investigating and enhancing performance of multiple antenna systems...
TRANSCRIPT
KTH Electrical Engineering
Investigating and Enhancing Performance of
Multiple Antenna Systems in Compact
MIMO/Diversity Terminals
Shuai Zhang
Doctoral Thesis in Electrical Systems
School of Electrical Engineering
KTH Royal Institute of Technology
Stockholm, Sweden 2013
Division of Electromagnetic Engineering
KTH School of Electrical Engineering
SE – 100 44 Stockholm, Sweden
ISSN 1653-5146
ISBN 978-91-7501-600-9
Akademisk avhandling som med tillstånd av Kungliga Tekniska Högskolan framlägges till
offentlig granskning för avläggande av teknologie doktorsexamen måndagen den 28 januari
2013 klockan 10.00 i sal F3, Kungliga Tekniska Högskolan,
Lindstedtsvägen 26, Stockholm.
© Shuai Zhang , 2013
Tryck: Universitetsservice US AB
To my grandparents…
Abstract
Today, owners of small communicating device are interested in transmitting or receiving
various multimedia data. By increasing the number of antennas at the transmitter and/or the
receiver side of the wireless link, the diversity/Multiple-Input Multiple-Output (MIMO)
techniques can increase wireless channel capacity without the need for additional power or
spectrum in rich scattering environments. However, due to the limited space of small mobile
devices, the correlation coefficients between MIMO antenna elements are very high and the
total efficiencies of MIMO elements degrade severely. Furthermore, the human body causes
high losses on electromagnetic wave. During the applications, the presence of users may
result in the significant reduction of the antenna total efficiencies and highly affects the
correlations of MIMO antenna systems. The aims of this thesis are to investigate and enhance
the MIMO/diversity performance of multiple antenna systems in the free space and the
presence of users.
The background and theory of multiple antenna systems are introduced briefly first.
Several figures of merits are provided and discussed to evaluate the multiple antenna systems.
The decoupling techniques are investigated in the multiple antenna systems operating at the
higher frequencies (above 1.7 GHz) and with high radiation efficiency. The single, dual and
wide band isolation enhancements are realized through the half-wavelength decoupling slot,
quarter-wavelength decoupling slot with T-shaped impedance transformer, tree-like parasitic
element with multiple resonances, as well as the different polarizations and radiation patterns
of multiple antennas.
In the lower bands (lower than 960 MHz), due to the low radiation efficiency and strong
chassis mode, the work mainly focused on how to directly reduce the correlations and enlarge
the total efficiency. A new mode of mutual scattering mode is introduced. By increasing the Q
factors, the radiation patterns of multiple antennas are separated automatically to reduce the
correlations. With the inter-element distance larger than a certain distance, a higher Q factor
also improved the total efficiency apart from the low correlation. A wideband LTE MIMO
antenna with multiple resonances is proposed in mobile terminals. The high Q factors
required for the low correlation and high efficiencies in mutual scattering mode is reduced
with another mode of diagonal antenna-chassis mode. Hence, the bandwidth of wideband
LTE MIMO antenna with multiple resonances mentioned above can be further enlarged while
maintaining the good MIMO/diversity performance.
The user effects are studied in different MIMO antenna types, chassis lengths, frequencies,
port phases and operating modes. Utilizing these usefully information, an adaptive
quad-element MAS has been proposed to reduce the user effects and the some geranial rules
not limited to the designed MAS have also been given.
Key words: Multiple antennas, planar inverted-F antenna, mutual coupling, slot antenna,
compact antenna, mobile antenna, Q factors, antenna array, ultra-wideband (UWB) antenna,
pattern diversity , polarization diversity, user effects, adaptive arrays, chassis mode.
Acknowledgements
This dissertation is a conclusion of my whole Ph.D. study. I have to say that the
acknowledgement is the most difficult part for me in this thesis, because there are too many
people I wish to acknowledge. Without these people, I don’t know how my life would be.
First, I would like to express my sincerest gratitude to my supervisor Prof. Sailing He for
his excellent guidance and abundant support. What he taught me are far beyond the scientific
knowledge, which can help me a lot in my future academic career.
I also wish to thank Dr. Zhinong Ying at Sony Mobile for his expert supervision in my
research area. Thanks to his support, I can understand not only the theory but also the
practical applications, which help me exploit deeply the research knowledge. I am also
grateful for Mr. Thomas Bolin and Dr. Peter Karlsson, who really facilitate my visiting time
in Sony Mobile.
Thanks then go to Prof. Buon Kiong Lau at Lund University. Without him, it would be
very hard for me to finish my work successfully. As I am a Ph.D. with few experiences, he
gave me a lot of suggestions and provided me with good experimental conditions.
I want to send my appreciation to Prof. Lars Jonsson with my full sincerity. He is the
internal reviewer for my thesis, who had to spend a lot of time on revising it. In addition, he
also gave me many helpful advices with his smile. That makes my work go smoother.
I am also thankful for Prof. Martin Norgren for his co-supervision. I thank my former
colleagues Dr. Xin Hu, Dr. Pu Zhang, Dr. Fei Xiao, Dr. Haiyan Qin, Dr. Xianqi Lin, Mr. Yi
Tan, Dr. Hui Li, Dr. Kelin Jia; fellow Ph.D. students and postdoc Dr. Helin Zhou, Ms. Xiaolei
Wang, Mr. Kun Zhao, Dr. Andrés Alayon Glazunov, Mr. Fei Sun… for all the good time
together.
With the full administrative support, Financial Administrator Ms. Carin Norberg and
System Administrator Mr. Peter Lönn are also the people I have to acknowledge.
Thank all my friends Dr. Jiang Xiong, Dr. Xiaopeng Wang, Mr. Zhenpeng Zhang, and Mr.
Xuan Li in China.
I also acknowledge EU Erasmus Mundus External Cooperation Window TANDEM for
the financial support of my study in KTH.
Last but not least, I want to send all my thankfulness to my parents, parents in law, my
wife Trang and my cute son for their endless love and support.
Shuai Zhang
January, 2013
List of papers
I. S. Zhang, S. N. Khan, and S. He. “Reducing mutual coupling for an extremely
closely-packed tunable dual-element PIFA array through a resonant slot antenna
formed in-between,” IEEE Trans. Antenna Propag., vol. 58, No. 8, pp. 2771-2776,
Aug. 2010.
II. S. Zhang, B. K. Lau, Y. Tan, Z. Ying, and S. He, “Mutual coupling reduction of
two PIFAs with a T-shape slot impedance transformer for MIMO mobile
terminals,” IEEE Trans. Antennas Propag., vol. 60, No. 3, pp. 1521-1531, Mar.
2012.
III. S. Zhang, Z. Ying, J. Xiong, and S. He, “Ultrawideband MIMO/diversity
antennas with a tree-like structure to enhance wideband isolation,” IEEE Antennas
wireless Propag. Lett., vol. 8, pp.1279-1282, Dec. 2009.
IV. S. Zhang, B. K. Lau, A. Sunesson, and S. He, “Closely-packed UWB
MIMO/diversity antenna with different patterns and polarizations for USB dongle
applications,” IEEE Trans. Antennas Propag., vol. 60, No. 9, pp. 4372-4380, Sep.
2012.
V. S. Zhang, A. A. Glazunov, Z. Ying, and S. He, “Reduction of the envelope
correlation coefficient with improved efficiency for mobile LTE MIMO antenna
arrays: mutual scattering mode,” IEEE Trans. Antennas Propag., accepted.
VI. S. Zhang, K. Zhao, Z. Ying, and S. He, “Diagonal antenna-chassis mode and its
application for wideband LTE MIMO antennas in mobile handsets,” IEEE Trans.
Antennas Propag., submitted.
VII. S. Zhang, K. Zhao, Z. Ying, and S. He, “Body loss and MIMO performance
investigation of different mobile terminal LTE MIMO antenna types with user
effects,” IEEE Trans. Antennas Propag., submitted.
VIII. S. Zhang, K. Zhao, Z. Ying, and S. He, “Adaptive quad-element multi-wideband
antenna array for user-effective LTE MIMO mobile terminals,” IEEE Trans.
Antennas Propag., accepted.
Papers and US patents not included in this thesis:
Journal papers:
IX. S. Zhang, J. Xiong, and S.He, "MIMO antenna system of two closely-positioned
PIFAs with high isolation," Electronics Letters, vol.45, No.15, pp.771-773, 2009.
X. S. Zhang, P. Zetterberg, and S. He, “Printed MIMO antenna system of four
closely-spaced elements with large bandwidth and high isolation,” Electronics
Letters, vol.46, No.15, pp.1052-1053, 2010.
XI. S. Zhang, S. N. Khan, and S. He, “Modified rhombic monopole antenna for low
loss frequency notched UWB applications,” J. Electromagn.. Waves Appl., vol. 23,
No. 2-3, pp. 361-368, 2009.
XII. K. Zhao, S. Zhang, Z. Ying, T. Bolin, and S. He, “SAR study of different MIMO
antenna designs for LTE application in smart mobile handsets,” IEEE Trans.
Antennas Propag., accepted.
XIII. K. Zhao, S. Zhang, and S. He, “Enhance the bandwidth of a rotated rhombus slot
antenna with multiple parasitic elements,” J. Electromagn. Waves Appl., vol. 24,
No. 2-3, pp.2087–2094, 2010.
XIV. S. N. Khan, S. Zhang, and S. He, “Low profile and compact size coplanar UWB
antenna working from 2.8 GHz to over 40 GHz,” Microw Opt Technol Lett., vol.
51, pp. 408-411, 2009.
XV. K. Zhao, S. Zhang, and S. He, “Closely-located MIMO antennas of tri-band for
Wlan mobile terminal applications,” J. of Electromagn. Waves Appl., vol. 24, No.
2-3, pp. 363–371, 2010.
XVI. K. Zhao, S. Zhang, Z. Ying, T. Bolin, and S. He, “Avoid the degradation of
radiation efficiency for LTE mobile antenna in CTIA talking and data modes due
to user’s hand effects,” IEEE Antennas wireless Propag. Lett., submitted.
Conference papers:
XVII. S. Zhang, K. Zhao, Z. Ying and S. He, “Body-Effect-Adaptive Compact
Wideband LTE MIMO Antenna Array with Quad-Elements for Mobile Terminals,”
IEEE International Workshop on Antenna Technology (iWAT), Tucson, Arizona,
USA, 2012. (Invited Paper)
XVIII. S. Zhang, Z. Ying and S. He, “Mutual scattering mode for LTE MIMO antennas,”
Asia-Pacific Conference on Antennas and Propagation (APCAP), Singapore, 2012.
(Invited Paper)
XIX. S. Zhang, Z. Ying and S. He, “Diagonal chassis mode for mobile handset LTE
MIMO antennas and its application to correlation reduction,” International
Workshop on Electromagnetics(IWEM), China, 2012. (Invited Paper)
XX. S. Zhang, K. Zhao, Z. Ying and S. He, “Diagonal antenna-chassis mode for
wideband LTE MIMO antenna arrays in mobile handsets,” IEEE International
Workshop on Antenna Technology (iWAT), 2013. (Invited Paper)
XXI. S. Zhang, B. K. Lau, A. Sunesson and S. He, “Closely-located PIFAs with high
isolation for MIMO applications,” IEEE International Symposium on Antennas
and Propagation, USA, 2011.
XXII. S. Zhang, B. K. Lau, A. Sunesson and S. He, “UWB MIMO antenna for USB
dongles with angle and polarization diversity,” International Symposium on
Antennas and Propagation, 2012.
XXIII. S. Zhang, B. K. Lau, A. Sunesson and S. He, “T-shape slot induced decoupling
for closely spaced dual PIFAs in MIMO terminals,” 1th COST IC1004
Management Committee Meeting, Sweden, 2011.
XXIV. S. Zhang, B. K. Lau, A. Sunesson and S. He, “Diversity-rich compact UWB
MIMO antenna for USB dongles,” 4th COST IC1004 Management Committee
Meeting, France. 2012.
XXV. S. Zhang, K. Zhao, V. Plicanic, Z. Ying and S. He, “MIMO reference antenna for
OTA applications,” The European Conference on Antennas and Propagation
(EuCAP), 2013. submitted
XXVI. K. Zhao, S. Zhang, Z. Ying, T. Bolin, and S. He, “SAR study of different MIMO
antenna designs for LTE application in smart mobile phones,” in Proc. IEEE
Antennas Propagation Soc. Int. Symp., Chicago, 2012.
XXVII. K. Zhao, S. Zhang, Z. Ying, and S. He, “Body loss study of MIMO antenna for
LTE application in smart mobile phones with different lengths,” The European
Conference on Antennas and Propagation (EuCAP), 2013.
XXVIII. Z. Wang, Q. Wang, S. Zhang, and R. Liu, “Design of an economical compact
broadband waveguide power divider,” 4th IEEE International Conference on
Circuits and Systems for Communications ICCSC, Shanghai, 2008.
US patents
XXIX. S. Zhang, S. He, K. Zhao, and Z. Ying “Multi-band wireless terminal with
multiple antennas along an end portion of a backplate,” Sony Mobile patent
pending and filed by US, Oct. 2011.
XXX. S. Zhang, and Z. Ying, “Multi-band co-located MIMO antenna with different type
antenna elements to enhance de-correlation bandwidth for LTE mobile terminal,”
Sony Mobile patent pending and filed by US, Feb. 2012.
XXXI. S. Zhang, Z. Ying, S. He, and K. Zhao “Selective distributed MIMO antenna
array for optimal OTA performance in LTE applications,” Sony Mobile patent
pending and filed by US, Apr. 2012.
