Professional Documents
Culture Documents
net/publication/326284306
CITATIONS READS
0 52
6 authors, including:
Yingsong Li
Harbin Engineering University
244 PUBLICATIONS 1,269 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
International Journal of Electronics and Communications (AEÜ) new scope and Editorial Board View project
All content following this page was uploaded by Yingsong Li on 11 July 2018.
Date of publication xxxx 00, 0000, date of current version xxxx 00, 0000.
Digital Object Identifier 10.1109/ACCESS.2018.Doi Number
ABSTRACT In this paper, an Antenna Interference Cancellation Chip (AICC) with high-pass response is
proposed to mitigate the mutual coupling of two antennas resonating in contiguous frequency bands. Using
the most up-to-date LTCC technology, the chip only occupies a compact volume of 1.6×0.8×0.6 mm3. Two
external tuning capacitors and two shunt inductors together with the LTCC AICC device which are in shunt
with two coupled antennas, are able to improve the antenna isolation near band-edge by more than 15 dB
without sacrificing antenna performance in its useful resonating bands. The high-pass property of the AICC
device prevent the decoupling design near 2.4GHz affecting the successful operation of GPS at 1.575 GHz.
The superiority of the proposed method is verified with active measurement of a Mi-Fi (mobile Wi-Fi)
device in Wi-Fi hotspot mode using the AICC device. The proposed AICC device and corresponding
decoupling method with the device can find plenty of applications in LTE and future 5G wireless platforms.
INDEX TERMS Antenna array mutual coupling, decoupling network, interference suppression, in-device
coexistence (IDC), high-pass filter.
and the LTE Band-40 antenna. Furthermore, today’s off- either the performance of the first few channels of Wi-Fi
the-shelf SAW (surface acoustic wave) or BAW (bulk band [5] or the performance of the last few channels of LTE
acoustic wave) even FBAR (film bulk acoustic resonator) band-40 [6-7].
filters are only able to produce at most out-of-band Another category of solutions aims to reduce antenna
rejection of 55~60 dB. The out-of-band emission of Wi-Fi mutual couplings (shorted as antenna decoupling) [8]-[26].
signal is still far beyond the LTE receiver sensitivity, which However, most of them are only proved to be effective for
will produce interference for the adjacent LTE receiver that mutual interfering antennas resonating at exactly the same
must be taken care of. band. Furthermore, it is well-known that almost all of the
TD-LTE Band 40 (B40) WiFi Channels (Ch) antennas in practical applications operate at multiple
Unit: MHz Unit: MHz
Ch13
frequency bands, for instance, the WBG (Wi-Fi-Bluetooth-
2461-2483
Ch12
2456-2478
GPS) antenna occupies the 1.575GHz, 2,4GHz bands; the
Ch11
2451-2473
Ch10
LTE and its diversity antennas must cover 690 MHz~960
2446-2468
EARFCN: 38750
2310
Ch9
2441-2463
MHz bands, 1.7 GHz~2.7 GHz and even 3.3 GHz ~3.7GHz
EARFCN: 38850 Ch8
2320
EARFCN: 38950 Ch7
2436-2458
bands for 5G systems. For most of the antenna decoupling
2330 2431-2453
EARFCN: 39050
2340
Ch6
2426-2448
solutions, only single band operation is assured, while the
EARFCN: 39150 Ch5
2350
EARFCN: 39250 Ch4
2421-2443
performance of the antennas at other operating frequency
2360 2416-2438
EARFCN: 39350
2370
Ch3
2411-2433
bands after antenna decoupling are untended, resulting in
EARFCN: 39450 Ch2
2380
EARFCN: 39550
2406-2428
Ch1
mismatch and efficiency loss.
2390 2401-2423
0 The proposed method in this paper for interference
fL f0 fH Frequency
suppression of contiguous frequency bands possesses the
Fig. 1. Mutually interfered Channels of the LTE band-40 and the 2.4GHz following unique features:
Wi-Fi band. (EARFCN: E-UTRA Absolute Radio Frequency Channel 1) The high-pass frequency selectivity makes the
Number) proposed AICC effective at around 2.3~2.5 GHz while it
Power Level
generates little interference on the operation bands below
Tx Signal Level
Pt
of Transceiver A
2.1 GHz.
Filter Rejection
(Typ. 55 ~ 60 dB)
Bandpass Filter 2) It uses the up-to-date Low Temperature Co-fired
Wi-Fi Response
Ceramic (LTCC) process to ensure a good product yield
and very compact size.
