GNSS RO Receiver Tracking and Ionospheric Irregularities Localisation Algorithms

ROSES 2014/A.26 GNSS Remote Sensing Science Team
NRA NNH14ZDA001N-GNSS Multi-GNSS Radio Occultation Algorithms
Table of Contents
1 Scientific / Technical / Management 1-1
1.1 Introduction and Background ........................................................................... 1-1
1.1.1 Challenge 1: Multi-GNSS RO Receiver Processing Algorithms ......... 1-1
1.1.2 Challenge 2: Ionosphere Irregularities Localization ............................. 1-1
1.1.3 Challenge 3: Polar Ionospheric Irregularities Characterization ............ 1-3
1.2 Objective and Expected Significance ............................................................... 1-3
1.3 Technical Approach and Methodology ............................................................. 1-5
1.3.1 Task 1 .................................................................................................... 1-5
1.3.2 Task 2 .................................................................................................... 1-9
1.3.3 Task 3 .................................................................................................. 1-12
1.4 Perceived Impact to State of Knowledge ........................................................ 1-13
1.5 NASA Programmatic Relevance .................................................................... 1-13
1.6 Plan of Work ................................................................................................... 1-14
1.6.1 Key Milestones ................................................................................... 1-14
1.6.2 Management Structure ........................................................................ 1-15
1.6.3 Contributions of PI and Key Personnel .............................................. 1-15
2 References and Citations 2-1
3 Biographical Sketches 3-1
3.1 Principal Investigator ........................................................................................ 3-1
3.2 Institutional Principal Investigator .................................................................... 3-3
3.3 Co-Investigators ................................................................................................ 3-5
3.4 Collaborator ...................................................................................................... 3-7
4 Current and Pending Support 4-1
4.1 Current Awards ................................................................................................. 4-1
4.2 Pending Awards ................................................................................................ 4-3
5 Budget Justification 5-1
5.1 Budget Narrative ............................................................................................... 5-1
5.2 Budget Details - CSU ........................................................................................ 5-3
5.3 Budget Details – JPL ........................................................................................ 5-4
5.4 Table of Personnel and Work Effort ................................................................. 5-4
5.5 Facilities and Equipment ................................................................................... 5-4
ROSES 2014/A.26 GNSS Remote Sensing Science Team
NRA NNH14ZDA001N-GNSS Multi-GNSS Radio Occultation Algorithms
1-­‐
1
1 Scientific / Technical / Management
1.1 Introduction and Background
Radio waves traversing ionosphere plasma irregularities experience refraction, scattering, and
attenuation [Yeh and Liu, 1982]. While these ionospheric effects may have an adverse impact on
the performance of space-based communication and satellite navigation, they have also provided
a powerful means of passively sensing the environment that created these effects. In the past two
decades, global navigation satellite systems (GNSS) signals have been widely used for
ionospheric monitoring through the establishment of numerous ground-based receiver networks
and satellite-based radio occultation (RO) systems [e.g., Basu et al., 2002; Komjathy et al., 2010;
Mannucci et al., 1999; Rocken et al., 2000]. Transmitted from medium Earth orbit (MEO) or
geostationary Earth orbit (GEO) satellites near 20,000 km altitude, the number of open multiconstellation
GNSS (multi-GNSS) signals with well-defined structures has been increasing at an
accelerated rate. By 2020, there will be over 160 GNSS satellites broadcasting over 400 signals
across the L band, nearly double the number today [Betz, 2013], providing increased
measurement accuracy with global coverage at a low cost. There are, however, many challenges
remaining in effectively utilizing the tremendous amount of multi-GNSS resources to accurately
detect, localize, and characterize disturbances and irregularities in the ionospheric plasma. This
proposal aims to address these three challenges.
1.1.1 Challenge 1: Multi-GNSS RO Receiver Signal Processing Algorithms
When propagating through ionosphere irregularities caused by space weather conditions and
ionosphere internal mechanisms, GNSS signals experience scintillation effects characterized by
simultaneous deep amplitude fading and random carrier phase variations [Morton et al., 2014].
A conventional GNSS receiver carrier phase lock loop (PLL) is not designed to handle such
stress. A number of algorithms have been developed to mitigate ionospheric scintillation effects
and to generate accurate estimations of scintillation signal parameters for ground-based
applications [e.g., Carroll et al, 2014; Humphreys et al., 2010; Kassabian and Morton, 2013,
2014; Mao and Morton, 2013; Peng et al., 2012; Psiaki et al., 2007; Xu et al., 2014; Xu and
Morton, 2015; Yin et al., 2014; Zhang and Morton, 2010, 2013]. RO receivers on LEO
platforms face more serious challenges due to their extensive signal path through the ionosphere,
their platform dynamics, and limited onboard processing resources. Similar challenges are
encountered for RO signals traversing the moist lower troposphere. Open loop (OL) tracking is
implemented as an alternative for LEO RO receivers on board COSMIC satellites when tracking
low altitude troposphere scintillation signals [e.g., Ao et al., 2009; Beyerle et al., 2003;
Sokolovskiy, 2001]. However, OL tracking relies on accurate models of the Doppler frequency,
which are difficult to establish during ionospheric scintillation and over extended time period.