Acronyms
ADG Actual Diversity Gain AoA Angle of Arrival
CSI Channel State Information DG Diversity Gain ECC Envelope Correlation Coefficient EDG Effective Diversity Gain LTE Long Term Evolution MAS Multiple Antenna Systems ME Multiplexing Efficiency MEG Mean Effective Gain
MIMO Multiple-Input Multiple-Output PCB Printed Circuit Board PIFA Planar Inverted-F Antenna
SER Symbol Error Rate SNR Signal to Noise Ratio SISO Single-Input Single-Output UWB Ultra-Wide Band WiMAX Worldwide Interoperability for Microwave Access WLAN Wireless Local Area Network XPR Cross Polarization Ratio
Contents
Chapter 1 Introduction .................................................................................................... 1
1.1 Background ......................................................................................................... 1
1.2 MIMO Wireless Communication ........................................................................ 1
1.3 Scope and Structure of This Thesis ..................................................................... 3
Chapter 2 Multiple-Antenna Performance Evaluation ................................................... 5
2.1 Diversity Performance ........................................................................................ 5
2.1.1 Balanced Branch Power-Mean Effective Gain (MEG) ................................ 5
2.1.2 Correlation .................................................................................................... 6
2.1.3 Diversity Gain ............................................................................................... 7
2.2 MIMO Performance ............................................................................................ 7
2.2.1 MIMO Capacity ............................................................................................ 7
2.2.2 Multiplexing Efficiency ................................................................................ 8
2.3 Discussions .......................................................................................................... 9
Chapter 3 Decoupling of Compact MIMO/Diversity Antennas Operating at High
Frequencies ................................................................................................................... 13
3.1 MIMO/Diversity Antennas with Single and Multiple Bands Isolation ............ 13
3.1.1 Isolation enhancement for an Extremely Closely-Spaced Dual-PIFA Array
through a Resonant Slot Antenna Formed In-between ....................................... 13
3.1.2 Mutual Coupling Reduction of Two PIFAs with a T-shape Slot Impedance
Transformer for MIMO Mobile Terminals ......................................................... 15
3.2 MIMO/Diversity Antennas for Ultra-Wideband (UWB) Isolation ..................... 18
3.2.1 Ultra-wideband MIMO/Diversity Antennas with a Tree-like Structure to
Enhance Wideband Isolation ................................................................................. 19
3.2.2 Closely-Packed UWB MIMO/Diversity Antenna with Different Patterns and
Polarizations for USB Dongle Applications .......................................................... 21
Chapter 4 Correlation Reduction with Improved Total Efficiency for the Low
Frequencies of the Mobile Handset MIMO/Diversity Antennas ................................ 27
4.1 Reduction of the Envelope Correlation Coefficient with Improved Efficiency for
Mobile LTE MIMO Antenna Arrays: Mutual Scattering Mode ............................. 27
4.1.1 Dual Monopoles on a Large Ground Plane with High Losses ................... 28
4.1.2 Dual-PIFA MIMO Antennas on Mobile Chassis ......................................... 30
4.1.3 Experiment .................................................................................................. 33
4.2 Diagonal Antenna-Chassis Mode and Its Application for Wideband LTE MIMO
Antennas in Mobile Handset ................................................................................... 34
4.2.1 Collocated Dual Monopoles ....................................................................... 35
4.2.2 Collocated Dual PIFAs ............................................................................... 39
4.2.3 Multiple Wideband LTE MIMO Antenna for Mobile Terminals .............. 41
4.2.4 Separately located LTE MIMO Antenna .................................................... 43
Chapter 5 Body Loss Investigation and User-Effect Reduction for LTE MIMO
Antennas in Mobile Terminals ..................................................................................... 47
5.1 Body Loss and MIMO Performance Investigation of the Different LTE MIMO
Antenna Types with the User Effects for Mobile Terminals .................................... 47
5.2 Adaptive Quad-Element Multi-Wideband Antenna Array for User-Effective
LTE MIMO Mobile Terminals ................................................................................. 49
Chapter 6 Conclusions and Summary of Papers .......................................................... 55
6.1 Conclusions ....................................................................................................... 55
6.1 Summary of Papers ........................................................................................... 56
1
Chapter 1
Introduction
1.1 Background
People’s demands for faster transmission and receiving of information seem endless and
have been the driving force behind the progress of wireless technology. Since 1985, wireless
communication systems have rapidly evolved from the analog systems (1G: first-generation
systems), to digital systems (2G: second-generation systems), and later to third-generation
systems (3G), which can realize multimedia transmission. In order to further increase the data
transmission rate, multiple-input and multiple-output (MIMO) technology has become an
important feature in fourth-generation (4G) wireless communication systems. As “a key to
gigabit wireless” [1], MIMO can linearly increase channel capacity with an increase in the
number of antennas, without needing additional frequency spectrum or power [2]-[4].
Moreover, popular wireless communication systems typically operate in a rich scattering
environment, which MIMO exploits to achieve the aforesaid large performance gain. In
practice this high data rate is highly dependent on the performance of multiple antennas. The
integration of several antennas operating in a polarization-diversity scheme at the base station
side of the link is a well-known and well-used solution for either 2G or 3G communication
standards today. However, it is much more complicated to put several radiators in a small
device due to the limited space allocated for them. Furthermore, the user body will also
significantly change the MIMO antenna performance. In this thesis the author will mainly
focus on performance investigation and enhancement of multiple antenna systems in compact
MIMO terminals.
1.2 MIMO Wireless Communication
In the conventional wireless system, there are one transmitter and one receiver, which are
called Single Input Single Output (SISO) system (as shown in Fig. 1.1 (a)). In the condition
of narrow band and static environment (frequency and time invariance), the scalar signal
mode can be given by:
𝑦 = ℎ𝑥 + 𝑛, (1.1)
where y is the received signal, x is the transmitted signal, n is characterized by additive white
Gaussian noise (AWGN) with zero mean and variance and h is the channel response.
Assuming the bandwidth of 1 Hz, the channel capacity of SISO is upper bounded by [5] [6]:
2
𝐶 = 𝑙𝑜𝑔 2(1 +𝐸𝑠
𝑁0|ℎ|2) (1.2)
where Es is the total transmit power, |ℎ|2is the power gain of the scalar channel and No is the
noise power spectrum density. The channel capacity of the SISO system increases
logarithmically with an increase in transmitting power.
Figure 1.1 Block diagram of SISO and MIMO systems.
In Fig. 1.1 (b), the block diagram of MIMO system is shown. Several pioneering works
[6]–[8] have been carried out on MIMO systems. The channel response is now described by a
channel matrix H, expressed by:
[
ℎ11 ⋯ ℎ1𝑀𝑅⋮ ⋱ ⋮
ℎ𝑀𝑇1 ⋯ ℎ𝑀𝑇𝑀𝑅
] (1.3)
where ℎ𝑚𝑛 is the complex transmission coefficient from the element n at Transmitter to the
element m at Receiver. In this case, the sampled vector signal model is given as
Y = HX + n (1.4)
where Y is the received signal vector at the MR receiving antennas, X is the transmitted signal
vector for the MT transmitting antennas, and n is the AWGN vector at the MR receiving
3
antennas. Without the channel state information (CSI) at the transmitter, the power is equally
allocated to the transmitter and the channel capacity of MIMO system is expressed by [6]:
Cequal power = ∑ log 2 (1 +Es
N0MTλi)
ri=1 (1.5)
where r is the number of orthogonal sub-channels (i.e., rank) and λi’s are the Eigen values of
the matrix HHH (if MT < MR) or H
HH (if MT > MR). (•)
H denotes the conjugate transpose (or
Hermitian) operator. Eq. (1.5) reveals that the high rate signal of MIMO systems is achieved
by opening multiple streams between the transmitters and receivers with the same operating
frequency channel.
1.3 Scope and Structure of This Thesis
This thesis is based on eight IEEE journal papers. The works can be divided into three
parts: (1) Paper I-IV: Decoupling for the multiple antenna system (MAS) operating on higher
frequencies (above 1700 MHz); (2) Paper V-VI: Correlation reduction and total efficiency
improvement for the mobile handset LTE MIMO MAS operating in lower bands (700-960
MHz) and multiple bands (both 700- 960 MHz and 1700-2700 MHz); (3) Paper VII-VIII:
User effect investigation and reduction for MAS. The thesis will be organized as follows:
In Chapter 2, the commonly used figure of merits is introduced to evaluate the
diversity/MIMO performance of MAS.
In Chapter 3, the single, multiple and wideband isolation enhancement methods are
presented for the higher bands.
In Chapter 4, the correlation reduction methods are investigated in the MAS of mobile
terminals. The total efficiencies of the MAS elements are also improved. During these studies,
the effects of chassis mode are considered.
In Chapter 5, the user effects are studied in different types and arrangements of MIMO
elements. An adaptive quad-element LTE MIMO antenna array is proposed to reduce the user
effects.
Finally, in Chapter 6 the conclusions of this thesis are presented, followed by a summary
of the author’s contributions to the Papers I-VIII.
References
[1] A. Paulraj, D. Gore, R. Nabar, and H. B¨olcskei, “An overview of MIMO
communications - a key to gigabit wireless,” Proceedings of the IEEE, vol. 92, pp. 198 –
218, Feb. 2004.
4
[2] R. D. Murch and K. B. Letaief, “Antenna systems for broadband wireless access,” IEEE
Commun. Mag., vol. 40, no. 4, pp. 76–83, Apr. 2002.
[3] G. Foschini, “Layered space-time architecture for wireless communication in a fading
environment when using multi-element antennas,” Bell Labs Tech. J., vol. 1, no. 2, pp.
41–59, 1996.
[4] J. Wallace, M. Jensen, A. Swindlehurst, and B. Jeffs, “Experimental characterization of
the MIMO wireless channel: Data acquisition and analysis,” IEEE Trans. Wireless
Commun., vol. 2, no. 2, pp. 335–343, Mar. 2003.
[5] C. E. Shannon, “A mathematical theory of communication,” Bell System Technology
Journal, vol. 27, pp. 623-656, Oct. 1948.
[6] J. Winters, “On the capacity of radio communication systems with diversity in a
Rayleigh fading environment,” IEEE Journal on Selected Areas in Communications, vol.
5, pp. 871 – 878, June 1987.
[7] G. J. Foschini and M. J. Gans, “On limits of wireless communications in a fading
environment when using multiple antennas,” Wireless Personal Communication, vol. 6,
pp. 311 – 335, 1998.
[8] I. E. Telatar, “Capacity of multi-antenna Gaussian channels,” European Transactions on
Telecommunications, vol. 10, pp. 585 – 595, 1999.
5
Chapter 2
Multiple-Antenna Performance Evaluation
A multiple antenna system can operate in diversity or MIMO schemes according to the
SNR level under a rich scattering circumstance. If the SNR is low, the diversity can be
applied to combat the fading. All the antennas at the transmitter (or receiver) send (or receive)
the same signals over the same channel. Since the transmitting (or receiving) antennas are
uncorrelated, the possibility of the fading deeps for all the antennas is reduced. The reliability
in the wireless link is improved. In the high SNR region, the MAS will work in the MIMO
scheme and utilize the fading to provide several uncorrelated channels. The different data are
simultaneously transmitted over different channels with the same operating frequency, where
the maximum data rate is achieved. In this chapter, the figure of merits for MAS is addressed
to help evaluate diversity and MIMO performance. The detail of the parameters for the single
antenna (impedance bandwidth, gain, radiation pattern, efficiency and so on) will not be
discussed in this thesis but can be found in [1].
2.1 Diversity Performance
2.1.1 Balanced Branch Power-Mean Effective Gain (MEG)
An imbalanced power of the diversity branches will result in a diversity loss which is
proportional to the imbalanced level [2]. This imbalance is affected by the total antenna
efficiency. However, since the diversity is expected to be used under any possible
circumstances, the antenna-channel mismatch is also quite important. Therefore, the mean
effective gain (MEG) is widely used, which accounts for all the effects of the total antenna
efficiency, antenna gain and wireless environment. The formula is given as follows [3]:
𝑀𝐸𝐺 = ∮(𝑋𝑃𝑅
𝑋𝑃𝑅+1𝐺𝜃(𝛺) ∙ 𝑃𝜃(𝛺) +
1
𝑋𝑃𝑅+1𝐺𝜙(𝛺) ∙ 𝑃𝜙(𝛺)) 𝑑𝛺 (2.1)
where 𝑑𝛺 = 𝑠𝑖𝑛 𝜃𝑑𝜃𝑑𝜙, XPR is the cross-polarization ratio, Gθ(Ω) and Gϕ(Ω) are the θ
and 𝜙 components of the realized antenna gain pattern, respectively. The antenna realized
gains in MEG are normalized to [4]:
∮ (𝐺𝜃(𝛺) + 𝐺𝜙(𝛺)) 𝑑𝛺 = 4𝜋 (2.2)
The statistical power spectrum distributions of both vertically and horizontally polarized
incident radio waves can be represented by 𝑃𝜃(Ω) and 𝑃𝜙(Ω), respectively. If 𝑃𝜃(Ω) and
6
𝑃𝜙(Ω) are separable in elevation and azimuth, they can be described by [5]:
𝑃𝜃(𝛺) = 𝑃𝜃(𝜃, 𝜙) = 𝑃𝜃(𝜃)𝑃𝜃(𝜙) (2.3)
𝑃𝜙(𝛺) = 𝑃𝜙(𝜃, 𝜙) = 𝑃𝜙(𝜃)𝑃𝜙(𝜙) (2.4)
𝑃𝜃(Ω) and 𝑃𝜙(Ω) are normalized to:
∮(𝑃𝜃(𝛺))𝑑𝛺 = ∮(𝑃𝜙(𝛺)) 𝑑𝛺 = 1 (2.5)
XPR (cross polar ratio) is the ratio of time average vertical 𝑃𝜃(𝛺) power to time average
horizontal 𝑃𝜙(𝛺) power [6]:
𝑋𝑃𝑅 = 𝑃𝜃(𝛺)
𝑃𝜙(𝛺) (2.6)
In an isotropic environment, it can be characterized by XPR=1, Pθ(Ω) = 𝑃𝜙(𝛺) =1
4𝜋 [7],
and MEG= η/2, where η is the total antenna efficiency.
In order to get an optimal diversity performance, the multiple antenna system has to
satisfy the balance power requirement
MEGantenna1 ≈ MEGantenna2 (2.7)
2.1.2 Correlation
The correlation of a multiple antenna system is used to describe how independent the
different antenna ports are. The envelope correlation coefficient 𝜌𝑒 is given in [8]:
𝜌𝑒 ≈ (∮(𝑋𝑃𝑅⋅𝐸𝜃𝑋𝐸𝜃𝑌
∗ 𝑃𝜃+𝐸𝜙𝑋𝐸𝜙𝑌∗ 𝑃𝜙)𝑑𝛺
√∮(𝑋𝑃𝑅⋅𝐸𝜃𝑋𝐸𝜃𝑋∗ 𝑃𝜃+𝐸𝜙𝑋𝐸𝜙𝑋
∗ 𝑃𝜙)𝑑𝛺∮(𝑋𝑃𝑅⋅𝐸𝜃𝑌𝐸𝜃𝑌∗ 𝑃𝜃+𝐸𝜙𝑌𝐸𝜙𝑌
∗ 𝑃𝜙)𝑑𝛺)
2
(2.8)
where Eθ,ϕX and Eθ,ϕY are the embedded, polarized complex electric field patterns of two
antennas X and Y, respectively in the multiple-antenna system. The rule of thumb for good
diversity/MIMO performance is 𝜌𝑒 < 0.5. The Eq. (2.8) reveals that the envelope correlation
coefficient is also a parameter related to wireless circumstances. In an MAS, the envelope
7
correlation coefficient can be affected by the element distance, directivity of antennas, ground
plane (or chassis), Q factors (Paper V), etc.
