued with HSPA (High-Speed Packet Access) supporting high-speed data transmissions. To meet continiously increasing data demands, 3GPP initiated a roadmap to develop a new standard which is called as LTE (Long Term Evolution) in 2004. Now the world’s first commercialized LTE networks launched in Stockholm and Oslo in December 2009. One of the most important specification coming with HSPA and LTE standard is the usage of smart antennas. MIMO (Multiple-Input Multiple-Output) which is called sometimes smart antenna, arises as a new technique that provides operators to increase their capacity with a cost-effective way. MIMO technique uses multiple antennas in receiver and transmitter. It has two modes; Diversity Space Time Coding and Spatial Multiplexing. In the first mode MIMO transmits the same data stream from two antennas but with different codes to improve cell coverage using Tx/Rx diversity. In the 2nd mode A different data stream is sent or received by each antenna, this led to double the amount of data for a given time. The purpose of this thesis is to investigate the effect of MIMO technique on coverage and also throughput. At the end of the thesis, the importance of MIMO technique will have been experimented for LTE networks and it has been made comparisons with HSPA Dual Carrier technology which gives one of the best throughput values nowadays by using another drive test setup. Keywords: MIMO, LTE, 3GPP, HSPA
III
ACKNOWLEDGEMENTS
I would like to thank my supervisor Prof. Dr. Hamit SERBEST for his
interests, supports, encouragements and primely trust me.
I would like to thank also my wife Derya and my doughter Beren for their
unlimited support and their patience.
IV
CONTENTS
ÖZ .............................................................................................................................I
ABSTRACT ............................................................................................................ II
ACKNOWLEDGEMENTS .................................................................................... III
CONTENTS ........................................................................................................... IV
LIST OF TABLE .................................................................................................... VI
LIS OF FIGURE.................................................................................................. VIII
LIST OF ABBREVIATONS .................................................................................XII
Simple presentation of cyclicly shifted subcarriers are shown in figure 6.3
below. Note that CP is also applied to each OFDM symbol before transmitting.
S1 S2 S3 S4
0 1 2 3 Ant03 0 1 2 Ant12 3 0 1 Ant21 2 3 0 Ant3
Figure 6.3. Cyclicly Shifted Subcarriers
6.2.2. Space Frequency Block Coding (SFBC)
In LTE, transmit diversity is implemented by using Space Frequency Block
Coding (SFBC). SFBC is a frequency domain adaptation of Space-time Block
Coding (STBC) also known as Alamouti coding where encoding is done in
6. MIMO (MULTIPLE INPUT MULTIPLE OUTPUT) İskender KARSLI
48
antenna/frequency domains rather than in antenna/time domains. In this approach
during any symbol period, two data symbols are transmitted simultaneously from the
two transmit antennas. In this scheme, the two symbols are sent on two different
frequencies, for example, on different subcarriers in an Orthogonal Frequency
Division Multiplexing (OFDM) system as shown in Figure 6.4. Let represent
the signal transmitted from the pth antenna on the kth subcarrier, then:
Figure 6.4. STBC and SFBC transmit diversity schemes for 2-Tx antennas. (Khan,
2009)
6.3. Spatial Mutiplexing
In the previous section, we examined how multiple antennas can be used to
provide the diverstiy gain. The transmission diversity improves the link performance
in the case of high mobilty. Transmit diversity is also useful for delay-sensitive
services. Since only a single data stream is always transmitted in transmit diversity, it
has no effect on data rates to reach peak values. Multiple transmission antennas
usage in eNB together with multiple receiver antennas at UE can be used to achieve
higher peak data rates by enabling multiple data stream transmission between eNB
and UE by using MIMO spatial multiplexing mode. The MIMO spatial multiplexing
6. MIMO (MULTIPLE INPUT MULTIPLE OUTPUT) İskender KARSLI
49
also improves cell capacity and throughput since UEs in good channel conditions can
use multiple streams transmissions. Spatial multiplexing is a special multiplexing
technique where different data streams are transmitted and recieved through several
independent (spatial) channels by using multiple antennas. As a result the higher the
number of antennas, the higher the number of data transmission rate. Main benefits
of spatial multiplexing;
• It is not needed additional power
• There is no additional bandwidth requirement
Of course, as in many areas of science, there is a theoretical limits in amount
of data that can be passed along a specific channel in the presence of noise. The
maximum amount of data that can be carried by a radio channel is limited by the
physical boundaries defined under Shannon's Law. Shannon's law defines the
maximum rate at which error free data can be transmitted over a given bandwidth in
the presence of noise. It is usually expressed in the form;
Where C is the channel capacity in bits per second, W is the bandwidth in
Hertz, and S/N is the SNR (Signal to Noise Ratio).