Po Antenna Isolation
(Typ. 15 dB)
3) It has external tuning mechanism to fine tune the
Pi High-pass extent of mutual coupling reduction as well as the antenna
AICC
Interference ANT B
ANT A Isolation Improved matching.
to Tranceiver B Rx Signal of by High-pass AICC
Transceiver B (Typ. 15 ~ 25 dB)
The remaining part of the paper is organized as the
Pr
LTE
following: Section II illustrates the design procedure using
Pn Noise floor
Band 40
SNR AICC with a practical example. Performance evaluation
Pa including active measurements of antennas with an AICC
Coexistence of Frequency
A and B Transceiver Transceiver in a commercially available Mi-Fi product is presented in
B A
Section III. Section IV gives the conclusions.
Fig. 2. The application of the proposed high-pass AICC for reducing
coexistence interference from transmitter A to receiver B working in II. DESIGN PROCEDURE ILLUSTRATED BY AN
contiguous frequency bands. EXAMPLE
Recently, both the industry and the academia strive to Unlike conventional antenna decoupling problems that
find a proper solution for the above-mentioned problem. two antennas are resonating at the same frequency band or
Their endeavors mainly fall into the following three even symmetrical in dimension, for antennas working in
categories: base band solution, active/passive filter solution contiguous frequency bands, the antennas could be distinct
and antenna decoupling solution. Coordinating between two from one another in shape and other characteristics. With
transceivers with adjacent operation frequency bands and the design theory well-illustrated in [11] and [17], it is easy
avoiding coexistence interference by time division to obtain the decoupling and matching conditions for the
multiplexing is a good choice to deal with the admittance parameters of the two antennas in contiguous
corresponding interferences [2]. An active interference bands:
suppression method, which uses a generated cancellation
signal with the help of complex system architecture and Decoupling Conditions:
algorithm, is proposed in [3]. Interference suppression
filters are used in [4] based on CMOS process. FBAR/SAW
Re Y21A ( f0 ) 0 , (1a)
based filtering solution proposed by Murata and Avago are
also available on the market, but they need to sacrifice
2169-3536 (c) 2018 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2018.2853709, IEEE Access
and
j Im Y21A ( f0 ) Y21C ( f0 ) 0 (1b)
Re Y11A ( f ) Y0 , f f0 , f H , (2a)
j Im Y11A ( f ) Y11C ( f ) 0, f f0 , f H , (2b) (a)
and
Re Y22A ( f ) Y0 , f f L , f0 , (3a)
j Im Y22A ( f ) Y22C ( f ) 0, f f L , f0 . (3b)
2169-3536 (c) 2018 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2018.2853709, IEEE Access
Port 1 Port 2
C2 C2
C C 50 ohm L1 50 ohm
Antenna1 Antenna2
t1 t2 C1
The Antennas
Port 1 Port 2
CM2 (a)
LM2 External External
C01 AICC Module C01 Coupling Coupling
AICC Module
TABLE I
DESIGNED VALUES OF THE COMPONENTS IN FIG. 4 (b)
Components Value Components Value
t1 25 degree C01 2.2 pF -Y12
t2 25 degree CM2 2.2 pF
L01 8.2 nH LM2 4.7 nH L01 Y11+Y12 Y22+Y12 L01
L02 12 nH / /
the mutual coupling between the two antennas at band-edge,
where filter rejection is not high enough yet.