2.1.3 Diversity Gain
Compared with the received Signal-Noise Ratio (SNR) from a single reference antenna,
the enhancement of the combined SNR from MAS is described as Diversity Gain (DG) [7],
[9]. If the best branch in MAS is selected as the reference antenna [7], [9], the diversity gain
(DG) at the possibility 𝑃(𝑟𝑐) can be expressed by [7]:
𝐷𝐺 =(𝑟𝑐𝛤𝑐)
(𝑟
𝛤)𝐵𝑒𝑠𝑡 𝑏𝑟𝑎𝑛𝑐ℎ
|
𝑃(𝑟𝑐)
(2.9)
where 𝑟 and 𝑟𝑐 are instantaneous SNRs, the detailed description of 𝑃(𝑟𝑐) is given in [9],
and 𝛤 and 𝛤𝑐 are mean SNRs for the combined and best single antenna branch signals,
respectively. Typically, the possibility 𝑃(𝑟𝑐) is 1% or 50%.
Generally, the total efficiency of the best branch in a compact MAS is lower than that of
the antenna in a single antenna system due to mismatch, mutual coupling and additional ohm
loss in the MAS. Therefore, Effective diversity gain (EDG) is also commonly utilized, by
selecting the antenna with 100% total efficiency as the reference [10]:
𝐸𝐷𝐺 =(𝑟𝑐𝛤𝑐)
(𝑟
𝛤)𝐵𝑒𝑠𝑡 𝑏𝑟𝑎𝑛𝑐ℎ
|
𝑃(𝑟𝑐)
∙ 𝜂𝐵𝑒𝑠𝑡 𝑏𝑟𝑎𝑛𝑐ℎ (2.10)
where 𝜂𝐵𝑒𝑠𝑡 𝑏𝑟𝑎𝑛𝑐ℎ is the total efficiency of the best single antenna in a compact multiple
antenna system.
The DG and EDG give two extreme values. In order to examine whether the realistic
single antenna in a compact terminal needs to be replaced by a multiple antenna system with
the same condition, actual diversity gain (ADG) is defined, which uses the realistic single
antenna as reference [9] and [10]:
𝐷𝐺 =(𝑟𝑐𝛤𝑐)
(𝑟
𝛤)𝑆𝑖𝑛𝑔𝑙𝑒 𝑎𝑛𝑡𝑒𝑛𝑛𝑎 𝑠𝑜𝑙𝑢𝑡𝑖𝑜𝑛
|
𝑃(𝑟𝑐)
(2.11)
2.2 MIMO Performance
2.2.1 MIMO Capacity
It has been mentioned in Chapter 1 that without the channel state information (CSI) at the
transmitter, the power will be equally allocated to the transmitter and the channel capacity of
the MIMO system can be expressed by [11]:
8
Cequal power = ∑ log 2 (1 +Es
N0MTλi)
ri=1 (1.5)
where r is the number of orthogonal sub-channels (i.e., rank) and λi’s are the Eigen values of
the matrix HHH (if MT < MR) or H
HH (if MT > MR). (•)
H denotes the conjugate transpose (or
Hermitian) operator. If the CSI is available at the transmitter, the power allocation can be
optimized to the stronger sub-channels rather than the weaker ones via a water filling
algorithm [7]:
𝐶𝑤𝑎𝑡𝑒𝑟 𝑓𝑖𝑙𝑙𝑖𝑛𝑔 = ∑ 𝑙𝑜𝑔 2(𝜆𝑖 ∙ 𝐷) 𝑟𝑖=1 (2.12)
and
𝐷 =1
𝜆𝑖+ 𝑝𝑖 (2.13)
where D is the “water level” for each of the sub-channels to be filled up to and pi = SNRi/𝜆𝑖
is the input power to the i-th sub-channel.
2.2.2 Multiplexing Efficiency
In order to evaluate the MIMO performance in a simple way, the Multiplexing Efficiency
(ME) or mux
is introduced according to [12]. ME is defined as the power penalty of a
realistic multiple antenna system in achieving a given capacity, compared with an ideal
antenna system with 100% total efficiency and zero correlation [12].
With the assumption of high SNR and an isotropic environment (i.e., equal likelihood of
impinging waves from any direction), ME can be expressed by [12]:
𝑚
= √𝜂1𝜂2 (1 |𝜌𝑐|2) , (2.14)
where η1 and η
2 are the total efficiencies of the MIMO antenna elements.
In a propagation channel with Gaussian distribution of the angle of arrival (AoA) given
by equation (9) in [13], the MIMO multiplexing efficiency can also be evaluated with the
following assumptions: The mean incidence direction is denoted by 𝜙0 and 𝜃0 (as opposed
to the isotropic channel, the likelihood of impinging waves is not the same in all directions
but has a maximum at AoA 𝜙0 and 𝜃0). It is assumed that the angular spread is the same in
both elevation and azimuth and approximately equals 30o. We have further restricted our
analysis to channels with balanced polarizations, i.e., with cross-polarization ratio XPR=1.
For reference, we have taken two ideal cross-polarized antennas, which give a zero
correlation. Following the above conditions, the multiplexing efficiency is a function of the
9
mean incidence direction:
𝑚
( 𝜙0, 𝜃0) =
√4 𝑀𝐸𝐺1( 𝜃0, 𝜙0)𝑀𝐸𝐺2( 𝜃0, 𝜙0)(1 |𝜌𝑐(𝜃0, 𝜙0)|2) (2.15)
where 𝑀𝐸𝐺1( 𝜃0, 𝜙0) and 𝑀𝐸𝐺2( 𝜃0, 𝜙0) are respectively the mean effective gains of each
antenna port [14], and 𝜌𝑐(𝜙0, 𝜃0) is the complex envelope correlation of the received signal.
2.3 Discussions
From the introduction and analyses of the merits for the MAS, we can see that the total
efficiencies of antennas and the correlation between multiple antenna elements affect both
diversity and MIMO performance. If the propagation channel environment is determined and
the multiple antenna elements are symmetrical in a compact terminal, a good
diversity/MIMO performance can be achieved through high total antenna efficiencies and low
envelope correlation coefficient. Furthermore, the compact MAS terminals also need to have
multiple and wide bands to satisfy different standards and functions.
For a multi-port antenna system, when only one port is fed and the others are terminated
with 50-Ω load, the total efficiency can be evaluated by the following equations:
𝜂𝑡𝑜𝑡𝑎𝑙 = 𝜂𝑟𝑎𝑑𝑖𝑎𝑡𝑖𝑜𝑛𝜂𝑚𝑖𝑠𝑚𝑎𝑡𝑐ℎ𝑖𝑛𝑔+𝑐𝑜 𝑝𝑙𝑖𝑛𝑔 (2.16)
𝜂𝑚𝑖𝑠𝑚𝑎𝑡𝑐ℎ𝑖𝑛𝑔+𝑐𝑜 𝑝𝑙𝑖𝑛𝑔 = 1 |𝑆𝑖𝑖|2 ∑ |𝑆𝑗𝑖|
2𝑗≠𝑖 (2.17)
where 𝜂𝑡𝑜𝑡𝑎𝑙 , 𝜂𝑟𝑎𝑑𝑖𝑎𝑡𝑖𝑜𝑛 and 𝜂𝑚𝑖𝑠𝑚𝑎𝑡𝑐ℎ𝑖𝑛𝑔+𝑐𝑜 𝑝𝑙𝑖𝑛𝑔 are the total efficiency, radiation
efficiency, and mismatching+ mutual coupling efficiency, respectively. Subscripts i, and j
represent the operating and terminated ports, respectively. According to Eq. (2. 17), mutual
coupling (∑ |𝑆𝑗𝑖|2
𝑗≠𝑖 ) can lead to a decrease of the total efficiency.
If the radiation efficiency of MAS is high, a high total efficiency and a low correlation
can be achieved by reducing the mutual coupling between MAS elements [15] [16]. Typically,
this method is quite effective when multiple antennas are operating at high frequencies
(above 1700 MHz).
If the radiation efficiency is low, the low mutual coupling may result from the high losses,
where the required high total efficiency and low correlation may not be satisfied. Therefore,
the target should be straightforwardly improving the correlation and total efficiencies, but not
reducing the mutual coupling any more. Mobile terminal MAS operating at the low
frequencies (below 960 MHz) belong to this situation. In addition, the wavelength of the
operating frequency in the lower bands is quite comparable with the mobile chassis
geometries, where the chassis (or ground plane) becomes an effective radiator. This may
10
result in similar radiation patterns of the MAS elements and degrade the MIMO/ diversity
performance.
Finally, the interactions between multiple antennas and users will reduce the total
efficiency and affect correlations of antennas by shifting the resonant frequencies and
absorbing some of the radiated/received power.
References
[1] C. A. Balanis, Antenna Theory: Analysis and Design, Wiley-Interscience, 3 edition, April
4, 2005.
[2] V. Plicanic, B. K. Lau, A. Derneryd, and Z. Ying, “Actual diversity performance of a
multiband diversity antenna with hand and head effects,” IEEE Trans. Antennas Propag.,
vol. 57, no. 5, pp. 1547-1556, May 2009.
[3] T. Taga, “Analysis for mean effective gain of mobile antennas in land mobile radio
environments,” IEEE Trans. Veh. Technol., vol. VT-39, no. 2, pp. 117–131, May 1990.
[4] Ogawa K., Matsuyoshi T., Monma K., “An analysis of the performance of a handset
diversity antenna influenced by head, hand and shoulder effects at 900 MHz: part
I-effective gain characteristics and part II-correlation characteristics,” IEEE Trans. On
Vehicular Technology, Vol.50, No.3, May 2001, pp.830-853.
[5] Pedersen G. F., Andersen J. B., “Handset antennas for mobile communications:
integration, diversity and performance,” Review of Radio Science 1996-1999, August
1999, pp. 119-137.
[6] Knudsen M. B., “Antenna systems for handsets,” ATV-Industrial PhD Project EF-755,
Aalborg University 2001.
[7] R. Vaughan and J. B. Andersen, Channels, Propagation and Antennas for Mobile
Communications. London, U.K.: Inst. Elect. Eng., 2003.
[8] M. B. Knudsen and G. F. Pedersen, “Spherical outdoor to indoor power spectrum model
at the mobile terminal,” IEEE J. Sel. Areas Commun., vol. 20, no. 6, pp. 1156–1168, Aug.
2002.
[9] V. Plicanic, “Characterization and enhancement of antenna system performance in
compact MIMO terminals,” a thesis of Lund University for the degree of Doctor of
Philosophy, 2011.
[10] P.-S. Kildal and K. Rosengren, “Correlation and capacity of MIMO systems and mutual
coupling, radiation efficiency, and diversity gain of their antennas: Simulations and
measurements in a reverberation chamber,” IEEE Commun. Mag., pp. 104–112, Dec.
2004.
[11] J. Winters, “On the capacity of radio communication systems with diversity in a
Rayleigh fading environment,” IEEE Journal on Selected Areas in Communications, vol.
5, pp. 871 – 878, June 1987.
[12] R. Tian, B. K. Lau, and Z. Ying, “Multiplexing efficiency of MIMO antennas,” IEEE
Antennas Wireless Propag. Lett., vol. 10, pp. 183-186, 2011.
[13] R. Tian, B. K. Lau, and Z. Ying, “Multiplexing efficiency of MIMO antennas in arbitrary
propagation scenarios,” European Conference on Antennas and Propagation (EuCAP),
Prague, Czech Republic, pp.373-377, 26-30 Mar. 2012.
11
[14] A. A. Glazunov, A.F. Molisch, and F. Tufvesson, “Mean effective gain of antennas in a
wireless channel,'' Microwaves, Antennas & Propagation, IET, vol.3, no.2, pp.214-227,
Mar. 2009.
[15] Hallbjorner, P., “The significance of radiation efficiencies when using S-parameters to
calculate the received signal correlation from two antennas,” IEEE Antennas Wireless
Propag. Lett., pp. 97–99, 2005.
[16] S. Blanch, J. Romeu, and I. Corbella, “Exact representation of antenna system diversity
performance from input parameter description,” Electron Lett., 39 (2003), 705–707.
12
13
Chapter 3
Decoupling of Compact MIMO/Diversity
Antennas Operating at High Frequencies
3.1 MIMO/Diversity Antennas with Single and Multiple Bands
Isolation
In a compact MIMO terminal, the space for the MAS is quite limited, which would result
in high mutual coupling and degrade the MIMO/diversity performance [1]. Planar inverted-F
antennas (PIFAs) have been widely used in portable devices due to its low profile and cost.
The results of [2]-[4] indicated that the inter-PIFA distance have to be large than half (or
quarter) of the free space wavelength in order to achieve a mutual coupling less than -20 dB
(or -10 dB), when two PIFAs share the same ground plane. Therefore, it is necessary to
develop mutual coupling reduction technologies for a compact MIMO/diversity PIFA array.
A number of studies have been done to enhance the isolation between closely-positioned
PIFAs [5]-[18]. Mushroom-like EBG structures [5]-[7] or defected ground structures (DGS)
[8]-[11] could reduce the mutual coupling by suppressing the propagation of the surface wave
or providing a band-gap effect. A compact MIMO/diversity antenna with two feeding ports
sharing the same radiator has been proposed in [12], [13]. In [14], a neutralization line is
inserted between two PIFAs to provide another field which can cancel the mutual coupling.
In the recent years, the inter-PIFA distance has been reduced rapidly maintaining the mutual
coupling less than -20 dB. In [15], by separating the common ground plane of two PIFAs, the
distance of 0.17λo has been achieved. The spacing has been reduced to less than 0.078λo
through a fish bone-shaped ground plane [16]. A decoupling element has been introduced in
[17] and a spacing of less than 0.0294λo has been realized. This decoupling element can
efficiently enhance the isolation, but has to occupy some areas between the two MIMO
elements. In [18] through removing the neutralization line in [14] to the outside edges, the
distance has been reduced to less than 0.027λo. However, the isolation cannot be enhanced
efficiently and the neutralization line is still available.