Since accurate carrier parameters are fundamental measurement quantities in RO applications,
novel GNSS carrier tracking algorithms with improved robustness and accuracy are needed to
maintain lock on scintillating signals and to generate accurate carrier phase and Doppler
estimations for signals traversing ionospheric plasma irregularities and the lower troposphere.
1.1.2 Challenge 2: Ionosphere Irregularities Localization Using RO Measurements
GNSS receiver tracking loop outputs, such as a signal’s carrier phase and Doppler frequency,
need to undergo inversion processes in order to obtain ionospheric and atmospheric profiles.
The most widely used RO inversion algorithm to obtain ionosphere electron density (Ne) profiles
is the Abel transform, despite several assumptions that introduce large errors [Hajj and Romans,
1998; Schreiner et al., 1999; Yue et al., 2010, 2011]. To improve the accuracy of ionospheric
ROSES 2014/A.26 GNSS Remote Sensing Science Team
NRA NNH14ZDA001N-GNSS Multi-GNSS Radio Occultation Algorithms
1-­‐
2
profiles, ground-based observations and/or ionosphere models have been used to remove the
spherical symmetry assumption [e.g., Garcia-Fernandez et al., 2003; Hernandez-Pajares et al.,
2000; Jakowski et al., 2002; Schreiner et al., 1999]. Innovative approaches, such as the
maximum entropy method [Hysell, 2007], and the use of data assimilation models to remove the
F region error contributions [Nicolls et al., 2009; Yue et al., 2011] have improved E and lower F
region profile retrievals.
There are, however, few studies in the literature that address retrieval of F region ionospheric
irregularities from RO measurements. And, irregularities observed in RO signals are not always
located at the tangent point along the raypath of the signal. Because the existence of F layer
irregularities seriously violates the horizontal homogeneity assumption in the Abel inversion
process, Ne profiles with irregularity features can only provide a qualitative indication that
significant structures or inhomogeneity exists near the occultation area. Accurate location of the
irregularities cannot be obtained from the retrieved Ne profiles.
A promising approach to localize irregularities at high altitude is physics-based backpropagation
of the complex electromagnetic (EM) fields recorded by LEO satellites along the
raypath. In this approach, the irregularity is treated as an equivalent phase screen [e.g.,
Sokolovskiy, 2000; Sokolovskiy et al., 2002; Vorob’ev et al., 1999]. Sokolovskiy et al. [2002]
applied 2-D back-propagation to high-rate RO signals collected from GPS/MET to localize a
number of ionospheric irregularities in the F layer and above 1000 km. The results were not
validated due to lack of co-located data and the rapidly changing state of the ionosphere. The
method is also limited by the assumption that the amplitude modulation induced by the
irregularities must be small inside the irregularity volume. Recently, Carrano et al. [2014]
successfully demonstrated a technique to back-propagate strong GPS amplitude and phase
scintillation signals from ground-based receiver measurements to construct phase screens.
Correlated RO receiver and ground-based common volume observations of amplitude and phase
scintillation are needed to validate and improve the physics-based techniques and evaluate the
accuracy of the irregularity estimations.
1.1.3 Challenge 3: Polar Ionospheric Irregularity Characterization Through Interferometry
The polar ionosphere has direct access to the interplanetary space and the magnetosphere, and
consequently mostly prone to space weather effects. During active solar conditions, the Sun
dumps massive magnetized plasma and kinetic energy into the terrestrial environment. A
geomagnetic storm is a manifestation of the response of the Earth’s upper atmosphere and the
polar ionosphere more directly. During geomagnetic storms, the polar ionosphere is substantially
distorted compared to the quiet-time characteristics, exhibiting rapid spatial and temporal
fluctuations of the ionization content and altered refraction index. Consequently, the phase and
amplitude of GNSS signals propagating through the polar ionosphere exhibit scintillation effects
[Jiao et al., 2013; Skone et al., 2008, 2009]. Figure 1 shows an example of the geomagnetic
field disturbances and affected GNSS satellite signals at Gakona, AK on July 15, 2012.