Step1: Extract the mutual admittance of the two antennas
and determine the type of AICC that is needed. -Y12
As shown in Fig 3 (a), before using the vector network
analyzer (Keysight E5080A in this example) to measure the ’+Y12
Y11 ’+Y12
Y22
scattering parameters of the antennas, port extension must
be done using the two rigid cables that are to be soldered
near the feeding points of two antennas. This operation is to
(c)
shift the reference planes to the feeding position to exact Fig. 5. Equivalent network model transformation with the shunt inductor
the scattering parameters with the correct phase, which is L01.
important for successfully transforming the measured S-
according to matching conditions (2b) and (3b) since the
parameters to Y-parameters. The extracted admittance
original antennas are already well matched.
parameters are shown in Fig. 3 (c). To satisfy decoupling
For any realizable passive networks, the slope of self-
condition (1a), two sections of transmission lines with
admittance parameters with respect to frequency (Y11C(f)
electric length of t1 and t2 respectively are connected to the
and Y22C(f)) in this case must always be positive [27]. To
two antennas to transform the mutual admittance of the two
balance the matching performance within the entire band of
antennas, Y21A, to purely imaginary [11], as shown in Fig. 4
interest, it is favorable to place the zero of Y11C(f) and Y22C(f)
(a). After admittance transformation, the remaining
in the center frequency of each desired operation band. To
imaginary part of the mutual admittance of the two
introduce a reflection zero, an L-C series resonant circuit
antennas, Im(Y21A), is positive. From equation (1b), it is
that is shorted to ground is introduced and shown in Fig. 5
found that an AICC with negative mutual admittance is
(a), with two capacitors are connected to the two ports of
demanded.
the network. The designed circuit model of the AICC, as
Step2: Design the circuit model of the AICC. indicated in Fig. 5 (a), has the following component values:
To create a two-port passive network/circuit with a C2 = 0.47 pF, L1 = 19 nH and C1 = 0.48 pF.
negative mutual admittance, capacitive Step3: Translate the circuit model of the AICC to physical
components/capacitors should be placed in-between the two realization using LTCC technology.
ports by intuition. Furthermore, since the admittance
The designed circuit can now be realized using LTCC
parameters for a lossless network are all purely imaginary,
process, which is given in Fig. 6 (a). The substrate used in
the self-admittances of the AICC, Y11C(f) and Y22C(f) must
this example has the relative permittivity of 9.8 and the loss
approach to zero within the frequency band of interest
2169-3536 (c) 2018 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2018.2853709, IEEE Access
Substrate C2 Smith
Ground Chart
C2 Open
Port 2
Circuit
S11
Band of interest:
Port 1
L1 1.5-2.1 GHz
C1
(a)
4 Layers Frequency:1 GHz to 4 GHz
Cond7 (a)
Cond6
Via5 5 Layers
Cond5 2 Layers
Via4 5 Layers
Cond4
Via3 5 Layers
Cond3
Via2 5 Layers
Cond2
Via1 5 Layers
Cond1
GND z
1 Layer 4 Layers
x y
(b)
Ground
(b)
Fig. 7 (a) Smith Chart plot of the AICC and (b) S-parameters of the
connected antennas and the AICC from 1.4 GHz to 2.8 GHz.
Port2
Port1 slight misalignments between the layers.
The inductor L1 is realized in a meandering fashion to be
fitted in a compact volume. The line width is designed as
narrow as possible to enlarge the inductance, which is
finally determined to be 80 μm according to our fabrication
Marker
Ground process tolerance. Two ground electrodes and two
input/output electrodes are plated on the outside of the
LTCC block, which is depicted in Fig. 6 (a). The two
(c) ground electrodes together with the ground layer could also
Fig. 6. (a) Physical model of the proposed high-pass AICC. (b) The layer
stack of the proposed high-pass AICC. (c) Photo of the AICC chip serve as shielding for the internal components. The
soldered on a PCB. finalized AICC, which is then soldered on the PCB board,
is shown in Fig. 6 (c).
tangent of only 0.003. The thickness of each finalized
substrate layer is 21μm, and the conductor thickness is 10 μ Step4: Adjust the external couplings and connect the
m. There are totally 8 layers of silver conductors including fabricated AICC module with the antennas in shunt to
the ground in this design. There are five via holes in the obtain the desired performance.
LTCC module indicated as Via1~ Via5 in Fig. 6 (b). The admittance parameters of the finalized AICC (Fig.