3.1.1 Isolation enhancement for an Extremely Closely-Spaced Dual-PIFA
Array through a Resonant Slot Antenna Formed In-between
In Paper I, the isolation between two PIFAs is reduced through a half wavelength slot
antenna formed by the slot etched on the ground plane and the adjacent edges of two PIFAs,
as shown in Fig. 3.1. The current distribution of the dual PIFA array in Fig. 3.1 is provided in
Fig. 3.2, when the left PIFA is operating with the other terminated by a 50 ohm load. We can
14
observe that due to the half wavelength slot antenna, the coupling current from the left PIFA
is trapped in-between and cannot flow to the other. Indeed, the S parameters of the proposed
MIMO antenna in Fig. 3. 3 show that a measured mutual coupling less than -20 dB is
achieved across the WLAN band of 2.4GHz-2.48GHz. The peak gain and efficiency are 2.12
dBi and 71.5%, respectively [Paper I]. The effectiveness of this method can be observed
clearly.
Figure 3.1.Geometry of the proposed dual-element PIFA array: (a) top view, (b) side view [Paper I].
Figure 3.2.The current distributions of the proposed MIMO antenna [Paper I]
In the method of Paper I there is only a narrow slot between the PIFAs. Therefore, the
inter-PIFA distance of the proposed MIMO antenna in Fig. 3.1 can be further reduced. In
Table I, the performances of the dual PIFA array with different inter-PIFA distances are listed.
We find that the proposed method in Paper I is still efficient even for an inter-PIFA spacing of
15
less than 0.0016λo (λo/600).
Figure 3.3. The comparison between the measured and simulated S-parameters of the proposed MIMO array
[Paper I].
TABLE I
SIMULATED PERFORMANCE OF ARRAYS WITH DIFFERENT INTER-PIFA DISTANCE
3.1.2 Mutual Coupling Reduction of Two PIFAs with a T-shape Slot
Impedance Transformer for MIMO Mobile Terminals
Compared with the previous results in [5]-[18], the method in Paper I can achieve a
smaller inter-PIFA distance and higher isolation. However, this method cannot be used for a
ground plane with arbitrary size and shape. This is because the impedance of the decoupling
slot is strongly affected by different sizes and shapes of ground plane, where good matching
cannot be guaranteed for the decoupling slot. Moreover, these methods in [5]-[18] and Paper
I cannot be used for dual-band applications, which further limit their versatility.
One of the solutions to these limitations can be found in Paper II. The geometry of the
proposed single-band PIFA array in Paper II are presented in Fig. 3.4. A T-shape slot (of
length Lhigh+Whigh/2) is inserted at the end of a quarter-wavelength decoupling slot antenna (of
Inter-PIFA
distance
-10 dB
impedance
bandwidth
(MHz)
20 dB
isolation
bandwidth
(MHz)
Peak
efficiency
(%)
Peak
gain
(dBi)
1.8 mm(0.0147λ0) 72 124 88.7 1.48
1.4 mm(0.0114λ0) 67 99 87.25 1.33
1 mm(0.0082λ0) 64 90 85.73 1.27
0.6 mm(0.0049λ0) 58 70 81.53 1.00
0.2 mm(0.0016λ0) 49 41 58.36 -0.64
16
length La+Ls+H+Llow) formed by the neighboring edges of two PIFAs. As an impedance
transformer for the decoupling slot, the T-shape slot can excite the decoupling slot
independent of the shape and size of the ground plane. The S parameters of two PIFAs with
T-shape slot are presented in Fig. 3.5. With the help of T-shape slot, the isolation between the
PIFAs is enhanced from 6.3 dB to more than 40 dB at the center frequency. The peak gain
and efficiency are also measured, which are 1.8 dBi and 70%, respectively.
Figure 3.4.Geometry of the proposed single-band PIFA array with T-shape slot: (a) top view, (b) ground plane,
(c) back view, (d) side view, (e) front view, (f) and (g) 3D view [Paper II]. All dimensions are given in mm.
Figure 3.5.Simulated S parameters with and without the T-shape slot impedance transformer, and measured S
parameters of the proposed single-band PIFAs for MIMO terminals [Paper II].
17
The method in Paper II can also realize a dual-band dual-isolation property. In Fig. 3.6 the
geometry of the proposed dual-band MIMO PIFA array are presented. Another slot is etched
on each PIFA to generate a higher frequency resonance. From the S parameters in Fig. 3.7,
we observe that two decoupling nulls are excited at the central frequencies of 2.4 GHz and
3.5 GHz, respectively. Based on the measured results, the proposed MIMO antennas can
cover the bands of 2.4-2.48 GHz (WLAN) and 3.4-3.6 GHz (WiMAX) with high isolations.
Figure 3.6.Geometry of the proposed dual-band PIFA array with a T-shape slot: (a) top view, (b) ground plane,
(c) back view, (d) side view, (e) front view, (f) and (g) 3D view [Paper II]. All dimensions are given in mm
Figure 3.7. Simulated S parameters with and without the T-shape slot impedance transformer, and measured S
parameters of the proposed dual-band PIFAs for MIMO terminals [Paper II].
18
The current distributions of the proposed dual-band dual-isolation PIFA array with
T-shape slot are given in Fig. 3.8 for 2.44 GHz and 3.5 GHz, with PIFA1 operating and PIFA
2 terminated with a 50 ohm load. At 2.44 GHz (the WLAN band), the currents are mainly
concentrated on part “B” (see Figs. 3.6 (g) and 3.8 (a)), which determines the operating
frequency of the lower band. In addition, most of the coupling currents are trapped in the slot
formed by the two PIFAs. At 3.5 GHz (the WiMAX band), the currents mainly focus on part
“A” (see Figs. 3.6 (g) and 3.8 (a)) which controls the operating frequency of the higher band.
The coupling currents are blocked by the L slot (see rout (3) in Fig. 3.6 (f)) and thus cannot
flow into the adjacent PIFA.
(a) (b)
Figure 3.8. Current distributions of the proposed dual-band PIFAs for MIMO terminals at (a) 2.44 GHz, (b) 3.5
GHz [Paper II].
3.2 MIMO/Diversity Antennas for Ultra-Wideband (UWB) Isolation
The decoupling methods for the single and multiple bands have been discussed above.
Next, the wideband isolation enhancement technology will be addressed.
MIMO and diversity technology can also be utilized in a UWB wireless communication
system, in order to increase the channel capacity or combat fading. In a portable UWB
MIMO/diversity antenna system, wideband isolation is required in a very limited space to
guarantee the good MIMO/diversity performance and compact structure. Many studies have
been carried out to achieve this target [24]-[28]. In [24], a T-shape reflector is inserted
between the two UWB monopole radiators to improve the isolation. In [25], a slot is etched
on a T-shape reflector in [24] to further reduce the mutual coupling. A diversity antenna for
PDA application has been proposed in [26] with three stubs to weaken the mutual coupling.
Different polarizations have been applied in [27]-[28]. However, in order to realize an
acceptable isolation, the size of the antennas in [24]-[28] are still relatively large. Recently, an
UWB diversity antenna with a small size of 37 × 45 mm has been presented in [29]. A good
isolation can be achieved from 3.1 GHz to 5 GHz (lower UWB band).
19
3.2.1 Ultra-wideband MIMO/Diversity Antennas with a Tree-like Structure
to Enhance Wideband Isolation
In order to further improve the isolation bandwidth, a tree-like structure, as shown in Fig.
3.9, is introduced in Paper III for a compact UWB MIMO/diversity antenna system. The
antenna system in Paper III has a compact size of 35 × 40 mm. which can cover the whole
UWB band (3.1-10.6GHz). Within the whole UWB band, a measured isolation level better
than -16 dB (-20dB in most of the band) is achieved, as illustrated in Fig. 3. 10. The
prototype and geometry of the UWB MIMO/diversity antenna are shown in Fig. 3. 9. The
compact structure can be observed clearly.
Figure. 3. 9. Geometry and prototype of the proposed UWB MIMO/diversity antenna system: (a) top view, (b)
back view. [Paper III]
The wideband isolation is realized with the tree-like structure mainly due to two reasons:
(1) Branch 1 (see Fig. 3. 9 (a)) can act as a reflector (see [24] and [29]) to separate the
radiation patterns of two MIMO elements, which can improve the wideband isolation. In
Paper III, the authors increase the number of the branches and the mutual coupling can be
further weakened. (2) More resonances can be introduced through increasing the number of
20
branches. Each of resonances is provided by a monopole of one branch or a quarter
wavelength slot formed by two neighboring branches. By properly adjusting the length and
position of each branch, the wideband isolation can be efficiently enhanced as the resonances
are in different frequency ranges. In Fig. 3.11 (b), improvement of the isolation across the
operating bands is clearly observed, with the total number of the branches increasing from
one to three. The simulation results show that branches 1, 2 and 3 mainly affect low, middle
and high frequencies, respectively. The current distributions with/without the tree-like
structures are shown in Fig. 3.12. The coupling currents of different frequencies from Port 1
(see Fig. 3. 9 (a)) are successfully blocked by the tree-like structure and cannot flow into Port
2.
Figure. 3. 10. Measured and simulated S-parameters of the proposed UWB MIMO/diversity antenna system.
[Paper III]
(a) (b)
Figure. 3. 11. Simulated S-parameters when the total number of the branches varies: (a) S11 and S22, (b) S12
and S21. [Paper III]
21
(a) (b)
(c)
Figure. 3. 12. Surface current distributions with Port 1 excitation: (a) 4 GHz, (b) 7 GHz and (a) 10 GHz. [Paper
III]
3.2.2 Closely-Packed UWB MIMO/Diversity Antenna with Different
Patterns and Polarizations for USB Dongle Applications
(a) (b)
Figure. 3. 13. (a) Detailed geometry and (b) the prototype of the proposed UWB MIMO/diversity antenna. The
grey area represents the copper layer and the dashed lines trace the edges of the copper layer on the opposite
side of the PCB. [Paper IV]
For the USB dongle applications, the UWB MIMO/diversity antenna needs to be much
22
smaller compared with those in the previous studies of [24]-[29] and Paper III. In Paper IV,
different patterns and polarizations are explored to realize the high isolation and extremely
compact structure In Fig. 3.13 the geometry of the proposed antenna is shown. A small size of
25 mm × 40 mm = 1000 mm2 (40% smaller than that of [29]) is observed. The S
parameters are given in Fig. 3.14. An isolation of higher than 20 dB is achieved across the
lower UWB band of 3.1-5.15 GHz
(a) (b)
Figure. 3. 14. (a) Simulated and (b) measured S parameters of the proposed UWB MIMO/diversity antenna.
[Paper IV]
Figure. 3. 15. Measured 3D radiation patterns for Port 1 and 2 at 3.2 GHz, 4.2 GHz, and 5.2 GHz. [Paper IV]
23
The isolation in Paper IV is mainly enhanced by the different radiation patterns and
polarizations of the two MAS elements. The 3D and 2D radiation patterns of the two
elements are presented in Fig. 3.15 and Fig.3.16, respectively. The polarization and pattern
diversity property are observed clearly.
Figure. 3. 16. The comparison between the simulated and measured and ф components in the x-z plane for
Port 1 at (b) 3.2 GHz, (d) 4.2 GHz, (f) 5.2 GHz and Antenna 2 at (a) 3.2 GHz, (c) 4.2 GHz, (e) 5.2 GHz [Paper
IV]
Gains and total efficiencies of two MIMO elements are also measured and obtained
through their 3D radiation patterns. During all the measurements (including S parameters,
radiation patterns, gains and efficiencies), two ferrites are utilized in order to mitigate
currents flowing from the ground plane onto the feed cables. The measured realized gains of
two MIMO elements, presented in Fig. 3. 17 (a), are larger than 1.5 dB within the lower
UWB band. Figure 3. 17 (b) shows the total efficiencies of two MIMO elements. Due to the
introduction the ferrites, the total efficiency are reduced by around 25%. However, it can be
observed in Fig. 3.17 (b) that the two MIMO elements are still quite efficient. In practical
applications the ferrites are not used and therefore the efficiency values can be much higher
than those shown in Fig. 3. 17 (b). The envelope correlation coefficient (ECC) obtained with
Formula (2.8) is less than 0.1 over the whole band. With the Formula (2.14) the multiplexing
24
efficiency is calculated from our measured efficiency and ECC, and shown in Fig. 3. 17 (b).
(a) (b)
Figure. 3. 17. (a) Measured peak gains and (b) efficiencies for Antennas 1 and 2. [Paper IV]
References
[1] S. Ko and R. Murch, “Compact integrated diversity antenna for wireless communications,”
IEEE Trans. Antennas Propag., vol. 49, no. 6, pp. 954–960, Jun. 2001.
[2] H. Carrasco, H. D. Hristov, R. Feick, and D. Cofre, “Mutual coupling between planar
inverted-F antennas,” Microwave Opt. Technol. Lett.,vol. 42, no. 3, pp. 224–227, Aug.
2004.
[3] T. Taga and K. Tsunekawa, “Performance analysis of a built-in planar inverted F antenna
for 800 MHz band portable radio units,” IEEE J. Select Areas Commun., vol. SAC-5, pp.
921–929, Jun. 1987.
[4] C. R. Rowell and R. D. Murch, “A capacitively loaded PIFA for compact mobile
telephone handsets,” IEEE Trans. Antennas Propag., vol. 45, no. 5, pp. 837–842, May.
1997.
[5] D. Sievenpiper, L. Zhang, R. F. J. Broas, N. G. Alexopolous, and E. Yablonovitch,
“High-impedance electromagnetic surfaces with a forbidden frequency band,” IEEE
Microw. Theory Tech., vol. 47, no. 11, pp. 2059–2074, Nov. 1999.
[6] F. Yang and Y. Rahmat-Samii, “Microstrip antennas integrated with electromagnetic
band-gap (EBG) structures: A low mutual coupling design for array applications,” IEEE
Trans. Antennas Propag., vol. 51, no. 10, pp. 2936–2946, Oct. 2003.
[7] L. Li, B. Li, H. X. Liu, and C. H. Liang, “Locally resonant cavity cell model for
electromagnetic band gap structures,” IEEE Trans. Antennas Propag., vol. 54, no. 1, pp.
90–100, Jan. 2006.