From this it can be seen that there is a limit on the capacity of a channel with
a given bandwidth. And the capacity is also limited by the signal to noise ratio of the
received signal. So a suitable way to increase the channel capacity is looking as
modulation scheme decision. The channel capacity can be increased by using higher
order modulation schemes, but these of course require a better signal to noise ratio.
Thus it can be said that, a balance exists between the data rate and the allowable error
rate, signal to noise ratio and power that can be transmitted.
As we mentioned to take advantage of the additional throughput capability,
MIMO utilizes multiple antennas in both transmitter and reciever part. In many
MIMO systems, just two antennas are used, but it is possible to use more antennas to
6. MIMO (MULTIPLE INPUT MULTIPLE OUTPUT) İskender KARSLI
50
increase throughput. Note that, in any case for MIMO spatial multiplexing the
number of receive antennas must be equal to or greater than the number of transmit
antenna . To take advantage of the additional throughput offered, MIMO wireless
systems utilize a matrix mathematical approach. Therefore A MIMO channel
consists of channel gains and phase information for links from each of the
transmission antennas to each of the receive antennas as shown in Figure 6.5. Then,
the channel for the M × N MIMO system consists of an N × M matrix given
as:
where represents the channel gain from transmission antenna j to the
receive antenna i. (Khan, 2009)
Figure 6.5. MxN Spatial Multiplexing
Consider 3x3 spatial multiplexing system. Then , , are transmitted the
data streams from antennas 1, 2, and 3. By this idea recieved signal streams , ,
are then;
6. MIMO (MULTIPLE INPUT MULTIPLE OUTPUT) İskender KARSLI
51
Where = signal received at antenna 1, is the signal received at antenna 2.
Then we can write a general formula for recieved signal;
To recover the transmitted data-stream at the receiver it is necessary to
perform a considerable amount of signal processing. First the MIMO system decoder
must estimate the individual channel transfer characteristic to determine the
channel transfer matrix. then the matrix can be produced and the transmitted data
streams can be reconstructed by multiplying the received vector with the inverse of
the transfer matrix.
6.4. Radio Configurations
In this section, it is given genaral explanations of radio configuratiton types
for each test cases and also properties of main units used in these configurations are
explained briefly. Test equipments and terminals are also included in the
explanations.
6.4.1. MIMO and SIMO
As explanined before a real LTE network has been used for LTE tests. The
network consists of cells with Ericsson RBS 6201s cabinet which is a new generation
cabinet type in Ericsson 6000 series family. It provides a flexible structure to
configure the cabinet with other technologies which are GSM, WCDMA and also
LTE at the same time. It consists of two radio shelves and each radio shelves can
6. MIMO (MULTIPLE INPUT MULTIPLE OUTPUT) İskender KARSLI
52
house up to 6 Radio Units (RU). Each RU has a 20 Mhz bandwidth and up to 60 W
of output power which is adjusted in steps of 20 W. The main functions of RU’s are;
• Transceiver (TRX)
• Transmitter (TX) amplification
• Transmitter/Reciever (TX/RX) Duplexing
• TX/RX filtering
Number of possible Radio Configurations are listed in following Table 6.1. In
our thesis first two configurations have been applied.