Thanks to the advanced fabrication process, we can make 6a), Y11C and Y21C are already shown in Fig. 3 (c) and (d),
all via holes padless (no via-pad needed). The diameters of respectively. Because the network is still symmetric at this
the via holes are all chosen to be 100 μm. All capacitors moment, Y22C is the same as Y11C. At 2.4 GHz, the mutual
(C1, C2) in this design are implemented using parallel admittance between the two antennas, Y21A is about 0.0064
plates. It is worth mentioning that the upper plates of S, while the mutual admittance parameter of the designed
capacitors C2 are intentionally made larger than their lower AICC, Y21C at the very same frequency is -0.0085 S. To
plates so that the capacitance can be stable while there are perfectly eliminate the mutual coupling at this frequency,
2169-3536 (c) 2018 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2018.2853709, IEEE Access
Y21C needs to be adjusted. Two external capacitors C01 are frequency f should have positive slope. It is obvious that
connected to the AICC to slightly modify the mutual the zeros of Y11’(f) and Y22’(f) denoted by f01’ and f02’
admittance at 2.4 GHz to be -0.0063 S (Y21D), which shall have the following relation if L01≠L02
cancels out Y21A at 2.4 GHz. The sum of Y21A and Y21D is
f01' f02' , (7)
plotted in Fig. 3 (c) as Y21T, which approach to zero around
2.4 GHz as we desired.
The matching conditions in (2) and (3) are more and
complicated to be satisfied. As is already pointed out in Step2,
to leverage the matching performance within the entire band
f01' , f02' f0 , (8)
of interest, it is demanded to set Y11C(f) and Y22C(f) to be zero
in the center frequency of the desired operation band. In
another word, the zero of the Y11 of the AICC should be It can also be concluded from (6) that if , we
around 2.45 GHz while the zero of the Y22 of the AICC will have:
should be around 2.35 GHz. In this case, the zeros of Y11D,
Y22D are all located at around 2.26 GHz, and then, effective f01' f02' f0 , (9)
measures must be taken to move their zeros to 2.35GHz and
2.45GHz, respectively. Two shunt inductors, L01 and L02, are
connected to the two I/O ports of the AICC which is In this example, we intentionally place the zero of the
symmetric AICC at frequency point of 2.26GHz that is
illustrated in Fig. 5 (b). The mechanism to move zeros of Yii
lower than 2.35GHz. Then, L01 is designed to be 8.2 nH and
(i=1,2) using shunt inductors is described in Fig. 5 (c). Using
L02 is designed to be 12 nH. By doing this, the admittance
Pi-equivalent circuit, any two-port microwave network can
parameter of the AICC with two shunt inductors become
be expressed by two shunt components: (Y11+Y12), (Y22+Y12), Y11M and Y22M which are shown in Fig. 3 (d). At this time,
and one series component: (–Y12). The two shunt we are able to move the zeros of Y11M and Y22M to 2.45
components will be combined with two extra shunt inductors GHz and 2.35 GHz, respectively.
L01 and L02, and they will be converted to (Y11’+Y12) and Finally, the AICC together with two external capacitors
(Y22’+Y12), where: C01 and two shunt inductors L01 & L02 are parallel
1 connected to the coupled antennas as shown in Fig. 4. The
Y11' ( f ) Y11 ( f ) j . (4a) precise values are listed in Table I. The responses of the
2 f L01
entire connected network in high frequency ranging from
and 2.2 GHz to 2.65 GHz are already shown in Fig. 3 (b) (with
1 solid markers). The isolation between the two antennas is
Y22' ( f ) Y22 ( f ) j . (4b) improved from around 15 dB to more than 30 dB near the
2 f L02
band edge of 2.4 GHz, showing more than 15 dB
Since all admittance parameters of the AICC improvement. The matching performance of Port 1 is better
are purely imaginary, it is easy to re-write than 15 dB within the Wi-Fi 2.4 GHz band, while the
equation (4) in the following form: matching performance of Port 2 is better than 10 dB within
the LTE band-40. It should be mentioned that a L-type
1
Im Y11' ( f ) Im Y11 ( f ) (5a) matching circuit (series CM2 and shunt LM2) is used at Port 2
2 f L01
.
to further enhance its performance.
and Step5: Check the open-circuit condition in low-band after
using the AICC.