[8] D. Ahn, J. S. Park, C. S. Kim, J. Kim, Y. Qian, and T. Itoh, “A design of the low-pass
filter using the novel microstrip defected ground structure,” IEEE Microw. Theory Tech.,
vol. 49, no. 1, pp. 86–93, Jan. 2001.
25
[9] C. Caloz, H. Okabe, T. Iwai, and T. Itoh, “A simple and accurate model for microstrip
structures with slotted ground plane,” IEEE Microwave Wireless Comp. Lett., vol. 14, no.
4, pp. 133–135, Apr. 2004.
[10] Y. J. Sung, M. Kim, and Y. S. Kim, “Harmonics reduction with defected ground
structure for a microstrip patch antenna,” IEEE Antennas Wireless Propag. Lett., vol. 2,
pp. 111–113, 2003.
[11] D. Guha, M. Biswas, and Y. M. M. Antar, “Microstrip patch antenna with defected
ground structure for cross polarization suppression,” IEEE Antennas Wireless Propag.
Lett., vol. 4, pp. 455–458, 2005.
[12] S. C. K. Ko and R. D. Murch, “Compact integration diversity antenna for wireless
communications,” IEEE Trans. Antennas Propag., vol. 49, no. 6, pp. 954–960, Jun. 2001.
[13] C. T. Song, C. K. Mak, R. D. Murch, and P. B. Wong, “Compact low cost dual polarized
adaptive planar phased array for WLAN,” IEEE Trans. Antennas Propag., vol. 53, pp.
2406–2416, Aug. 2005.
[14] A. Diallo, C. Luxey, P. L. Thuc, R. Staraj, and G. Kossiavas, “Study and reduction of the
mutual coupling between two mobile phone PIFAs operating in the DCS1800 and UMTS
bands,” IEEE Trans. Antennas Propag., vol. 54, Nov. 2006.
[15] Y. Gao, X. D. Chen, Z. N. Ying and S. He, “Design and performance investigation of a
dual-element PIFA array at 2.5GHz for MIMO terminal,” IEEE Trans. Antennas
Propag., vol. 55, no. 12, pp. 3433–3441, Dec. 2007.
[16] C.-Y. Chiu, C.-H. Cheng, R. D. Murch, and C. R. Rowell, “Reduction of mutual
coupling between closely-packed antenna elements,” IEEE Trans. Antennas Propag., vol.
55, no. 6, pp. 1732–1738, Jun. 2007.
[17] A. C. K. Mak, C. R. Rowell and R. D. Murch, “Isolation enhancement between two
closely packed antennas,” IEEE Trans. Antennas Propag., vol. 56, no. 11, pp. 3411–3419,
Nov. 2008.
[18] A. Chebihi, C. luxey, A. Diallo, P. H. Thuc and R. Staraj, “A novel isolation technique
for UMTS mobile phones,” IEEE Antennas Wireless Propag. Lett., vol. 7, pp. 665–668,
2008.
[19] C.-Y. Chiu, K. M. Shum, C. H. Chan, “A tunable via-patch loaded PIFA with size
reduction,” IEEE Trans. Antennas Propag., vol. 55, no. 1, pp. 65–71, Jan. 2007.
[20] J. Villanen, J. Ollikainen, O. Kivekas, and P. Vainikainen, “Coupling element based
mobile terminal antenna structure,” IEEE Trans. Antennas Propagat., vol. 54, no. 7, pp.
2142-2153, Jul. 2006.
[21] W. L. Schroeder, A. A. Vila, and C. Thome, “Extremely small wideband mobile phone
antenna by inductive chassis mode coupling,” in Proc. 36th Europ. Microw. Conf.,
Manchester, 2006, pp. 1702-1705.
[22] W. L. Schroeder and C. T. Famdie, “Utilization and tuning of the chassis modes of a
handheld terminal for the design of multiband radiation characteristics,” in Proc. IEE
Wideband Multiband Antennas and Arrays, Sep. 7, 2005, pp.117-122.
[23] S. R. Best, “The significance of ground-plane size and antenna location in establishing
the performance of ground-plane-dependent antennas,” IEEE Antennas Propagat. Mag.
vol. 51, no. 6, pp. 29-42, Dec. 2009.
26
[24] K. L. Wong, S. W. Su, and Y. L. Kuo, “A printed ultra-wideband diversity monopole
antenna,” Microw Opt Techno Lett., vol. 38, no. 4, pp. 257–259, 2003.
[25] L. Liu, H. P. Zhao, T. S. P. See, and Z. N. Chen, “A printed ultrawideband diversity
antenna,” in Proc. Int. Conf. for Ultra-Wideband, Sep. 2006, pp. 351–356.
[26] S. Hong, K. Chung, J. Lee, S. Jung, S. S. Lee, and J. Choi, “Design of a diversity
antenna with stubs for UWB applications ,” Microw Opt Techno Lett., vol. 50, no. 5, pp.
1352–1356, 2008.
[27] E. Antonino-Daviu, M. Gallo, B. Bernardo-Clemente and M. Ferrando- Bataller,
“Ultra-wideband slot ring antenna for diversity applications,” Electron. Lett., vol. 46, no.
7, pp. 478-480, 2010.
[28] Y. Lu and Y. Lin, “A compact dual-polarized UWB antenna with high port isolation,” in
Proc. IEEE Int. Symp. Antennas Propagation Society International Symposium
(APS’2010), Toronto, Canada, Jul. 11-17, 2010.
[29] T. S. P. See, and Z. N. Chen, “An ultrawideband diversity antenna,” IEEE Trans.
Antennas Propag., vol. 57, no. 6, pp. 1597–1605, Jun. 2009.
27
Chapter 4
Correlation Reduction with Improved Total
Efficiency for the Low Frequencies of the
Mobile Handset MIMO/Diversity Antennas
In current and future wireless telecommunications systems, such as the long-term
evolution (LTE) and LTE-Advanced, the multiple-input and multiple-output (MIMO) systems
are an integral part of mobile terminals. In the LTE standards, several new channels are
allocated to the lower bands of 700-960MHz. As mentioned in Chapter 2, the elements in the
MAS should have a low correlation and a high total efficiency to guarantee a good
multiplexing MIMO performance. Unlike the higher bands, the mobile handset MAS
operating in the lower frequencies will not focus on the reduction of the mutual coupling but
rather the improvement of the correlation and efficiency directly due to the low radiation
efficiency [1]. The wavelengths in the lower frequencies are much longer than those in the
higher bands and this poses some new challenges on the practical realization of the good
MIMO performance in mobile terminals: (1) each MIMO antenna element has to be
redesigned to obtain a compact structure of the device; (2) the structures for decorrelation
have to be small enough and still work well; (3) the MIMO elements and the decorrelating
structures are more closely positioned, causing high correlation and low efficiencies; (4) the
chassis mode will be efficiently excited, which makes the radiation pattern of each MIMO
element quite similar leading to a very high correlation. An envelope correlation coefficient
less than 0.5 and a total efficiency higher than 40% are good values for cellular LTE MIMO
antennas in the lower bands according to industry researches, including field trials and mock
ups [2]. Some studies have been done trying to solve these problems such as: the
neutralization line method for the single band LTE MIMO antenna in [3], the decoupling
networks for the lower bands (i.e., 900 MHz) in [4][5]. However, these methods can only be
used for very narrow bands and will cause a large radiation efficiency reduction in practice.
In this chapter, the mutual scattering mode and diagonal antenna-chassis mode will be
introduced to solve these problems. As one example of the utilization of these two modes, a
multiple wideband LTE MIMO antenna will be proposed with a low correlation and high
efficiency for mobile terminal applications.
4.1 Reduction of the Envelope Correlation Coefficient with Improved
Efficiency for Mobile LTE MIMO Antenna Arrays: Mutual Scattering
Mode
28
Generally, the correlation of a MIMO antenna system is determined by the antenna
element types and element distance. The mutual scattering mode in paper V illustrates that
the antenna Q factor is another important parameter strongly affecting the correlation.
Normally, in any given MIMO antenna array, the inter-element distance, the element types,
and the radiation efficiency are determined beforehand. In practice, the Q factors can be
straightforwardly tuned through different input impedance matching. The correlation of a
lossy LTE MIMO antenna array can be reduced efficiently, even down to zero, by increasing
the Q factors of the MIMO antenna elements. The Q factors in this chapter are calculated by
[24, Eq. 96]. The resistance and reactance of the input impedance required in [24, Eq. 96] is
obtained from [19, Eq. 10], where the load on the non-operating port is set to 50 ohm.
4.1.1 Dual Monopoles on a Large Ground Plane with High Losses
Figure. 4. 1. Geometry of dual monopoles on a large ground plane with an inter-element distance of 50
mm.[Paper V]
In Fig. 4.1, dual monopoles on the large lossy ground plane (100 S/m) are analyzed first
with an inter-element distance of 50 mm. As we go from Matching 1 to Matching 5, the Q
factor (see Fig. 4. 2(a)) increases and the envelope correlation coefficients (ECC) are
decreasing (see Fig. 4. 2(c)) with the improved total efficiency (TE) (see Fig. 4. 2(d)). This is
because by increasing the Q factor, each element will become a scatter to the other elements.
Consequently, without adding any decorrelating structure into the MIMO system, the
radiation patterns in Fig. 4. 3 are separated automatically achieving a low correlation.
When the distance between dual monopoles decreases to 20 mm, the higher Q factors can
still give lower ECCs, but worse TE. The reason is that with a distance of 20 mm, the relative
contribution of S11/S22 in improving the total efficiency is slower than that of S21/S12 in
reducing the total efficiency. Hence, a new distance of the Critical Distance is defined, where
the two opposite contributions of S11/S22 and S21/S12 to the TE are equal. When the
distance between MIMO elements exceeds the Critical Distance, a better impedance
matching (or higher Q factor) can not only reduce the correction but also improve the total
efficiency.
29
(a) (b)
(c) (d)
Figure. 4. 2. (a) Q factors, (b) S parameters, (c) envelope correlation coefficients (ECC), and (d) radiation
efficiency (RE) and total efficiency (TE) for different impedance matching levels of dual monopoles on a large
ground plane with an inter-element distance of 50 mm. [Paper V]
Figure. 4. 3. Radiation patterns (Realized Gain): (a) Antenna 1(Port 1) in Matching 5; (b) Antenna 2 (Port 2)
in Matching 5; (c) Antenna 1 in Matching 1; and (d) Antenna 2 in Matching 1. [Paper V]
30
4.1.2 Dual-PIFA MIMO Antennas on Mobile Chassis
Figure. 4. 4. The geometries of collocated dual PIFAs for MIMO applications. (unit: mm) [Paper V]
(a) (b)
(c) (d)
31
(e)
Figure. 4. 5. (a) Q factors, (b) S parameters, (c) envelope correlation coefficients (ECC), (d) radiation efficiency
(RE) and total efficiency (TE) and (e) multiplexing efficiency (ME) for different impedance matching levels of
dual PIFAs. [Paper V]
The Critical Distance is also investigated in mobile terminals. The dual MAS elements
are collocated at the same end of the chassis with an inter-element distance of 10 mm, as
shown in Fig. 4.4. The Q factors, S parameters, ECCs and TEs are presented in Fig. 4. 5 (a),
(b), (c) and (d), respectively. We find that the inter-element distance of 10 mm is larger than
the Critical Distance: a higher Q factor leads to a lower correlation and higher efficiency.
Therefore, the Critical Distance is not determined by the physical distance but the current
distributions. A small inter-element distance is also possible to achieve quite a large Critical
Distance.
The Multiplexing Efficiencies (MEs) of the dual PIFAs with different Q factors
(impedance matchings) are studied and given in Fig. 4.5 (e). With the Q factor increasing, an
improvement of ME can be obtained efficiently.
Figure. 4. 6. Radiation patterns (Realized Gain) of dual PIFAs: (a) Antenna 1(Port 1) in Matching 5; (b)
Antenna 2 (Port 2) in Matching 5; (c) Antenna 1 in Matching 1; and (d) Antenna 2 in Matching 1. [Paper V]
32
Like the dual-monopole with 50 mm case, the radiation patterns of dual PIFAs will be
changed when the Q is greater. In Fig. 4.5, the radiation patterns are more scattered in
Matching 5 than in Matching 1.
When the two MAS elements are allocated at the two ends of the mobile chassis the
higher Q factor will lead to the lower efficiency. Therefore, the inter-element distance is less
than the Critical Distance in this case.
As an example of the wideband application of the method described above, a wideband
LTE MIMO antenna is shown in Fig. 4.7. As we can see, the MIMO array volume is kept the
same as the one in Fig. 4.4. However, one additional branch for each PIFA has been
introduced to generate an extra resonance at the lower frequencies. The losses of capacitors
and inductors are 0.1 ohm and 0.2 ohm, respectively. In Fig. 4.8, the S Parameters, ECC and
Efficiency are presented. We find that the proposed MIMO antenna covers the bands of
746-870 MHz (15.35% fractional bandwidth) with an ECC less than 0.5 and a total efficiency
greater than -3 dB.
Figure. 4. 7. The geometry of the wideband collocated MIMO antenna with mutual scattering mode. [Paper V]
(a) (b)
33
(c)
Figure. 4. 8. (a) S parameters, (b) envelope correlation coefficient, and (c) efficiency of the proposed wideband
collocated MIMO antenna. [Paper V]
4.1.3 Experiment
(a)
(b) (c)
Figure. 4. 9. (a) prototypes, (b) measured S parameters and (c) measured envelope correlation coefficients
(ECC), of fabricated dual PIFAs with different matching levels. [Paper V]
34
To verify the simulated results, some prototypes of collocated dual PIFAs with different
matchings (as well as Q factors) are fabricated and are shown in Fig. 4.9 (a). From the
measured S parameters and ECCs, as presented in Fig. 4.9 (b) and (c), respectively, it can be
obtained that the conclusions from the measurements agree with those from the simulations.
A generally available mutual scattering mode has been introduced. This mode can be
excited through increasing the Q factors of MIMO antenna elements. In practice, for a given
array, the only two requirements for low correlation and improved efficiency are a high Q
factor (matching) and an element distance larger than the Critical Distance. This method
has the following advantages:
(1) Easy to realize: impedance matching techniques (including impedance matching network)
have been successfully utilized in industry for many years.
(2) No specific requirement for geometry of each MIMO antenna element (i.e. they don’t
need to be the same and may have arbitrary structures).
(3) Not only valid for the single band, but also for wide band and multiple bands: in this
paper we have studied a single band case. Actually, in the lower bands the antennas can have
multiple resonances. A wide band and low correlation MIMO antenna can be proposed with
improved efficiency.