Table 6.1. Possible Radio Cofigurations Configuration No of Radio Units Output Povver 3x20MHzMIMO 6 20+20 or 40+40 or 60+60 3x20 MHzSIMO 3 20 or 40 or 60 6x20 MHz 6 60 6x20 MHzMIMO 12 60 + 60
Simple hardware representation of MIMO and SIMO configuration has been
shown in figure 6.6. From the table1 in MIMO configuration there are 2 Radio Units
used in each cell and totally 6 RU’s in each site. Since it is used 2 Transceiver Radio
units in our cells we can say our system as 2*N MIMO system. On the other hand for
SIMO confıguration it requires 1 RU for each cells and totally 3 RUs in a site. But
please don’t forget, only one cell is activated and other cells are halted during the
tests.
Figure 6.6. Simple Hardware Representation of MIMO and SIMO configurations
6. MIMO (MULTIPLE INPUT MULTIPLE OUTPUT) İskender KARSLI
53
6.4.2. Dual Carrier
DC HSPA (Dual Cell HSPA) test has been carried out in a real Ericsson 3G
HSPA network that consists of cells with 6000 series cabinets. DC-HSDPA is a new
feature that comes with 3GPP Release 8. Major benefits of this feature is;
• Doubling DL Throughput with a peak throughput of 42Mb/s within the
cell
• Increase of HSDPA coverage
• Increase cell capacity
We know that the current bandwidth in UMTS/HSPA is 5 MHz. In Release 8
downlink, it is possible to increase data rates using either a combination of MIMO
and 64QAM or dual-cell HSDPA for operation on two 5MHz carriers with 64QAM,
data rates reach up to 42Mbps. In our thesis we used dual-cell HSDPA for the test
case. Dual cell approach provides higher throughput rates by combining two adjacent
5 MHz carriers as shown in figure 6.7 below. In this configuration, it is possible to
achieve a doubling of the 21 Mbps maximum rate available on each channel to 42
Mbps with a less expensive infrastructure upgrade. Since it doubles the user
throughout, we can say that this feature also increases cell capacity. Dual Cell
operation looks cheaper way than the usage of MIMO. Because MIMO needs
additional hardware usage which effects costs.
Figure 6.7. Dual Carrier Operation
6. MIMO (MULTIPLE INPUT MULTIPLE OUTPUT) İskender KARSLI
54
Dual-Cell assumes that both the 5MHz bands are adjacent. If they are not
adjacent then the better term to refer for DC is Dual-Carrier. A dual-carrier user can
be scheduled in the primary serving cell as well as in a secondary serving cell with
two parallel HS-DSCH transport channels as shown in figure 6.8. One of the carrier
can be configured as primary serving cell.
Figure 6.8. Parallel Operation of Dual Carrier HSPA
All physical layer procedures and non HSDPA related channels are included
in the primary serving cell. As a consequence, the dual-carrier feature also provides
an efficient load balancing between carriers in one sector. In Dual carrier
configuration, two transport blocks can be transmitted on their respective cells using
a different number of channelization codes as shown in the figure 6.7. We know
from the UMTS knowledges, if a UE is served by two cells of same node-b, we call
this event as softer handover. DC case can be accepted similar to this event but there
are minor differences in both Layer 2 and physical layer design. There are two
HARQ processes per TTI for dual carrier transmission/reception. So, at the physical
layer, dual carrier transmission can be tought as independent transmissions over 2
HS-DSCH channels, each having associated downlink and uplink signalling.
Figure 6.10 shows the architecture of MC-HSDPA. There are two
connections established for UE, one is serving cell and the other one is secondary
cell as shown in the figure. Notice that, in DL, two carriers are used while in UL
direction only one carrier is used for transmission.
6. MIMO (MULTIPLE INPUT MULTIPLE OUTPUT) İskender KARSLI
55
Figure 6.9. MC-HSDPA Architecture Overview
Physical layer structure can be seen from the figure 6.11 below. Primary
serving cell has all the physical channels, including DPCH/F-DPCH, EHICH , E-
AGCH, and E-RGCH. Secondary HS-DSCH cell functions as suplemantary cell
which have CPICH, HS-SCCH and HS-PDSCH physical channels. Please note that
in UL there is only used cell is primary serving cell.
