DSpace Collection:
http://dspace.cityu.edu.hk:80/handle/2031/746
Thu, 08 Dec 2016 09:47:30 GMT2016-12-08T09:47:30ZStable photonic microwave generation using semiconductor laser dynamics
http://dspace.cityu.edu.hk:80/handle/2031/8624
Title: Stable photonic microwave generation using semiconductor laser dynamics
Authors: Zhuang, Junping (莊俊平)
Abstract: Photonic generation of stable microwave signals has attracted considerable attention for
transmission of microwave signals over optical fibers with low loss, electromagnetic
interference immunity, and large bandwidth. The generation techniques have found
a range of potential applications including radio-over-fiber communication, photonic
microwave signal processing, photonic microwave beamforming, and radar. Among the
various techniques for photonic microwave generation, the approaches based on semiconductor
lasers are strong competitors due to compactness and potential for integration.
In this thesis, the dynamics of both optically injected semiconductor lasers and passively
mode-locked semiconductor lasers are investigated for photonic generation of microwave
signals. Techniques for improving the frequency stability are explored. Firstly, an
optical injection drives the laser into nonlinear dynamical period-one (P1) oscillation
at a microwave frequency, which is widely tunable up to 100 GHz. Owing to the laser
nonlinearities, the P1 microwave linewidth can even be smaller than the free-running
optical linewidth, though it is still affected by intrinsic laser noise. Optical feedback to
the injected laser are introduced to further stabilize the P1 frequency fluctuation. Via
theoretical analyses and numerical simulations, the P1 microwave linewidth is found to
follow an inverse-square dependence on the feedback strength and feedback delay time
asymptotically. By modification to a dual-loop feedback, noisy side peaks around the
central P1 frequency are effectively suppressed through the Vernier effect. The numerical
simulations are in good agreement with the experimental results. Photonic generation at
a millimeter-wave frequency of 45.4 GHz is experimentally demonstrated with a clean
spectrum and a linewidth below 50 kHz. Secondly, utilizing the optical tunability of the
P1 frequency by introducing modulation on the injection power, photonic generation of
frequency-modulated continuous-wave (FMCW) microwave signals is also investigated.
This generation scheme exhibits superior stability and tunability in terms of central
frequency, bandwidth, and chirp rate. The FMCW microwave signal is generated with
central frequency, sweep bandwidth, and chirp rate tunable up to 22 GHz, 7.7 GHz,
and 0.42 GHz/ns, respectively. Thirdly, monolithic photonic microwave generation
is demonstrated by using a passively mode-locked semiconductor laser, where an
anticolliding-pulse mode-locking (ACPML) configuration is introduced. The laser
consists of a gain section and a saturable absorber (SA) section. With proper DC biasing
on the laser, a millimeter-wave signal at around 34.7 GHz with a linewidth on the
order of 100 kHz is demonstrated. To realize the ACPML configuration, low-reflection
and high-reflection coatings are deposited on the SA section and gain section facets,
respectively. After the deposition, the threshold is unchanged, while the modulation of
the SA is increased. As a result, simultaneous improvements in pulse peak power, pulse
amplitude noise, and timing jitter are achieved. These photonic schemes generate stable
microwave signals that offer opportunities in photonic microwave applications.
ii
Notes: CityU Call Number: TK7872.S5 Z45 2015; xxvii, 167 pages : illustrations (some color) 30 cm; Thesis (Ph.D.)--City University of Hong Kong, 2015.; Includes bibliographical references (pages 141-164)Thu, 01 Jan 2015 00:00:00 GMThttp://dspace.cityu.edu.hk:80/handle/2031/86242015-01-01T00:00:00ZMemory-based hardware architectures for regular expression matching
http://dspace.cityu.edu.hk:80/handle/2031/8622
Title: Memory-based hardware architectures for regular expression matching
Authors: Or, Nga Lam (柯雅琳)
Abstract: Signature-based intrusion detection systems (IDS) and anti-virus systems detect
potential threats via byte-by-byte content inspection. Today, more and more intrusion
signatures and computer viruses are specified using regular expression (regex).
Matching an input data stream against a set of regex patterns is a computation
intensive task. Accompanying the rapidly increasing data traffic over the Internet, the
design of hardware accelerator to speed up regex matching has been an active
research area.