(4) Easy to use together with other known decoupling methods: our proposed method only
requires a high Q factor or good matching to achieve decorrelation.
(5) Possible to use for the MIMO antennas with more than two elements: however, due to
more coupling elements, the Critical Distance will become larger compared to that of
dual-element MIMO antennas. Therefore, if the design purposes are reducing correlation as
well as improving efficiency, the element distances should be larger than that in the
dual-element case.
(6) The conclusions from this paper are also valid for several kinds of MIMO antennas,
including the on-ground kind.
4.2 Diagonal Antenna-Chassis Mode and Its Application for Wideband LTE
MIMO Antennas in Mobile Handsets
The mobile chassis will become an effective radiator when a small self-resonant antenna
operates in the lower bands [6] [7]. In [8] [9] it is discovered that the chassis modes are only
dependent on the shape of the chassis. Therefore, it is very easy for the MAS elements
operating at lower frequencies to excite the same chassis mode and then result in similar
radiation patterns, which will dramatically degrade the MIMO/diversity performance. The
effects of chassis modes on the MIMO antenna in the higher bands (around 2.4 GHz) and
lower bands are investigated in [10]-[12] and [13] [14], respectively. Generally speaking,
there are three ways to reduce the chassis mode effects: (1) only make one of the MIMO
antenna elements excite the chassis mode to achieve a good bandwidth, and decouple the
other MIMO element from the chassis [14]. However, this method leads to clearly smaller
bandwidth of the MIMO element without the chassis mode, typically less than 1% at 900
35
MHz [7]. Furthermore, the method with decoupling of the chassis mode in [9] is not so
practical and the collocated MIMO antenna case is not investigated. (2) Excite the orthogonal
chassis mode for different MIMO antenna elements. The research in [12] has used this
method, but only in the higher bands. In the lower bands, due to the geometry of the mobile
chassis the number of chassis modes is quite limited and it is even more difficult to find the
orthogonal modes to use. (3) Utilize mutual interactions of MIMO elements with high Q
factor to separate the radiation patterns. This method is not limited to the mobile chassis and
has been studied in Paper V. In addition, other specific LTE MIMO antenna designs have
been proposed in [15]-[18].
In Paper VI a diagonal antenna-chassis mode is introduced. Similar to the definition of
the bandwidth in [19], in Paper VI a MIMO bandwidth is defined as the overlap bandwidth of
the ECC lower than 0.5, total efficiency higher than -4 dB and impedance matching better
than -6 dB at the frequencies lower than 960 MHz, while the ECC is lower than 0.1, total
efficiency higher than -2 dB and impedance matching better than -6 dB at frequencies higher
than 1700 MHz.. By properly exciting the diagonal antenna-chassis mode, the MIMO
bandwidth can be enhanced efficiently at frequencies lower than 960 MHz. This is realized
through moving the three bandwidths to the same range and enlarging the low-ECC
bandwidth without the degradation of impedance bandwidth and total efficiency.
4.2.1 Collocated Dual Monopoles
The diagrams of the diagonal antenna-chassis mode in the collocated arrangement for 700
MHz bands and 800 MHz-960 MHz bands are shown in Fig. 4.10 (a) and (b), respectively.
When one antenna operates in bands lower than 960 MHz, the antenna itself and the chassis
can be treated as two arms of a half wavelength dipole. If we move the antenna from the
middle to one side of the short chassis edge, the arm on the chassis will lean in the diagonal
direction. As the radiation pattern is mainly determined by the longer arm (on the chassis) of
a dipole, the dipole-type radiation pattern will consequently also be slanted. If we put another
electrical small antenna on the other side of the short edge to form a collocated MIMO
antenna array, the two MIMO elements will have two crossed dipole-like radiation patterns.
Furthermore, due to the mutual coupling between two elements, the coupled element will
become one part of the ground plane to extend the current component in the short edge
direction. This will lead to a ground plane “wider” than its actual width, which will further
enlarge theθvalue (see Fig. 4.10). With these properties a low correlation coefficient of
MIMO antennas will occur.
In order to verify the theory, two collocated MIMO monopole arrays with near feedings
(NF) (d=14) and far feedings (FF) (d=50) are proposed in Fig. 4. 11. Considering the mutual
scattering effects in Paper V, the matching levels for the NF and FF MIMO monopoles are
kept the same (-13 dB). The matching level and central frequency can be tuned by the shunt
capacitor of the port and the series inductor (see Fig. 4.11), respectively. 740 MHz is selected
as the central frequency for the 700MHz bands. The detailed inductor and capacitor values
for all the designs can be found in Paper VI.
36
(a)
(b)
Figure. 4. 10. The diagonal antenna-chassis mode in the collocated arrangement for (a) 700 MHz bands and (b)
800 MHz-960 MHz bands. [Paper VI]
The S Parameters, ECCs and efficiencies of the dual monopoles with NF and FF at the
central frequency of 740 MHz are shown in Fig. 4.12 (a), (b), and (c), respectively. In Fig.
4.12 (a), we can find that the impedance bandwidth and mutual coupling (S21 in S
Parameters) of the FF are wider and stronger than those of the NF. The reasons are as follows:
The FF has a longer on-chassis dipole arm, which can improve the impedance bandwidth.
Meanwhile, the larger θ(see Fig. 4.10(a)) in the FF will also cause a stronger x-component
current (see Fig. 4.11) and higher voltage difference between two ports, and thus the mutual
coupling becomes stronger. As we expect, the larger θ in the FF can make the null and the
0.5-specified ECC bandwidth deeper and wider, respectively, as illustrated in Fig. 4.12 (b).
One can also notice that compared to the NF case the null of ECC in the FF moves to a lower
frequency. From our study, either the longer on-chassis dipole arm or the larger θ value
will result in this lower frequency shift of the ECC null. In the FF case, both the on-chassis
arm and θ have increased due to the feeding locations. In addition, because of the stronger
mutual coupling in FF, the width of the ground plane can be “extended” more to further
37
enlarge the on-chassis arm and θ. Consequently, in the FF the ECC null can be shifted to
lower frequencies dramatically. In Fig. 4. 12 (c), from the FF to the NF, the peak of total
efficiency will move to a higher frequency (the opposite behavior as ECC). Therefore, the
diagonal antenna-chassis mode is necessary, but if it is too strong or too weak this kind of
mode may result in the mismatch between the best ranges of the total efficiency, ECC and
impedance bandwidth. In practice, the optimal goal is to select a feeding point to make the
peak of total efficiency, the null of S11and ECC match at the same frequency. A maximum
MIMO bandwidth can be achieved. In addition, although the mutual couplings are available
here, the total efficiencies in Fig. 4.12 (c) are still not lower (even higher) than those in
[11]-[14]. The reason is that in our proposed method no additional decoupling structures and
losses are introduced into the MIMO array and thus the radiation efficiencies in our methods
are much higher than those in [11]-[14].
Figure. 4. 11. The geometries of the collocated dual-monopole MIMO antenna with: (a) near feedings (d= 14)
and (b) far feedings (d= 50). (Unit: mm) [Paper VI]
(a) (b)
38
(c)
Figure. 4. 12. The simulated (a) S Parameters, (b) envelope correlation coefficients and (c) efficiencies of
different collocated MIMO antennas at the central operating frequency of 740 MHz. [Paper VI]
Figure. 4. 13. Current distributions at 740 MHz for dual monopoles with (a) NF and (b) FF. [Paper VI]
The current distributions (when port 1 operates) at 740 MHz for dual monopoles with NF
and FF are shown in Fig. 4.13 (a) and (b), respectively. It can be observed that the actual
diagonal antenna-chassis mode is the superposition of current path A and B. Both the NF and
FF case can excite the diagonal antenna-chassis mode, but the current path A in FF (see Fig.
4.13 (b)) is longer than that in NF (see Fig. 4.13 (a)). Therefore, path C in FF can lean more
and achieve a lower correlation.
The simulated 3D radiation gain patterns at 740 MHz are provided in Fig. 4.14. We can
see that the two ports in the FF have more severely crossed patterns than those in the NF,
especially at the elevation angle of 0-50 degrees.
39
Figure. 4. 14. 3D radiation patterns for dual monopoles with FF and NF at 740 MHz. [Paper VI]
4.2.2 Collocated Dual PIFAs
Figure. 4. 15. The geometries of the collocated dual-PIFA MIMO antennas with: (a) near shortings (d= 20) and
(b) far shortings (d= 40). (Unit: mm) [Paper VI]
In mobile handsets, PIFAs are another kind of commonly used antennas that can more
easily achieve wide band at lower frequencies through multiple resonances in a small volume
than monopoles, i.e. the antenna in [20]. Furthermore, different from monopoles, one PIFA
has two excitations on the ground plane-the feeding and the shorting, which will greatly
affect the antenna-chassis mode. Therefore, it is very important to investigate the diagonal
antenna-chassis mode in the collocated dual-PIFA MIMO antennas.
40
In order to simplify the research model, we fix each feeding close to one side of the short
chassis edge and only study the locations of shorting points. The dual-PIFA MIMO antennas
with near shorting (NS) points (d=20) and far shorting (FS) points (d=40) are proposed and
shown in Fig. 4.15. Two groups of matching levels of -13dB and -10 dB are selected at 740
MHz. The matching level and central frequency can be tuned by the series capacitor (see Fig.
4.15) of the port and the series inductor of the shorting pin, respectively. The detailed
inductor and capacitor values for all PIFAs with the NS and FS are given in Paper VI.
The S Parameters, envelope correlation coefficient (ECC) and efficiency in the 700 MHz
bands are shown in Fig. 4.12. In Fig. 4.12 (a) it can be observed that similar to the monopole
case, for each matching level the MIMO PIFAs with FS always have a wider bandwidth. The
mutual couplings in all the dual PIFAs are quite similar, and lower than those in collocated
dual monopoles due to the availability of shorting pins. The conclusions from the ECC in Fig.
4.12 (b) are similar to the case with dual monopoles. However, one can also find that the null
of ECC with FS does not move to a lower frequency as much as that in the case of FF dual
monopoles. The reason is that due to the insignificant increase of the mutual coupling from
FS to NS in PIFAs the width of the ground plane is not “extended” further. In addition, due to
the mutual scattering mode in [10] the nulls of MIMO PIFAs with FS and NS in -13 dB
matching level are deeper than those in -10 dB matching level. However, the PIFAs with FS
and -10 dB matching level have even wider 0.5-ECC bandwidth than those with NS and -13
dB matching level. This property can reduce the matching level requirement for low ECC in
[10] and decrease the Q factor to further improve the bandwidth. This can be clearly observed
in the following analysis. In Fig. 4.12 (c) the peaks of total efficiencies move in the same
manner as the MIMO monopoles.
Figure. 4. 16. Current distributions at 740 MHz for dual PIFAs (-10 dB matching) with FS and NS. [Paper VI]
41
The current distributions (when port 1 operates) for dual PIFAs (-10 dB matching) with
NS and FS at 740 MHz are shown in Fig. 4.16. The dual PIFAs with FS (see Fig. 4.17 (a))
have the similar current distributions to those in the dual monopoles, which means that if the
feeding and shorting point of each PIFA element are both located close to one corner of the
short edge, a diagonal antenna-chassis mode similar to that in the monopole case can be
excited successfully. Meanwhile, from current distributions for the dual PIFAs with NS in Fig.
4.16 (a) we can know that the chassis mode has already degraded to the traditional one which
has been well studied in [6]-[14]. For each PIFA the currents from shorting points are out of
phase with those from the feeding (or port). The dual PIFAs are located on the same end of
the chassis where four excitations are available, and all the adjacent excitations are out of
phase. If these four excitations are averagely arranged (i.e., NS case), the currents on path A
would be cancelled out by each other. Therefore, the superposed path C is parallel with the
long edge of the chassis and the diagonal antenna-chassis mode cannot be excited. Due to the
change of superposed current paths, the 3D radiation patterns of the FS and NS dual PIFAs
become quite different, as illustrated in Fig. 4. 17.
Figure. 4. 17. 3D radiation patterns for dual PIFAs (-10 dB matching) with FS and NS at 740 MHz. [Paper VI]
4.2.3 Multiple Wideband LTE MIMO Antenna for Mobile Terminals
As one example for the wideband application of the diagonal antenna-chassis mode, a
multiple wideband collocated LTE MIMO antenna is proposed for mobile terminals, as
shown in Fig. 4.18. The MIMO array is mounted on a FR4 substrate with a loss tangent of
0.025 and a permittivity of 4.4. The whole array is quite compact with a total volume of only
60 mm× 12 mm × 7 mm. A 1.5 mm- thick plastic mobile phone-cover with a total volume of
130 mm×10 mm×65 mm is used with a 1 mm gap around the antenna array to simulate the
practical situations. The shorting points have been carefully selected to excite a proper
amount of the diagonal antenna-chassis mode in order to achieve the maximum MIMO
bandwidth in the lower bands. For each PIFA, there are two long arms, which can provide
two resonances in the lower bands, and the inter-digital structure is utilized to tune the
42
matching of these two resonances. This can be observed from the S parameters in Fig. 4.19 (a)
and a band of 740-960 MHz can be covered. Since there is no other structures between two
PIFAs a metal strip operating around 1700 MHz can be inserted as a parasitic decoupling
strip [21]. Furthermore, this metal strip together with the (far) shorting pins of the PIFAs can
also form two slots, as illustrated in (1) in Fig. 4.18, and can provide resonances around 1700
MHz for the impedance matching. Therefore, due to the introduction of the metal strip and
far shortings, the combination of slot-monopole-slot is formed, where both the impedance
matching and decoupling property will occur simultaneously at 1700 MHz (Fig. 4.19 (a)).
Similar phenomena can also be found in [22] and [23] but they are monopole-slot-monopole
combinations. In order to further improve the bandwidth in higher bands (greater than 1700
MHz), the feeding line (2) (see Fig. 4.18) of each PIFA is optimized to provide another
resonance at 2500 MHz. This way, in the higher bands, both PIFAs can cover a band of
1700-2700 MHz with isolation greater than 15 dB, as shown in Fig. 4.19 (a).
Figure. 4. 18. The geometries of the wideband collocated dual-PIFA MIMO antennas. (Unit: mm) [Paper VI]
The ECC and efficiencies in the operating bands are given in Fig. 4.19 (b) and (c),
respectively. In the lower bands, because of the diagonal antenna-chassis mode, the
bandwidths of 0.5 ECC and -4 dB total efficiencies have been tuned into the same range as
that of the impedance bandwidth. In the higher bands, since the mutual coupling is quite low,
within the 1700-2700 MHz band, the ECC and total efficiency are better than 0.5 and -2 dB,
respectively.