Figure 6.10. Physical Channel Configuration
6. MIMO (MULTIPLE INPUT MULTIPLE OUTPUT) İskender KARSLI
56
Physical channels in Downlink;
• P-CPICH - Primary Common Pilot Channel
• DPCH (Dedicated Physical Channel) One per UE
• HS-SCCH High Speed Shared Control Channel
• HS-PDSCH High Speed Physical Downlink Shared Channel
• E-HICH
• E-AGCH
Physical channels in Uplink;
• DPCCH - Dedicated Physical Control Channel
• HS-DPCCH High Speed Dedicated Physical Control Channel
• E-DPCCH
• E-DPDCH
6.5. Test Terminals
2 types of terminals have been used in the tests. These terminals are;
• Samsung 4G USB modem (GT-B3710)
Samsung 4G USB modem is a MIMO compatible modem which is used in
Test Case 1 and Test Case 2. The modem supports 2.6GHz band for LTE service.
Samsung’s LTE solution is fully compliant with the latest 3rd Generation Partnership
Project (3GPP) LTE Release 8 (Rel-8) standard and its LTE UE category is CAT 3.
Table 6.2 shows the peak data rates according to LTE UE categories. From the table
since our modem is category 3 peak downlink data rate is 102.048Mb/s, uplink is
51.024Mb/s and it does not support 64 QAM in UL direction.
6. MIMO (MULTIPLE INPUT MULTIPLE OUTPUT) İskender KARSLI
E372 Huawei support 42 Mb/s HSPA dual carrier downlink and 11 Mb/s
uplink services. E372 has been used in test case 3. It support maximum 15 HS-
DSCH codes. It’s HSPA UE category is 24 and 3gpp release 8.
Table 6.3. List of HSPA UE Categories (Q: QPSK, 16: 16QAM, 64: 64QAM) based on downlink performance 4G Americas, 2011b)
6.6. Test Tools
TEMS investigations 11.0.2 has been used as a test tool for all tests. TEMS is
a well known tool which is a product of ASCOM and TEMS investigations 11.0.2
version is released on July 23, 2010. It supports GSM, HSPA and LTE technologies.
6. MIMO (MULTIPLE INPUT MULTIPLE OUTPUT) İskender KARSLI
58
7. TEST CASES İskender KARSLI
59
7. TEST CASES
One of the purpose of this thesis is to explain the actual reason of throughput
increase if it is related with technology change or MIMO usage. To explain this, It
has been investigated the effect of MIMO usage in a real LTE system. At the same
time It has been also carried out another test scenario in a real HSPA network to
make a comparison with LTE cases. Test scenarios have been carried out in 3 cases.
• Test Case 1: Multiple Input Single Output (MIMO)
• Test Case 2: Single Input Single Output (SIMO)
• Test Case 3: HSPA Dual Carrier
To carry out the test scenairos it has been selected an industrial region
environment and taken outdoor drive test measurements for a fixed predefined route.
The reason for selecting the industrial environment is to recieve better signal strength
for having better throuhput since the spaces between buildings are much more than
other environments. During test, it has been especially cared about to take all signal
strength ranges to observe the performace efficiently. In Results and Analysis
chapter, signal strengths and throughput results have been concluded and compared
in a table for each test cases. Physical layer values of the cases have been calculated
and compared with each other for analysis.
Figure 7.1. Test Route
7. TEST CASES İskender KARSLI
60
7.1. Test Case 1: Multiple Input Multiple Output (MIMO)
MIMO is a multiple antenna technique to enhance throughput or coverage of
a cell in a mobile system. There are two modes of MIMO which are spatial
multiplexing and transmit diversity as explained in previous chapters. Since the main
subject always emphasised in broadband market is high throughput while subscribers
are in mobile state so it is more proper to test MIMO in spatial multiplexing mode. In
this thesis LTE cell is configured as 2x2 spatial multiplexing mode of MIMO.
7.1.1. Materials and Configurations for Test Case 1
In this case, data is transmitted from two TX antennas and recieved by two
RX antennas as shown in the figure 7.1. TX part represents the base station side RX
side represents terminal side in the figure. It has been used an 18 dBi gain cross
polarized antenna which has a horizontal beamwidth of 58 degrees and 6.2 degrees
of vertical beamwidth in the cell. Other feeder and antenna properties are given in the
Table 7.1.