1
Im Y22' ( f ) Im Y22 ( f )
(5b)
2 f L02
.
To ensure that the decoupling of the two antennas near
2.4 GHz does not affect the performance in the GPS band
For a symmetrical network, supposing the zeros of the WBG antenna (connected to Port 1), one of the AICC
of Y11(f) and Y22(f) locate at f0, we shall have: port that is connected to WBG antenna shall be open-
1 circuited at least in the GPS band whose center frequency is
Im Y11' ( f0 ) 1.575GHz. This is achieved by design the AICC model
2 f L01
. (6a)
with very small I/O capacitors, C2 together with the high-
pass nature of the circuit model shown in Fig. 5 (a). In the
physical model shown in Fig. 6, this small capacitor is
and
realized by increasing the distance between the upper and
1
Im Y22 ( f0 )
' lower plate of the capacitors. In this LTCC design, we
2 f L02
. (6b)
increase this distance to two substrate layers to reduce the
I/O capacitance.
Considering the positive real nature of any realizable
networks [27], both Im[Y11’(f)] and Im[Y22’(f)] against
2169-3536 (c) 2018 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2018.2853709, IEEE Access
(a)
With AICC 20.9 20.2 20.1 91.58 90.35
LTE
B40 Wi-Fi A. Measurement Setup
Mobile
Phone
Mi-Fi As shown in Fig. 8. The active measurement of the Mi-Fi
device with coupled LTE Band-40 and Wi-Fi antennas is
conducted in a SATIMO SG-24 near field anechoic
Link chamber [28]. To eliminate interference from other wireless
Antenna
Down Link
transmission, other antennas are removed, keeping only
LTE Band-40 and Wi-Fi antennas. The DUT (Mi-Fi) is
placed on the azimuth positioner and put in the center of the
probe array. A mobile phone is also placed in the chamber
Up Link near the DUT.
Communication Tester
The TIS (total isotropic sensitivity) measurement of the
(RS. CMW500) DUT is conducted using conventional method but with the
(b) DUT’s Wi-Fi in hotspot mode. That is to say, as the
Fig. 8 (a) The photo; (b) The block diagram of the active measurement
configuration in a near field 24-probe anechoic chamber
measurement is ongoing, the DUT’s Wi-Fi module is
always communicating with the mobile phone while the
LTE Band-40 transceiver of the DUT is only in its
The smith chart of the designed AICC is plotted in Fig. 7 receiving mode as shown in Fig. 8 (b). To ensure that our
(a). It is obvious that we are able to control the open circuit decoupling design do not affect the efficiency of the Band-
frequency of the AICC to work around 1.575 GHz. The 40 antennas, TRP (total radiated power) measurement is
overall S-parameters of the coupled antennas only as well also conducted.
as the connected coupled antennas and AICC from 1.4 GHz
to 2.8 GHz are also superposed in Fig. 7 (b). At the GPS B. Results and Discussion
band, the reflection and isolation curves are almost The measured TRP values of the following three channels
unchanged after adding the AICC. of LTE Band-40 are recorded for the DUT with and without
the AICC we designed: channel 38750 (2.30262 GHz),
III. PERFORMANCE EVALUATION BY ACTIVE channel 39150 (2.35 GHz) and channel 39550 (2.39738
MEASUREMENT GHz). TIS measurement is only carried out in channel
To verify that our decoupling measures are effective in 39150 (2.35 GHz) and channel 39550 (2.39 GHz).
real-world applications, active measurements are carried The measurement is conducted twice, and their
out in an anechoic chamber. respective results are listed in Table II, showing good
2169-3536 (c) 2018 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2018.2853709, IEEE Access
repeatability. We can draw the following conclusions from [11] L. Zhao, L. K. Yeung, and K.-L. Wu, " A coupled resonator
decoupling network for two-element compact antenna arrays in
the tests:
mobile terminals," IEEE Trans. Antennas Propag., vol. 62, no. 5, pp.