A systematic approach to detect regex is based on finite automaton. The
space-time trade-off between deterministic finite automaton (DFA) and
non-deterministic finite automaton (NFA) is well-known. DFA can offer constant
throughput but it may suffer from the state explosion problem. Therefore,
implementation of DFA-base match engine for large pattern set on embedded device
with limited on-chip memory is not viable. NFA-based match engine requires linear
space but the throughput is unacceptably low if it is implemented using the
conventional lookup-table based method. Implementing the NFA-based match engine
with hardwired circuits may overcome the speed deficiency by exploiting the massive
parallelism offered by dedicated hardware circuitries, but this approach does not
support efficient dynamic updates to the pattern set.
In this thesis I shall present cost-effective memory-based hardware architectures
for regex matching. Results presented in this thesis are based on the regex patterns
taken from the Snort IDS and the ClamAV anti-virus system. The size of the pattern
set in a typical IDS installation is in the order of hundred, whereas the size of the
pattern set in an anti-virus system is in the order of 10 thousand. Hence, scalability of
the matching method is a major issue for anti-virus application. One of the objectives
in this research is to implement a virus detection engine on a single FPGA. This task
is unprecedented and is extremely challenging given the fact that the size of the virus
database is greater than the amount of on-chip memory available in state-of-the-art
FPGA.
The major contributions of this thesis are two folds. First, a memory-based NFA
regex matching architecture, called MX-NFA, for intrusion detection is developed.
MX-NFA is based on a novel concept of active transition rule table that can be
implemented using on-chip block RAM with simple embedded control circuits. Each
transition rule only contains a transition symbol and the associated control flags, and
no explicit state IDs are maintained. Individual transition rules can be enabled or
disabled by the associated control circuits based on the input sequence. The system
does not explicitly maintain the set of active states. Instead, all the out-going
transition rules from an active state of the underlining NFA are enabled, and out-going
transition rules from an inactive state are disabled. When the input character matches
the transition symbol of an enabled transition rule, the corresponding rule will be fired
and it may in turn activate the transition rules associated with the corresponding next
state. The MX-NFA regex matching engine requires linear space and offers constant
processing speed of one character per cycle. In addition, dynamic updates to the
pattern set can be handled easily by simply modifying the data stored in the internal
tables and registers without any hardware reconfiguration.
The number of regex patterns in a typical virus database is in the order of 10
thousand. It is not possible to implement the regex matching engine for such a large
pattern set on a single FPGA using conventional byte-oriented matching methods. The
second contribution of my thesis is the proposed information reduction approach to
regex matching for very large pattern set in anti-virus application. Instead of
following the conventional byte-oriented matching approach, the virus signatures are
treated as equivalent token-based patterns. A token is a non-trivial sub-pattern (which
may contain regex features) extracted from the underlying patterns. The tokens can be
seen as a hypothetical alphabet set with very large cardinality, e.g. the number of
unique tokens is greater than the number of regexes in the pattern set. The input
byte-stream is transformed into a token-stream using cost-effective hardware modules,
where the number of tokens in the token-stream is much smaller than the number of
bytes in the original input stream. A byte value is usually contained in a couple of
thousand patterns, but in general a given token is contained in one or a few patterns
only. When a token is received, the NFA-based detection system will only need to
make a few state transitions. As a result, the complexity of the underlying automaton
can be reduced substantially, and cost-effective implementation in hardware is viable.
The proposed method to detect the 9700 regexes in the ClamAV virus database only
requires 2.24MB of on-chip memory. Together with the previous results of our
research group on the design of string matching engine, it is possible to implement a
hardware detection engine for the full set of ClamAV embedded virus signatures with
90 thousand strings and 9700 regexes on a single FPGA.
Notes: CityU Call Number: TK5105.59 .O7 2015; xi, 126 pages : illustrations (some color) 30 cm; Thesis (Ph.D.)--City University of Hong Kong, 2015.; Includes bibliographical references (pages 120-125)Thu, 01 Jan 2015 00:00:00 GMThttp://dspace.cityu.edu.hk:80/handle/2031/86222015-01-01T00:00:00ZIterative channel estimation methods in large antenna systems
http://dspace.cityu.edu.hk:80/handle/2031/8621
Title: Iterative channel estimation methods in large antenna systems
Authors: Ma, Junjie (馬俊傑)
Abstract: Large antenna system, also known as massive multiple-input multiple-output
(MIMO) system, has been identified as a promising technique for future 5G cellular
systems. Through the use of a large number of antennas, huge gain can
be achieved through coherent reception (matched filtering) or coherent transmission
(beamforming). To achieve such gain, channel state information (CSI)
is the key. This fact motivates the research activities on the channel estimation
problem in large antenna systems. Iterative (turbo) detection/estimation principles
have been successfully applied to many coding, communication and signal
processing problems. In this thesis, we will study iterative channel estimation
methods in the large antenna context.