Figure 7.2. 2x2 MIMO Configuration
7. TEST CASES İskender KARSLI
61
Table 7.1. Antenna and Feeder Spesifications Antenna And Feeder SpesificationsAntenna Height 23 (m)Antenna Type Kathrein 800 10541 Antenna Frequency Range 2300–2690 MHzAntenna Gain 18 dBiImpedance 50 ohmsPolarization +45° and -45°Front-to-back ratio >25 dB Horizontal beamwidth 2490–2690 MHz 58 degrees (half power)Vertical beamwidth 2490–2690 MHz 6.2 degrees (half power) Antenna Tilt 3Azimuth 0Feeder Type 7/8" 50 ohmsFeeder Length 12 (m)
In MIMO case, it has been utilized 2 RUs for 2*2 MIMO configuration and
remember that an RU has a 20Mhz bandwidth and up to 60W of power. Since there
is only one subscriber within the cell and we desire to take maximum throughput, it
will be dedicated one subcarrier with 20Mhz bandwidth to only one subscriber to
take maximum throughput as shown in Figure 7.3.
Figure 7.3. One subcarrier with 20Mhz bandwdith
Table 7.2. RU Configuration Radio Unit Configurations for MIMORadio Configuration for MIMO 1×20 MHz *Number of Radio Units 2Output power (W) 40+40TX Frequency 2620-2640 (Mhz)RX Frequency 2500-2520 (Mhz) * One cell is active and other cells are halted during the tests.
7. TEST CASES İskender KARSLI
62
7.1.2. Test Results
Reference Signal Recieved Power (RSRP) and Physical Layer PDSCH
Throughput measured by the terminal is plotted in figure 7.4. It can be easily
concluded that there is a direct correlation between RSRP and throughput. For low
RSRP samples it has been measured low throughput. But it is very exciting that even
in the values below -100dB’s, the throughput is still maintain above 30Mb/s.
Figure 1.25. Serving Cell RSRP Plot for MIMO Case
Figure 7.4. RSRP vs PDSCH Phy Throughput Plot for MIMO Case
It can be seen from the figures below that PDSCH throughput values of
MIMO case is accumulated between 35Mb/s and 70Mb/s which is 93.8% of all
samples. And it is seen that the MIMO configured cell has served better than 70Mb/s
in 5.8 percent of all test.
Figure 7.5. Serving cell RSRP and PDSCH Phy distribution for MIMO case
7. TEST CASES İskender KARSLI
63
During the MIMO test, 3 types of modulation scheme has been used in DL
direction on the other hand 2 types of modulation scheme have been used in UL
direction. The left of the Figure 7.8 shows that majority of the modulation types are
64QAM and 16QAM. It has been recieved QPSK scheme only in small part of the
test which the lowest signal strength of the test. We had emphasised the correlation
between throughput and signal strength. Actually we know that throughut is
depended on used modulation scheme. At majority of the test it has been recieved
64QAM scheme and even in lower signal levels still we have 16QAM scheme as can
be seen from the signal plots below. It tends to a satisfied and sufficient avarege
throughput experience for user. In UL direction the main modulation scheme used is
16QAM except a small quantity of samples.
Figure 7.6. PSDCH and PUSCH Modulation Plot for MIMO Case We know that in MIMO 2 transport blocks (TB0 and TB1) are trasfered
simultaneously. Since they are transfered at same modulation in majority of the test.
It has been only included TB0 plot for analysis as shown in figure 7.8. We know
that;
• QPSK: 1 symbol 2 bits
• 16QAM: 1 symbol 4 bits
• 64QAM: 1 symbol 6 bits
7. TEST CASES İskender KARSLI
64
Distribution of the modulations for PDSCH and PUSCH is given in the table;
Table 7.3. Modulations Distribution for MIMO Test Case.
64QAM 55% 16QAM 83%16QAM 40% QPSK 17%QPSK 5%
DL UL
Figure 7.10 shows that during test duration it has been always used 2
transport blocks. 2 Transport Block usage percentage is 100.