Generation of ionospheric irregularities is mostly due to storm-time free energy from plasma
density gradients, external electric fields, particle precipitations, velocity shears, field-aligned
currents, etc. The irregularities have broadband scales and accordingly interact selectively and
differently with different GNSS signals. Comprehensive studies of the polar ionospheric
interaction with and reaction to solar and geomagnetic activities require a large network of
receivers that could track all visible multi-GNSS satellites at all times to produce interferometric
imaging of the ionosphere. The results of Challenges 1 and 2 discussed above are critical to
perform accurate estimates of temporal and spatial variability of ionospheric irregularities.
ROSES 2014/A.26 GNSS Remote Sensing Science Team
NRA NNH14ZDA001N-GNSS Multi-GNSS Radio Occultation Algorithms
1-­‐
3
Figure 1. Geomagnetic field variations and percent of GNSS satellite signals exhibiting scintillation
effects at Gakona, Alaska, on July 15, 2012.
1.2 Objectives and Expected Significance
GPS RO limb-sounding techniques have evolved in parallel with the advancement of GNSS
from a proof-of-concept to operational systems that provide global weather forecasting, climate
monitoring, and ionosphere studies. The COSMIC-2 constellation will utilize the latest
advancements in multi-GNSS through its tri-GNSS receivers to track open signals from GPS,
GLONASS, and Galileo satellites. This new generation of RO systems is expected to increase
the number of atmospheric and ionospheric profiles by an order of magnitude, and to drastically
improve their measurement resolution. To maximize these anticipated benefits, we propose
studies to achieve the following objectives by addressing the challenges presented above:
Objective 1: Develop robust and accurate multi-GNSS receiver tracking algorithms to handle
strong ionospheric and lower tropospheric RO scintillation signals.
Objective 2: Develop data-driven, physics-based methodologies to accurately localize
ionospheric irregularities and simulate RO scintillation signals.
Objective 3: Perform mixed-scale multi-GNSS interferometry to characterize polar ionospheric
irregularities with unprecedented spatial and time resolutions.
Objective 1 will be achieved by exploiting space, time, frequency, and constellation diversity
of modern multi-GNSS signals. In recent years, we have established a unique event-driven
wideband multi-GNSS data collection network [Jiao et al., 2014; Morton et al., 2014; Peng and
Morton, 2012; Pelgrum et al., 2011; Taylor et al., 2012] at strategically selected locations shown
in Figure 2. The network has amassed a large amount of data containing strong ionospheric
scintillation. These data, as well as simulated RO signals and wideband RO samples to be
collected from Haleakala, a high elevation mountaintop in Hawaii, will be used to characterize
scintillation signal structures and support algorithm development and performance evaluation.
Objective 2 addresses the challenges to localize the irregularities at high altitudes. Our
proposed data-driven, physics-based technique is based on joint processing of common volume
RO and ground-based GNSS measurements. The process will start with identification of
ionospheric scintillation from a ground-based network and from RO profiles obtained using Abel
inversion algorithm. Back-propagation of EM fields from both ground-based receiver arrays and
space-based RO receivers will be implemented to localize equivalent phase screens of
0 2 4 6 8 10 12 14 16 18 20 22 24
0
3
6
0
100
-800
0
800
Time (Hours)
Number of SV Percent B Field Variation (nT)
H
D
Z
No. SV Above 30o Elevation
No. SV Affected
Percent of SV Affected
0 2 4 6 8 10 12 14 16 18 20 22 24
0
3
6
0
100
-800
0
Time (Hours)
Number of SV Percent B Field Variation (nT)
H
D
Z
No. SV Above 30o Elevation
No. SV Affected
Percent of SV Affected
ROSES 2014/A.26 GNSS Remote Sensing Science Team
NRA NNH14ZDA001N-GNSS Multi-GNSS Radio Occultation Algorithms
1-­‐
4
irregularities and quantitatively validate the results. The proliferation of multi-GNSS signals in
space may provide multi-dimensional characterization of ionospheric irregularities, and be used
to further constrain the RO inversion algorithm to improve the accuracy of retrieved ionospheric
profiles. Forward propagation of GNSS signals through identified equivalent phase screens will
allow us to validate the back-propagation algorithm and generate simulated multi-GNSS RO
scintillation signals for algorithm testing and evaluations.
Figure 2. Event-driven wideband multi-GNSS data collection networks established and/or operated by the
proposal team. The horizontal and vertical labels are geodetic latitude and longitude in degrees.
Objective 3 will be accomplished by utilizing measurements from multi-scale multi-GNSS
networks at northern hemisphere high latitudes to establish interferometric imaging of the spatial
and temporal evolution of the dynamic ionosphere. The majority of our ground-based multi-
GNSS receiver arrays are also co-located at major ionosphere research facilities where active RF
sounding instruments, incoherent scatter radars, and optical imagers are available to augment the
GNSS measurements and provide validation support. The results obtained from achieving
objective 1 and 2 will allow us to utilize RO measurements to further refine and validate the
interferometric images. The high-resolution interferometry results will make it possible to
characterize the production, distribution, and evolution of ionospheric irregularities.