After introducing the background work in Chapter 1, we consider in Chapter
2 the channel estimation problem for an uplink multi-cell system where each
cell contains only one user. We analyze the performance of a data-aided channel
estimation technique where partially decoded data is used to estimate the
channel. We show that there are two types of interference components in this
scheme that do not vanish even when the number of antennas grows to infinity.
The first type, referred to as cross-contamination, is due to the correlation
among the data signals from different users. The second type, referred to as
self-contamination, is due to the dependency between the channel estimate and
the estimation error. For efficient use of the channel, the data part in a signaling
frame is typically much longer than the pilot part for a practical system. Consequently,
compared with pilot signals, data signals naturally have lower cross
correlation. This fact reduces the cross-contamination effect in the data-aided
scheme. Furthermore, self-contamination can be effectively suppressed by iterative
processing. These results are confirmed by both analyses and simulations.
Compared to other channel estimation schemes for large antenna systems, the
data-aided channel estimation scheme is non-cooperative and does not rely on
specific channel conditions. Therefore it is simpler and more flexible in practice.
In Chapter 3, we extend the analysis of the data-aided channel estimation
scheme to the multiple-user scenario. We consider joint linear minimum meansquared
error (LMMSE) channel estimation and matched filtering (MF)-based
data detection with soft-interference cancellation.
In Chapter 4, we study a scheme where pilots are superimposed on data.
This scheme avoids the rate loss incurred by dedicate pilot positions. We investigate
the power allocation problem between pilot and data for the superimposed
scheme. Intuitively, there is an optimal pilot power ratio that achieves
the best tradeoff between pilot power overhead and channel estimation quality.
Interestingly, through the use of a recently developed relationship between
minimum mean-squared error (MMSE) and mutual information (referred to as
the MMSE-I relationship), and based on several assumptions, we show that the
optimal portion of the power allocated to pilot approaches zero for Gaussian
signaling. We also consider the practical realization of the theoretical prediction
and show that considerable performance improvement can be achieved by
matching the transfer functions of the forward error correction (FEC) code and
the channel estimator/data detector.
In Chapter 5, we propose a compressed sensing channel estimation algorithm
for millimeter-wave large antenna systems. Millimeter-wave communication
channel is sparse due to the directional transmission nature at high frequencies.
Compressed sensing techniques can be leveraged to exploit the spatial
sparsity. We first propose a generic turbo signal recovery (TSR) algorithm for
partial unitary sensing matrix-based compressed sensing problems. We analyze
the proposed TSR algorithm using state evolution. We show that the stationary
point of the state evolution of the proposed algorithm is consistent with that
of the Bayesian optimal MMSE performance derived using the replica method.
This indicates the potentially excellent performance of the proposed algorithm,
as verified by numerical results. For a millimeter-wave system with a large
uniform linear array, the array manifold can be approximated by a discrete
Fourier transform (DFT) matrix. Motivated by this fact, we apply the proposed
turbo signal recovery algorithm to the millimeter-wave sparse channel estimation
problem. We follow the hybrid analog-digital processing architecture where
the entries of the analog processing matrix are constrained to have constant
magnitudes. A technique to design the analog processing matrix is proposed.
We provide numerical results to demonstrate the effectiveness of the proposed
algorithm.
Finally, Chapter 6 concludes the thesis and outlines possible future research
directions.