Figure 7.7. Used Transport Blocks in MIMO 7.2. Test Case 2: Single Input Multiple Output (SIMO)
Between MIMO and SIMO hardware configuration, main difference is the
number of Radio Units. Additionally it has been made some parameter changes.
Antenna and feeder spesificatios are all same with SIMO case which is given in table
7.1.
7.2.1. Materials and Configurations for Test Case 2
In SIMO case spatial multiplexing is deactivated. To configure SIMO, some
parametric changes are applied and also one radio unit is deactivated. In SIMO case
only 1 TB is transferred to the terminal but terminal has two RX antenna as shown in
figure 7.11.
7. TEST CASES İskender KARSLI
65
Figure 7.8. SIMO Configuration
In SIMO case, 20Mhz bandwidth is still dedicated to single user to give
maximum throughput. The RU configuration is given in Table 7.4.
Table 7.4. RU Configuration Radio Unit Configurations for SIMORadio Configuration for MIMO 1×20 MHz *Number of Radio Units 1Output power (W) 40TX Frequency 2620-2640 (Mhz)RX Frequency 2500-2520 (Mhz)
7.2.2. Test Results
From the SIMO signal test plots below it seems a negative effect on coverage
compared to the MIMO. It is an expected situation since there is only one TB
transferred. Average RSRP in SIMO is -86 dbm while in MIMO -82.18dbm.
However in SIMO it is transferred one TB instead of two as in MIMO, maximum
throughput is not decreased to half of the maximum throughput in MIMO. Maximum
throughput in SIMO is 49.9Mb/s while in MIMO 76Mb/s.
7. TEST CASES İskender KARSLI
66
Figure 7.9. RSRP vs PDSCH Phy Throughput Plot for SIMO Case
PDSCH Throughput values are accumulated above 35Mb/s. Even in low
signal strengths throughput is not decreased below 17Mb/s.
Figure 7.10. Serving Cell RSRP and PDSCH Phy Distribution for SIMO Case
There are two modulation schemes used in DL which are 64QAM and QPSK
on the other hand in UL used modulation schemes are 16AQM and QPSK as shown
in the figure 7.15. In the region shown in left plot, it is seen that the radio conditions
are not so bad but used modulation scheme is QPSK. Because of that reason it has
been served with low throughput in this region. However it looks like a problem with
test measurements, it doesn’t effect our results since we calculate median values of
each KPI. And the other interesting point is there is not 16QAM modulation used in
PDSCH modulation, even in lowest signal levels it is still continious to serve with
6QAM modulation.
7. TEST CASES İskender KARSLI
67
Figure 7.11. PSDCH and PUSCH Modulation Plot for SIMO Case
Remember that there is only one TB used in SIMO case. And modulation
types disribution for PDSCH and PUSCH is given in te table;
Table 7.5. Modulations Distribution for SIMO test case. DL UL
64QAM 67 % 16QAM 52 %
QPSK 33 % QPSK 48 %
Maximum througput served by the SIMO configured cell is 1,336Mb/s as
shown in figure 7.17. SIMO cell has served better than 1Mb/s in 47 percent of all
test. As a result it seems MIMO has a better UL and DL performance in throughput
and also coverage.
Figure 7.12. Used Transport Blocks in SIMO
7. TEST CASES İskender KARSLI
68
7.3. Test Case 3: HSPA Dual Carrier
In previous sections it is given MIMO and SIMO tests which are carried in a
trial LTE network. As we mentioned before these tests are carried out within a single
cell, all other cells are deactivated during the tests. But in Dual Carrier HSPA case,
test is carried out within a real network. But from the signal plot below shows that
there are there cells served during the test and these cells are belong to same node b.