Summarizing the discussions above, the proposed multiple wideband MIMO antennas
can cover a MIMO bandwidth of 740-960 MHz and 1700-2700 MHz within a very small
volume.
The measurements have also been carried out for the fabricated multiple wideband
MIMO antennas and compared with the simulations, as presented in Fig. 4. 19. A good
agreement can be found there, which verifies the correctness and effectiveness of applying a
diagonal antenna-chassis mode into the practical wideband MIMO antenna design.
43
(a) (b)
(c)
Figure. 4. 19. The simulated (a) S Parameters, (b) envelope correlation coefficients and (c) efficiencies for
different wideband collocated dual-PIFA MIMO antennas. [Paper VI]
4.2.4 Separately located LTE MIMO Antenna
(a) (b) (c) (d)
Figure. 4. 20. The diagonal antenna-chassis mode of the adjacent corner located arrangement for (a) 700-800
MHz bands and (b) 800-960 MHz bands. The diagonal antenna-chassis mode of the diagonal corner located
arrangement for (c) 700-800 MHz bands and (d) 800-960 MHz bands. [Paper VI]
44
If two LTE MIMO antenna elements are separately located on the adjacent corners, the
diagrams for 700 MHz bands are shown in Fig. 4.20 (a) and (b), respectively. It can be
observed that the diagonal antenna-chassis mode and crossed current paths of two MIMO
elements are still available. It is different from the collocated case, due to the mutual coupling
the ground plane extension is along the long chassis edge, which will reduce the θ value in
Fig. 4.20 and weaken the diagonal antenna-chassis mode. For each frequency, the stronger
diagonal antenna-chassis mode will lead to a shorter on-chassis dipole arm and larger θ in
Fig.4.21. This will result in the following phenomena: the frequency shift effects on the ECC
null will be cancelled out but the ECC can still be reduced (due to the larger θ). Due to the
weaker long-edge current components, the mutual coupling and the total efficiency can be
improved. In summary, in the adjacent corner located case, the diagonal antenna-chassis
mode should be made as strong as possible to achieve a good MIMO bandwidth. In addition,
compared with the collocated case, the diagonal antenna-chassis mode in this arrangement is
quite weak, but due to the much larger inter-element distance the performance will not
degrade so much.
The diagram of the diagonal antenna-chassis mode in the diagonally located arrangement
is shown in Fig. 4.20 (c) and (d). It can be clearly observed that the on-chassis dipole arms of
MIMO elements are parallel with each other. It means that the radiation patterns of two MAS
elements will be similar causing a high ECC and low efficiency. Therefore, the diagonal
corners are the worst located arrangement for MIMO bandwidth. In this arrangement the
diagonal antenna-chassis mode should be as weak as possible to enlarge the distance of the
two parallel on-chassis dipole arms in order to improve performance.
References
[1] Hallbjorner, P., “The significance of radiation efficiencies when using S-parameters to
calculate the received signal correlation from two antennas,” IEEE Antennas Wireless
Propag. Lett., pp. 97–99, 2005.
[2] Z. Ying, “Antennas in cellular phones for mobile communications,” Proceedings of the
IEEE, vol. 100, no. 7, pp. 2286-2296, Jul. 2012.
[3] H. Bae, F. J. Harackiewicz, M. Park, T. Kim, N. Kim, D.Kim, and B. Lee, “Compact
mobile handset MIMO antenna for LTE700 applications”, Microwave Opt. Technol. Lett.,
vol. 52, no. 11, pp. 2419–2422, Nov. 2010
[4] B. K. Lau and J. B. Andersen, “Simple and efficient decoupling of compact arrays with
parasitic scatterers,” IEEE Trans. Antennas Propag., vol. 60, no. 2, pp. 464‐472, Feb.
2012.
[5] K. Karlsson, and J. Carlsson, “Analysis and optimization of MIMO capacity by using
circuit simulation and embedded element patterns from full-wave simulation," iWAT 2010,
Lisban, Portugal, Mar., 2010.
45
[6] P. Vainikainen, J. Ollikainen, O. Kivekäs, and I. Kelander, “Resonator- based analysis
of the combination of mobile handset antenna and chassis,” IEEE Trans. Antennas
Propag., vol. 50, no. 10, pp. 1433–1444, Oct. 2002.
[7] J. Villanen, J. Ollikainen, O. Kivekas, and P. Vainikainen, “Coupling element based
mobile terminal antenna structure,” IEEE Trans. Antennas Propagat., vol. 54, no. 7, pp.
2142-2153, Jul. 2006.
[8] R. F. Harrington and J. R. Mautz, “Theory of Characteristic modes for conducting bodies,”
IEEE Trans. Antennas Propagat., vol.19, no.5, pp. 622-628, Sep. 1971.
[9] M. C. Fabres, E. A. Daviu, A. V. Nogueiram, and M. F. Bataller, “The theory of
characteristic modes revisited: a contribution to the design of antennas for modern
applications,” IEEE Trans. Antennas Propagat. Magazine, vol. 49, no. 5, pp. 52-68, Oct.
2007.
[10] S.K. Chaudhury, H.J. Chaloupka, and A.Ziroff, “Novel MIMO Antennas for Mobile
Terminal,” EuMC2008, Amsterdam, Netherlands, Oct. 2008.
[11] R. Martens, E. Safin, and D. Manteuffel, “Selective excitation of characteristic modes
on small terminals,” EUCAP2011, Rome, Italy, Apr. 2011.
[12] D. Manteuffel, and R. Martens, “A concept for MIMO antennas on small terminals based
on characteristic modes,” iWAT2011, Hong Kong, China, Mar., 2011.
[13] A. Krewski, W. L. Schroeder, K. Solbach, “Bandwidth limitations and optimum
low-band LTE MIMO antenna placement in mobile terminals using modal analysis,”
EUCAP2011, Rome, Italy, Apr. 2011.
[14] H. Li, Y. Tan, B. K. Lau, Z. Ying, and S. He, “Characteristic mode based tradeoff
analysis of antenna-chassis interactions for multiple antenna terminals,” IEEE Trans.
Antennas Propagat., vol. 60, no. 2, pp. 490–502, Feb. 2012.
[15] Y. K. Hong ; S. Bae ; G.S. Abo, W. M. Seong, and G. H. Kim, “Miniature Long-Term
Evolution (LTE) MIMO Ferrite Antenna,” IEEE Antennas Wireless Propag. Lett., vol. 10,
pp. 603-606, Jun. 2011.
[16] J. Y. Chung, T. Yang, J. Lee, and J. Jeong, “Low correlation MIMO antenna for LTE
700MHz band,” APSURSI2011, Spokane, America, Jul. 2011.
[17] S. Lee, J. W. Lee, “Internal MIMO antenna to selectively control isolation characteristic
by isolation aid in multiband including LTE band,” APMC2010, Yokohama, Japan, Mar.
2010.
[18] R. G. Vaughan and J. B. Andersen, “Antenna diversity in mobile communications,”
IEEE Trans. Veh. Technol., vol. 36, no. 4, pp. 149–172, Nov. 1987.
[19] B. K. Lau, J. Bach Andersen, G. Kristensson, and A. F. Molish, “Impact of matching
network on bandwidth of compact antenna arrays”, IEEE Trans. Antennas Propag., vol.
54, no. 11, pp. 3225-3238, 2006.
[20] S. Jeon, S. Oh, H.H. Kim and H. Kim, “Mobile handset antenna with double planar
inverted-E (PIE) feed structure,” Electon.Lett., vol. 48, no. 11, pp. 705–707, May. 2012.
[21] A. C. K. Mak, C. R. Rowell and R. D. Murch, “Isolation enhancement between two
closely packed antennas,” IEEE Trans. Antennas Propag., vol. 56, no. 11, pp. 3411–3419,
Nov. 2008.
46
[22] S. Zhang, B. K. Lau, Y. Tan, Z. Ying, and S. He, “Mutual coupling reduction of two
PIFAs with a T-shape slot impedance transformer for MIMO mobile terminal,” IEEE
Trans. Antennas Propag., vol. 60, no. 3, pp. 1521-1531, 2012.
[23] S. Zhang, B. K. Lau, A. Sunesson, and S. He, “Closely-packed UWB MIMO/diversity
antenna with different patterns and polarizations for USB dongle applications,” IEEE
Trans. Antennas Propag., vol. 60, no. 9, pp. 4372-4380, Sep. 2012.
[24] A. D. Yaghjian and S. R. Best, “Impedance, bandwidth, and Q of antennas,” IEEE Trans.
Antennas Propag., vol. 53, no. 4, pp. 1298–1324, Apr. 2005.
47
Chapter 5
Body Loss Investigation and User-Effect
Reduction for LTE MIMO Antennas in Mobile
Terminals
In Chapter 4, the methods of proposing a high performance wideband LTE MIMO system
have been introduced. The low envelope correlation coefficient and high total efficiencies can
be realized in free space. However, in practice the interactions between the mobile terminal
and the user will reduce the total efficiency of handset antennas by shifting the resonant
frequencies and absorbing some of the radiated/received power. The effects of the user’s
hand and head on single mobile terminal antennas have been studied in [1]-[3]. The effect of
the user’s hands on the operation of lower UHF-band mobile terminal antennas can be found
in [4]. Unfortunately, the research in [1]-[4] only focus on the user effects on a single antenna.
A dual-element MIMO antenna array with hand effects has been studied in [5] for the
diversity performance at 2 GHz in mobile terminals. Another research about actual diversity
performance of a multiband diversity antenna with hand and head effects can be found in [6].
The diversity mobile antenna elements utilized in [6] are one on-ground (OG) PIFA and one
ground-free PIFA, located at each end of the chassis. However, the user effects on MIMO
channel performance and body loss have not been well investigated for the multi-wide bands
of 750-960 MHz and 1700-2700 MHz. Other different types of MIMO antennas (collocated
GF MIMO, orthogonal OG MIMO, and parallel OG/GF MIMO antennas) have not been
studied. Furthermore, the methods of reducing the user effects still are not presented in [6].
5.1 Body Loss and MIMO Performance Investigation of the Different LTE
MIMO Antenna Types with the User Effects for Mobile Terminals
In Paper VII, the body loss and MIMO performance of MIMO antennas with user effects
are studied over different mobile terminal lengths (90mm, 110mm, 130mm and 150mm).
Different MIMO antenna types are investigated, as shown in Fig. 5.1. Three kinds of user
effects, namely, SAM head and PDA hand (talk mode), PDA hand (data mode), and dual
hands (read mode) are utilized to simulate the actual user positions. The model of the user
effects are presented in Fig. 5.2. The relative positions of the whole MIMO antenna array and
the user’s body (SAM head and PDA hand; PDA hand) are in accordance with the CTIA
revision 3.1 [7]. For the case of dual hands, since there is no set standard yet, the whole
antenna array and dual hands are arranged in the commonly handsets-holding way. The body
loss studies of the four kind of MIMO antennas are carried out at four frequency points of
48
0.75, 0.85, 1.9 and 2.6 (or 2.1) GHz, as well as on different MIMO modes (one receiving/one
transmitting, and simultaneously transmitting/receiving). The dielectric properties of the
human tissue used in this paper can be found in [8]. The body loss is calculated by the
definition formula:
Body Loss= Radiation Efficiency free space (dB) – Radiation Efficiency user effect (dB)
(5.1)
Figure. 5. 1. Diagrams of different types of MIMO antennas.
(a) (b)
(c)
Figure. 5. 2. The models for (a) talking mode (head and hand), (b) data mode (single hand) and (c) reading
mode.
49
The detailed analyses and figures can be found in Paper VII. It is discovered that the body
losses and MIMO performances with the user effects are highly affected by the ground status
(ground free or on ground), element locations (collocated, separately located or orthogonally
located) and feeding positions. If an adaptive function is exciting to change these factors
above (ground, element and feeding locations) according to the user positions, the body loss
can be efficiently reduced and hence MIMO performance can also be improved. Furthermore,
the phase of two MIMO antennas will also result in different body loss and some phases
causing high body losses should be avoided in practice.
5.2 Adaptive Quad-Element Multi-Wideband Antenna Array for
User-Effective LTE MIMO Mobile Terminals
Figure. 5. 3. Geometry of the proposed quad-element LTE MIMO antenna array. [Paper VIII]
(a) (b)
50
(c) (d)
Figure. 5. 4. (a) The S parameters, (b) Total efficiency, (c) Envelope correlation coefficients and (d)
Multiplexing efficiency, of the different dual-element combinations in the adaptive quad-element LTE MIMO
antenna array. [Paper VIII]
In Paper VIII an adaptive concept is introduced for the user-effect reduction. This
function can be realized through a quad-element multi-wideband MIMO antenna array, as
given in Fig. 5.3. With the existence of human body, the best two elements out of four will be
selected as a dual-element MIMO array, with the other two ports and grounding points open.
The performances of the proposed quad-element MIMO array in free space, talking mode,
data mode and reading mode can be obtained from Fig. 5.4, Fig. 5.5, Fig.5.6 and Fig. 5.7,
respectively.
From the total efficiency, envelope correlation coefficient and multiplexing efficiency for
the three kinds of user effects, some general rules and physical explanations can be given as
follows (not restricted only to our proposed MIMO antenna array):
1. In the talk mode, compared to the PDA hand, the SAM head will bring more efficiency
loss for the Antenna 1 and Antenna 2 position than the position of Antenna 3 and Antenna 4.
2. The total efficiency variations between different combinations in the lower band
(750-960 MHz) are relatively small compared with those in the higher band (1.7-2.7 GHz). In
other words, the lower band efficiency is less sensitive to the body coverage than the higher
band. This is because in the higher band the array is the only main radiating part, while in the
lower band, besides the dual-element array, the ground plane is also a main radiator [9].
3. The envelope correlation coefficient in the lower band can be reduced effectively if the
dual-element combination has the following two characteristics: large human body coverage
of the whole dual-element antenna array, and user hand placed approximately symmetrically
between two ports. The reason is as follows: when the dual-element array satisfies these two
characteristics the user hand can be viewed as a source of scattering. This scattering will
efficiently separate the radiation patterns of the two MIMO antenna elements and
consequently achieve a low correlation.