In summary, in this thesis we have studied the applications of iterative channel
estimation methods in large antenna systems. Our results show that by
appropriately exploiting available information, such as soft feedback of a decoder
and a priori sparsity information, we can reduce the training overhead
and improve the channel estimation accuracy. The contributions contained in
this thesis show that iterative channel estimation methods provide a powerful
means to enhance the performance of massive MIMO systems
Notes: CityU Call Number: TK5103.4836 .M3 2015; xviii, 140 pages : illustrations 30 cm; Thesis (Ph.D.)--City University of Hong Kong, 2015.; Includes bibliographical references (pages 131-140)Thu, 01 Jan 2015 00:00:00 GMThttp://dspace.cityu.edu.hk:80/handle/2031/86212015-01-01T00:00:00ZDesign of millimeter-wave planar antenna arrays with substrate integrated waveguide feeds
http://dspace.cityu.edu.hk:80/handle/2031/8620
Title: Design of millimeter-wave planar antenna arrays with substrate integrated waveguide feeds
Authors: Li, Yujian (李雨鍵)
Abstract: The thesis presents a series of planar antenna arrays for emerging millimeter-wave
wireless applications. All proposed antennas, which are excited by coupling apertures
etched on the copper-clad surface of the substrate integrated waveguide (SIW), exhibit
wide bandwidths and good radiation characteristics. Due to the use of the SIW to
compose feed networks, low feed network losses and high-gain performance are
obtained forthe designed antenna arrays. Moreover, the feeding scheme of the aperture
coupling with a two dimensional geometry is easy to realize at millimeter-wave
frequencies, which enables the proposed antenna array to be fabricated conveniently
by different manufacturing technologies.
First, cavity-backed microstrippatch antenna arrays with full corporate SIW feed
networks are demonstrated. A prototype of 16 × 16 radiating elements with a SIW to
waveguide transition is fabricated by applying standard printed circuit board (PCB)
facilities. A gain up to 30.1 dBi with a 3-dB gain bandwidth of 16.1%, an impedance
bandwidth of 15.3% for SWR < 2, and symmetrically broadside radiation patterns with
-40 dB cross-polarizations are achieved. The performance of the proposed antenna
array is also systematically evaluated. The result serves as a reference for designing
large antenna arrays operating at millimeter-wave frequencies.
Second, a microstrip patch antenna array with an improved wide bandwidth is studied.
A novel wideband SIW-fed dual-aperture-coupled microstrip patch antenna is
proposed. A single layered SIW feeding network with wideband T-junctions and the
wideband radiating element are employed to achieve high-gain and wideband
performance simultaneously. The simulated and measured impedance bandwidths of a
4×4 antenna array are 27.5% and 22.6% for |S11| < -10 dB. A gain up to 19.6 dBi and
symmetrical unidirectional radiation patterns with low cross polarization are also
achieved.
Third, anovel perforated dense dielectric (DD) patch antenna array fed by a
single-layered SIW feed network is investigated. The proposed antenna array offers
advantages of wide bandwidth and good radiation performance and can be fabricated
conveniently by standard PCB technology. A 4 × 4 prototype operated at the 60-GHz
band is designed, fabricated and measured to demonstrate the design. An impedance
bandwidth of 23% for SWR < 2, a gain up to 17.5 dBi with a 3-dB gain bandwidth of
24%, and symmetrical unidirectional radiation patterns with cross polarizations of less
than -20 dB are obtained.
Fourth, a series of novel SIW-fed aperture-coupled ME-dipole antennas with linearly,
circularly and dual-polarized radiations are developed. Compared with reported
designs, the proposed designsmaintain the advantages of the ME-dipole antennas,
including wide bandwidth, symmetrical E- and H- plane radiation patterns, low
backward radiation and high gain, but are easier to realize at millimeter-wave
frequencies. By employing the dual-polarized aperture-coupled ME-dipole antenna, a
2 × 2 wideband antenna array working at the 60-GHz band that can generate
two-dimensional multi-beams with dual-polarizations is designed, fabricated and
tested. A measured impedance bandwidth wider than 21.9% for |S11| < -10 dB and a
gain up to 12.5 dBi are obtained. Owing to the superiority of the ME dipole, the
radiation pattern of the array is also stable and nearly identical in two orthogonal
planes for both polarizations.
Notes: CityU Call Number: TK7871.67.M53 L5 2015; xxvii, 206 pages : illustrations 30 cm; Thesis (Ph.D.)--City University of Hong Kong, 2015.; Includes bibliographical references (pages 185-201)Thu, 01 Jan 2015 00:00:00 GMThttp://dspace.cityu.edu.hk:80/handle/2031/86202015-01-01T00:00:00Z