Figure 7.13. Serving Cells signal plot
Serving Cell Ratios are given as;
SCR 153 : 58.59 %
SCR 161 : 23.14 %
SCR 169 : 18.27 %
7.3.1. Materials and Configurations for Test Case 3
Since there is not only one cell in our test case it is allowed to make
handovers. It is given main serving cell properties in this chapter. In our test
scenario, it has been used Kathrein 742 215 antenna whose horizontal and vertical
pattern characteristics and gain are similar with LTE cases. Note that the antenna
height in DC case is 26m which is 3 meters higher than LTE cases. Other difference
is related with feeder type. Since It is used Ericsson 6601 cabinet as we mentioned
before, there is no feeder loss in DC case. Table 7.6 shows all physical properties of
the cell.
7. TEST CASES İskender KARSLI
69
Table 7.6. Phsical Properties of DC test case. Antenna And Feeder SpesificationsAntenna Height 26 (m)Antenna Type Kathrein 742 215 Antenna Frequency Range 1710–2200 MHzAntenna Gain 18 dBiImpedance 50 ohms Polarization +45° and -45°Front-to-back ratio >25 dB Horizontal beamwidth 1920–2200 MHz 65 degrees (half power)Vertical beamwidth 1920–2200 MHz 6.4 degrees (half power) Antenna Tilt 5Azimuth 0Feeder Type F/O
Table 7.7. Radio Unit Configurations for DC Radio Unit Configurations for DCConfiguration 3*2 (Totally 3 sectors, each sector have 2 carriersNumber of Radio Units 3Output power (W) 30TX Frequency 2136RX Frequency 2982
7.3.2. Test Results
In DC case, physical layer measurements have been analysed as we made in
previous cases. Figure 7.19 shows CPICH RSCP and CPICH EcNo plots. Now, the
serving HSPA cell is not a stand alone cell as in LTE cases, then it will be more
convenient to include the EcNo plot. Because in this case intercell interference is
available. So in this case both CPICH RSCP and EcNo is key factors for throughput.
As can be seen from the figure coverage plot looks better than LTE cases in terms of
signal strength. The reason for that is allowing handovers to other cells. It can be
shown that at lowest signal strenghts EcNo decrases to lowest values. But it can be
easily seen that even in best CPICH RSCP values EcNo might be at worst values
because of the interference issues. Other necessary factor is the network load. As we
mentioned HSPA is a power dependent system. Since each user consumes an amount
of power of the cell, throughput will be effected from the load of the system.
7. TEST CASES İskender KARSLI
70
Figure 7.14. CPICH RSCP and CPICH EcNo Plot
Ideal value of EcNo is accepted around -12 dBm. EcNo figure below looks
the test route is a little bit polluted because 73 percent of the EcNo samples are less
then -12 dBm. It has a negative effect on throughput even if the coverage is
satistified. Median value of RSCP is already calculated as -64.54 dBm.
Figure 7.15. CPICH RSCP and CPICH EcNo Distribution Lets look at the region 1 in the figure 7.21 below, in this region we have good
RSCP samples but bad EcNo values. This causes low physical DL and UL
througputs. If we look at the region 2, we see bad RSCP and bad EcNo values but
throughput values look better than the region 1. It is tought that the system load in
region 2 is more relaxed than region 1.
7. TEST CASES İskender KARSLI
71
Figure 7.16. Physical Layer Served Thr and HS UL EDCH Throughput Plot
HS Physical served median throughput in DL is calculated 10.676 Mb/s and
median throughput in UL is 258 Kb/s. DC HSPA cell has served better than 10Mb/s
in 55 percent of all test. It is seen that in HSPA DC case throughput vales
accumulated between 5Mb/s and 15 Mb/s. And the maximum value taken in this case
is 26.949 Mb/s.
1
2
7. TEST CASES İskender KARSLI
72
8. RESULTS AND ANALYSIS İskender KARSLI
73
8. RESULTS AND ANALYSIS
In this thesis, 3 test cases have been carried out for analysing the performance
effects of the multiple antenna usage in an LTE system together with technology
comparisons with LTE and Dual carriers HSPA networks. All test cases are studied
and analysied in the previous sections in detail. The following table 8.1 gives a result
table that consists of throughput and coverage values of each test cases. And the
latter table gives theorical speeds of each cases.
Table 8.1. Throughput and Coverage Comparisons Between Test Cases LTE MIMO LTE SIMO Dual Carrier