4. In the lower band, the multiplexing efficiency of the dual-element combination with
the largest coverage of the user’s body can sometimes perform better than the other
combinations. This is somewhat counterintuitive. Low body coverage does not always give a
51
good result. In the higher band, the best performance should result from the combination with
the least body coverage. The mechanism behind this phenomenon is as follows: the
multiplexing efficiency is determined by two factors– total efficiency and the envelope
correlation coefficient. In the lower band, the correlation between two antennas is relatively
high. This is the main reason for the decrease of the MIMO channel capacity. As mentioned
in items 2 and 3, the large body coverage can reduce the correlation very effectively and the
efficiency will not decrease a lot. Thus the large body coverage can sometimes give a better
multiplexing efficiency than the others. In the higher band, the correlation has fallen to a
quite low level and the total efficiency is the main influencing factor on the multiplexing
efficiency. Reducing the body coverage (the loss of efficiency) can directly increase the
MIMO performance. Additionally, if the MIMO antenna elements have a quite low envelope
coefficient in the lower band, it can be expected that the low body coverage will also obtain a
higher multiplexing efficiency than a high coverage. However, due to the limited space in
mobile terminals and the strong chassis mode in the lower band, it is very difficult to make
sure all the dual-element combinations have a low correlation.
(a) (b)
(c) (d)
Figure. 5. 5. (a) Model and antenna elements locations, (b) Total efficiency, (c) Envelope correlation coefficient
and (d) Multiplexing efficiency, of the SAM head and PDA hand (talk mode). [Paper VIII]
52
(a) (b)
(c) (d)
Figure. 5. 6. (a) Model and antenna elements locations, (b) Total efficiency, (c) Envelope correlation coefficient
and (d) Multiplexing efficiency, of the PDA hand (data mode). [Paper VIII]
(a) (b)
53
(c) (d)
Figure. 5. 7. (a) Model and antenna elements locations, (b) Total efficiency, (c) Envelope correlation coefficient,
and (d) Multiplexing efficiency, of the dual hand (read mode). [Paper VIII]
The experiment results agree well with the simulations, as illustrated in Paper VIII. To
summarize the simulation and experimental results, in practical applications, Antenna pair (3,
4) and Antenna pair (1, 2) can be utilized for the talk mode (with the consideration of the
SAR), and data mode, respectively. For the read mode case, Antenna pair (1, 4) and Antenna
pair (2, 3) will be used in the lower and higher bands, respectively.
The quad-element MIMO array in Paper VIII is just one example (or solution). Some
other adaptive concepts also need to be further developed based on the research in Paper VII.
References
[1] W. Yu, S. Yang, C. L. Tang, and D. Tu, “Accurate simulation of the radiation
performance of a mobile slide phone in a hand-head position,” IEEE Antennas Propag.
Mag, vol. 52, no. 2, pp. 168-177, Apr. 2010.
[2] M. Pelosi, O. Franek, M. B. Knudsen, G. F. Pedersen, and J. B. Andersen, “Antenna
proximity effects for talk and data modes in mobile phones,” IEEE Antennas Propag.,
Mag, vol. 52, no. 3, pp. 15-26, Jun. 2010.
[3] J. IIvonen, O. Kivekas, J. Holopainen, R. Valkonen, K. Rasilainen, and P. Vainikainen,
“Mobile terminal antenna performance with the user’s hand: effect of antenna
dimensioning and location,” IEEE Antennas Wireless Propag. Lett., vol. 10, pp. 772-775,
2011.
[4] J. Holopainen, O. Kivekas, J. IIvonen, R. Valkonen, C. Icheln, and P. Vainikainen,
“Effect of the user’s hands on the operation of lower UHF-band mobile terminal antennas:
focus on digital television receiver,” IEEE Trans. Electromagn. Compat., vol. 53, no. 3,
pp. 831-841, Aug. 2011
[5] A. A.-. Azremi, J. Ilvonen, R. Valkonen, J. Holopainen, O. Kivekäs, C. Icheln and P.
Vainikainen, “Coupling element-based dual-antenna structures for mobile terminal with
54
hand effects”. International Journal of Wireless Information Networks, 18(3), pp.146-157,
2011.
[6] V. Plicanic , B. K. Lau, A. Derneryd and Z. Ying, “Actual diversity performance of a
multiband diversity Antenna with hand and head Effects,” IEEE Trans. Antennas Propag.,
vol. 57, no. 5, pp. 1547-1556, May. 2009.
[7] “Test plan for mobile station over the air performance,” CTIA revision 3.1, Jan., 2011.
[8] K. Fujimoto and J. R. James, Mobile Antenna Systems Handbook, 3rd ed. Reading, MA:
Artech House, 2008, ch. 5.
[9] K. Solbach and C. T. Famdie, “Mutual coupling and chassis-mode coupling small phases
array on a small ground plane,” Eur. Conf. Antennas Propag. (EuCAP), Edinburgh, U.K.,
Nov. 11–16, 2007.
55
Chapter 6
Conclusions and Summary of Papers
6.1 Conclusions
In this thesis, we have investigated and enhanced the performance of multiple antenna
systems in compact MIMO/diversity terminals. The background has been introduced briefly.
Several parameters, such as correlation, diversity gain, MIMO capacity, multiplexing
efficiency, etc., have been studied to help evaluate the MAS performance.
For a MAS operating in the higher bands and with high radiation efficiency, the major
topic is how to reduce the mutual coupling between MAS elements. This is because in this
case the low mutual coupling can successfully lead to low envelope correlation coefficients
(between MAS elements) and high total efficiencies (of MAS elements) to improve the
MIMO/diversity performance. A half-wavelength decoupling slot antenna, formed by the
neighboring edges of two PIFAs and a slot on the ground plane, has been introduced. The
coupling current from one element to the other was efficiently blocked by the slot antenna
and the high isolation was achieved. Since there was only one slot between two PIFA, the
inter-PIFA distance could further decrease while maintaining the low mutual coupling. In
order to extend the idea of the decoupling slot into the MAS with the arbitrary sizes and
shapes of the ground plane, a quarter-wavelength decoupling slot antenna with an impedance
transformer were studied. By adjusting the length of the T-shape slot, the decoupling slot
antenna could be excited in any size and shape of the ground plane, and furthermore a
dual-band dual-isolation property was also realized in a dual-PIFA MIMO/diversity antenna
array. A tree-like decoupling parasitic element was inserted between two UWB radiators.
Wideband isolation was achieved through reflecting or blocking the coupling with the
tree-like structure. Different polarizations and radiation patterns of two MAS elements were
also exploited to design a compact UWB MIMO/diversity antenna for USB dongle
applications.
In the lower bands (below 960 MHz), due to the low radiation efficiency and strong
chassis mode, the mutual couple cannot estimate the envelope correlation coefficients
accurately any more. The targets should be changed to how to realize the low correlations and
high total efficiencies directly. A new mode of mutual scattering mode has been investigated.
Utilizing this mode, each MAS element was viewed as a scatterer of the others when the Q
factors of MAS elements increased. The radiation patterns of MAS elements were separated
automatically due to this scattering effect. With the inter-element distance of the MAS larger
than the Critical Distance, a higher Q factor also improved the total efficiency. A wideband
LTE MIMO antenna with multiple resonances was designed in mobile terminals, taking all
the advantages of the properties of this mode. In order to further improve the impedance
bandwidth while maintaining the good MIMO/diversity performance of MAS, another mode
56
of diagonal antenna-chassis mode were introduced. The high Q factors required for the low
correlation and high efficiencies in mutual scattering mode were reduced. Hence, the
bandwidth of wideband LTE MIMO antenna with multiple resonances mentioned above was
enlarged significantly without the degradation of the MIMO/diversity performance.
Human body as a kind of material with high permittivity and losses would highly change
the performance of MAS in a handset terminal. These effects were studied in this thesis under
different MIMO antenna types, chassis lengths, frequencies, port phases and operating modes.
An adaptive quad-element MAS were proposed and investigated under different user
positions. In the applications, the best two elements out of four were selected as a MIMO
antenna array. It was found that: (1) the total efficiency variations between different
combinations in the lower band (750-960 MHz) were relatively small compared with those in
the higher band (1.7-2.7 GHz). (2) The envelope correlation coefficient in the lower band
could be reduced effectively if the whole dual-element antenna array had the large human
body coverage and the user hand was placed approximately symmetrically between two ports.
(3) In the lower band, the multiplexing efficiency of the dual-element combination with the
largest coverage of the user’s body might perform better than the other combinations. Some
other general rules not limited to the designed MAS have also been given.
As for the future work, there are still several important directions or topics in this area.
The correlations and total efficiencies in a MAS operating in the lower bands need to be
further improved in a wider bandwidth. How to further significantly reduce the body loss,
especially in talking mode, still has a long way. MIMO antenna systems, applied to some
other technology, i.e., RFID, body centric communications, also need to be investigated.
6.2 Summary of Papers
Paper I: This paper introduced an efficient isolation enhancement method for a
closely-positioned tunable dual-PIFA array. Low mutual coupling was realized through a λ0/2
folded slot antenna which is formed by a slot etched on the ground plane and the neighboring
edges of the two PIFAs. Direct coupling was trapped by the slot antenna which radiated the
coupling power into free space. A measured mutual coupling of less than -36.5 dB can be
achieved between two PIFAs at 2.45 GHz with a very small inter-element distance.
Paper II: This paper proposed a mutual coupling reduction method for two closely spaced
PIFAs with arbitrary size and shape of the ground plane. Through a T-shape slot impedance
transformer, both single-band and dual-band MIMO PIFAs can be well proposed with high
isolations. The locations of the single-band and dual-band MIMO PIFAs on the ground plane
were also studied. An eight-fold increase in the bandwidth of one PIFA was achieved, when
the single-band PIFAs were positioned at one corner of the ground plane, with the bandwidth
of the other PIFA and the low mutual coupling unchanged.
Paper III: This paper studied a wideband isolation enhancement technology. With a tree-like
structure on the ground plane, a compact printed UWB MAS with a compact size of 35 × 40
57
mm and a wide band of 3.1-10.6 GHz was designed. Within that band, a low mutual coupling
can be achieved. The effectiveness of the tree-like structure was analyzed. Measured results
agreed well with the simulations.
Paper IV: This paper proposed a closely-positioned UWB MAS (of two elements) with a
small size of 25 mm by 40 mm for USB dongle applications. Wideband high isolation was
achieved through the different patterns and polarizations of the two UWB MAS elements.
Moreover, the slot formed between the monopole and the ground plane of the half slot
antenna was utilized to further reduce the mutual coupling at the lower frequencies and to
provide an additional resonance at one antenna element in order to increase its bandwidth.
The underlying mechanisms of the antenna’s wide impedance bandwidth and high isolation
were analyzed in detail. Based on the measurement results, within the operating wide band,
the isolation is higher than 26 dB. Furthermore, a chassis mode was excited when a physical
connection was required between the ground planes of the two UWB MAS elements. With
the performance of the half slot element unchanged, the monopole could cover a band of
1.78-3 GHz, in addition to the UWB band.
Paper V: This paper introduces a mutual scattering mode. Utilizing this mode, the correlation
of a lossy LTE MIMO antenna array can be reduced efficiently, even down to zero, by
increasing the Q factors of the MIMO antenna elements. In practice, the Q factors can be
straightforwardly tuned through different input impedance matching, and the zero correlation
occurs with the Q factor higher than that from the conjugate input impedance matching. On
one hand, when the inter-element distance is larger than a certain distance (what we
denominate as the Critical Distance), the total efficiency could also be improved in addition
to reducing the correlation. On the other hand, when the inter-element distance is less than the
critical distance, a reference MIMO antenna with high correlation and high total efficiency
could be proposed for OTA measurement applications. This mode is investigated for the dual
monopoles with a large lossy ground plane (by a general analysis) and various mobile
terminal MIMO antennas. A wideband MIMO antenna, with multiple resonances, covering
a band of 746-870 MHz is proposed with the envelope correlation coefficient and total
efficiency less than 0.5 and higher than 50% (-3dB), respectively. Measurements and
simulations agree well for all the fabricated prototypes. The envelope correlations and the
multiplexing efficiencies of the prototypes are also investigated in Gaussian distributed
propagation channels.
Paper VI: This paper introduced a diagonal antenna-chassis mode. A parameter of MIMO
bandwidth was defined as the overlap range of the low-envelope correlation coefficient
(ECC), high-total efficiency and -6 dB-impedance matching bandwidths. Utilizing the
diagonal mode, the MIMO bandwidth of the collocated MIMO antennas can be improved
efficiently at frequencies lower than 960 MHz. This was realized through moving the three
bandwidths to the same range and enlarging the low-ECC bandwidth without the degradation
of impedance bandwidth and total efficiency. As a practical example a wideband collocated
58
LTE MIMO antenna was proposed covering the bands of 740-960 MHz and 1700-2700 MHz.
Within the operating lower and higher bands the total efficiencies were larger than -3.4 and
-1.8 dB with the ECC lower than 0.5 and 0.1, respectively. Furthermore, the diagonal
antenna-chassis mode was also investigated in the adjacent corner and diagonal corner
MIMO element locations. Some useful conclusions were given to enlarge the MIMO
bandwidth. Some measurements of the S Parameters, envelope correlation coefficients, total
efficiency and 3D radiation gain patterns were carried, and agreed well with the simulations.
Paper VII: This paper studied the body loss and MIMO performance for LTE MIMO
antennas in mobile handset systems with the effects of different MIMO antenna types, chassis
lengths, phase difference and operating mode (separated mode or simultaneous modes) under
the mobile phone frequency points in 700-960 and 1700-2700 MHz. Three kinds of practical
cases were investigated: SAM head+ PDA hand, single PDA hand, and dual hands. Some
general rules were provided.
Paper VIII: This paper proposed a compact MAS adaptive to the effects of the user’s body for
LTE MIMO mobile terminals. The bands 750-960 MHz and 1700-2700 MHz were covered
with high efficiency in free space. Three kinds of user effects were studied: SAM head and
PDA hand; PDA hand; and dual hands. The array was formed by selecting the best two
elements out of four, with the other two ports and grounding points open. The user effects
were reduced through the spacing of the element feeds on the mobile chassis, and the MIMO
channel capacity was improved by the element selection. The total efficiency, envelope
correlation coefficient, and multiplexing efficiency were presented for the three user effects.
In the lower band, the decreased correlation from the selection action improved the
multiplexing efficiency, and the underlying physical mechanisms were discussed.
Experiments for the three user effects agreed well with the simulation results.