1) The strong mutual coupling between the LTE Band- 2767-2776, May 2014.
40 and Wi-Fi antennas significantly deteriorates the [12] M. Mavridou, A.-P. Feresidis, and P. Gardner, “Tunable Double-
TIS in the high band of LTE Band-40. The Layer EBG Structures and Application to Antenna Isolation,” IEEE
Trans. Antennas Propag., vol. 64, no. 1, pp. 70-79, JAN. 2016.
measured TIS of the DUT without AICC in channel
[13] J. Y. Lee, S. H. Kim and J. H. Jang, "Reduction of Mutual Coupling
39550 located at 2.39 GHz, is around 76~78 dBm, in Planar Multiple Antenna by Using 1-D EBG and SRR Structures,"
which is about 12~14 dB lower compared to the in IEEE Transactions on Antennas and Propag., vol. 63, no. 9, pp.
TISs in channel 38750 and channel 39150. 4194-4198, Sept. 2015.
[14] M. J. Al-Hasan, T. A. Denidni, and A.-R. Sebak, “Millimeter-Wave
2) In channel 38750 and channel 39150, the Wi-Fi
Compact EBG Structure for Mutual Coupling Reduction
filter rejection characteristic successfully helps to Applications.” IEEE Trans. Antennas Propag., vol. 63, no. 2, pp.
eliminate unwanted interference from Wi-Fi. 823-828, FEB. 2015.
3) Using the AICC will not affect the TRP of LTE [15] S.-W. Su, C.-T. Lee, and F.-S. Chang, “Printed MIMO-antenna
system using neutralization-line technique for wireless USB-dongle
Band- 40, as the TRP values in all three measured
applications,” IEEE Trans. Antennas Propag., vol. 60, no. 2, pp.
channels of the DUT with and without the AICC 456–463, Feb. 2012.
differ from each other by no more than 1 dB, which [16] B. K. Lau and J. B. Andersen, “Simple and efficient decoupling of
is within measurement uncertainty. compact arrays with parasitic scatterers,” IEEE Trans. Antennas
Propag., vol. 60, no. 2, pp. 464-472, Feb. 2012.
4) A good decoupling design using the AICC, not only
[17] Z. Wang, L. Zhao. Y. Cai, S. Zheng and Y. Yin, “A Meta-Surface
improves the passive isolation between the two Antenna Array Decoupling (MAAD) Method for Mutual Coupling
coupled antennas, but also recovers the TIS of the Reduction in a MIMO Antenna System,” Scientific Reports, 2018,
channel 39550 of the LTE Band-40 transceiver to vol. 8, Article 3152.
[18] L. Zhao and K.-L. Wu, “A Dual-band Coupled Resonator
around 90 dBm.
Decoupling Network for Two Coupled Antennas,” IEEE Trans.
Antennas Propag., vol. 63, no. 7, pp. 2843-2850, Apr. 2015.
ACKNOWLEDGEMENT [19] K. Wei, J. Y. Li, L. Wang, Z. J. Xing and R. Xu, "Mutual Coupling
The authors would like to thank Mr. Fujimoto at Hirai Reduction by Novel Fractal Defected Ground Structure Bandgap
Filter," in IEEE Transactions on Antennas and Propagation, vol. 64,
corporation for the help on LTCC fabrication. We would also no. 10, pp. 4328-4335, Oct. 2016.
express our gratitude to the testing engineers: Chao Jing, [20] S. M. Amjadi and K. Sarabandi, "Mutual Coupling Mitigation in
Xudong Zhang at Innowave Wireless Ltd. for the active Broadband Multiple-Antenna Communication Systems Using
measurement. We are also grateful to Dr. Kewei Qian and Feedforward Technique," IEEE Transactions on Antennas and
Propagation, vol. 64, no. 5, pp. 1642-1652, May 2016.
Mr. Li Liu from Filter Group for preparation of the samples [21] K. Wang, L. Li and T. F. Eibert, "Comparison of Compact Monopole
and helpful discussions. Antenna Arrays With Eigenmode Excitation and Multiport
Conjugate Matching," IEEE Transactions on Antennas and
Propagation, vol. 61, no. 8, pp. 4054-4062, Aug. 2013.
REFERENCES
[22] J. C. Coetzee and Y. Yu, “Port decoupling for small arrays by means
[1] 3GPP TR 36.816: “Study on signalling and procedure for
of an eigenmode feed network”, IEEE Transactions on Antennas and
interference avoidance for in-device coexistence,” V11.2.0, 2011.
Propagation, vol. 56, no. 6, pp.1587-1593, Jun 2008.
[2] Z. Hu, R. Susitaival, Z. Chen, I.-K. Fu, P. Dayal, and S. K. Baghel,
[23] Y. Wang and Z. Du, "A Wideband Printed Dual-Antenna With Three
“Interference avoidance for in-device coexistence in 3GPP LTE-
Neutralization Lines for Mobile Terminals," IEEE Transactions on
advanced: Challenges and solutions,” IEEE Commun. Mag., vol. 50,
Antennas and Propagation, vol. 62, no. 3, pp. 1495-1500, March
no. 11, pp.60-67, Nov. 2012.
2014.
[3] A. Raghavan, E. Gebara, E. Tentzeris, and J. Laskar, “Analysis and
[24] G. H. Zhai, Z. N. Chen, X. M. Qing, “Enhanced Isolation of a
design of an interference canceller for collocated radios,” IEEE
Closely Spaced Four-Element MIMO Antenna System Using
Trans. Microwave Theory Tech., vol. 53, no. 11, pp. 3498–3508, Nov.
Metamaterial Mushroom”, IEEE Trans. Antennas Propag., vol. 63,
2005.
no. 8, pp. 3362-3370, Aug. 2015.
[4] H. Khatri, P. S. Gudem, and L. E. Larson, “Integrated RF
[25] H.-L. Peng, R. Tao, W.-Y. Yin, and J.-F. Mao, “A Novel Compact
interference suppression filter design using bond-wire inductors,”
Dual-Band Antenna Array With High Isolations Realized Using the
IEEE Trans. Microw. Theory Techn., vol. 56, no. 5, pp. 1024–1034,
Neutralization Technique,” IEEE Trans. Antennas Propag., vol. 61,
May. 2008.
no. 4, pp. 1956-1962, Apr. 2013.
[5] Murata Manufacturing Co., Ltd, “SAW Single Filter for Band40”,
SAFEB2G35FB0F0A datasheet, Feb., 2014. [26] S. Ghosh, T. N. Tran and T. Le-Ngoc, "Dual-Layer EBG-Based
[6] Murata Manufacturing Co., Ltd, “SAW Single Filter for WLAN Miniaturized Multi-Element Antenna for MIMO Systems," in
2.4GHz”, SAFEA2G45MA1F0A datasheet, June, 2014. IEEE Transactions on Antennas and Propagation, vol. 62, no. 8,
[7] Avago Technologies, “ISM Bandpass Filter”, ACFF-1024 datasheet, pp. 3985-3997, Aug. 2014.
Oct., 2014. [27] M. E. van Valkenburg, Network Analysis, Prentice-Hall, Englewood
[8] F. Yang and Y. R. Samii, “Microstrip antennas integrated with Cliffs, NJ, 1955.
electromagnetic band-gap EBG structures: A low mutual coupling [28] [Online]. Available: http://www.satimo.com/
design for array applications,” IEEE Trans. Antennas Propag. vol. 51,
no. 10, pp. 2936-2946, Oct. 2003.
[9] L. K. Yeung and Y. E. Wang, “Mode-based beamforming arrays for
miniaturized platforms,” IEEE Trans. Microw. Theory Tech., vol. 57,
no. 1, pp. 45-52, Jan. 2009.
[10] C. Y. Chiu, C. H. Cheng, R. D. Murch, and C. R. Rowell, “Reduction
of mutual coupling between closely-packed antenna element,” IEEE
Trans. Antennas Propag. vol. 55, no. 6, pp. 1732-1738, Jun. 2007.
2169-3536 (c) 2018 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See
View publication stats http://www.ieee.org/publications_standards/publications/rights/index.html for more information.