The proposed activities will enable us to develop more accurate, efficient, and robust GNSS
RO systems for next-generation remote sensing applications and to demonstrate the usefulness of
combining space-based and ground-based multi-GNSS measurements for distributed sensing of
the dynamic ionosphere. Accurate representations of ionospheric profiles are important not only
for ionospheric investigations, they also impact the quality of corresponding lower atmospheric
profiles, and affect characterization of perturbations that are driven by other natural and manmade
processes occurring at the Earth’s surface. The objectives of the proposed work are
therefore in line with the NRA’s goal of seeking innovative approaches to the development of
GNSS remote sensing techniques and algorithms to advance Earth system science objectives.
Gakona
Alaska
Arecibo,
Puerto
Rico
Hong
Kong
Singapore
Jicamarca,
Peru
Ascension
Island
Established
Sites
Planned
Sites
Magnetic
equator
±15o
Magnetic
latitude
50%
Auroral
oval
90%
Auroral
oval
Background
map
and
geomagnetic
boundaries
of
interests
courtesy
of
James
Secan,
Northwest
Reseach
Associates,
Inc.
Poker
Flat
Sondrestrom
Greenland
Haleakala
Hawaii
Colorado
Resolute
Bay
Canada
Temporary
Sites
ROSES 2014/A.26 GNSS Remote Sensing Science Team
NRA NNH14ZDA001N-GNSS Multi-GNSS Radio Occultation Algorithms
1-­‐
5
1.3 Technical Approach and Methodology
We propose the following three tasks aimed to achieve the objectives outlined above.
1.3.1 Task 1 - Developing Robust and Accurate Multi-GNSS RO Receiver Tracking Algorithms
to Handle Ionospheric and Lower Tropospheric RO Scintillation Signals
Accurate carrier phase tracking of strong scintillation signals is a challenging task because of the
conflicting design criteria imposed by simultaneous deep amplitude fading and high carrier
dynamics. For this reason, most groundbased
GNSS receiver networks
established to monitor ionosphere
scintillations do not perform well during
strong scintillations. To address this
issue, we developed an event-driven
multi-GNSS intermediate frequency (IF)
data collection system. Figure 3 shows
the schematic of the system. A
conventional ionosphere scintillationmonitoring
(ISM) receiver continuously
processes multi-constellation signals. An
array of wideband RF front ends samples
IF inputs and temporarily stores the data
in circular buffers. If the ISM receiver
detects scintillation, our custom
designed trigger software retrieves the
circular buffer contents to a permanent
storage system. The stored data are postprocessed
using our custom receiver
tracking software. Figure 4 shows an
example of phase scintillation index for GPS PRN24 L1, L2C, and L5 signals over Ascension
Island on March 10, 2013. The ISM receiver (solid lines) had numerous carrier phase cycle slips
and lost lock of signals. In contrast, our software-defined radio (SDR) algorithms were able to
maintain lock of the same signals recorded by the IF data collections system (dotted lines).
Figure 4. GPS Phase
scintillation index for L1,
L2, and L5 signals during
a strong scintillation
event. The solid lines are
outputs of an ionospheric
scintillation-monitoring
(ISM) receiver. The
dotted lines are postprocessed
results from
recorded IF data using our
software-defined radio
(SDR) carrier tracking
algorithms. From [Morton
21:00 21:05 21:10 21:15 21:20 et al., 2014].
0
10
20
30
40
50
60
70
80
σ
φ
(degrees)
UTC Time (HH:MM)
Ascension Island. March 10, 2013. GPS PRN 24
SDR
ISM L1
L2
L5
Cycle
slip
or
loss
of
lock
Figure 3. Schematic of event-driven wideband
reconfigurable multi-GNSS data collection system.
ROSES 2014/A.26 GNSS Remote Sensing Science Team
NRA NNH14ZDA001N-GNSS Multi-GNSS Radio Occultation Algorithms
1-­‐
6
We designed several novel GNSS receiver carrier-tracking algorithms to handle strong
scintillations for ground-based receivers. For example, Peng et al. [2012] presented a mixed
PLL and a vector tracking loop (VTL) algorithm in which undisturbed signals are processed by a
conventional PLL, while strong scintillation signals are tracked using feedbacks from VTL
outputs. A signal integrity-monitoring module generates indicators of scintillation level or
“health” status of each channel. A VTL computes the position, velocity, and timing (PVT)
solutions based on outputs from healthy PLL channels to construct the code phase and carrier
Doppler frequency model as feedback for the stressed channels. Xu et al. [2014] expanded this
algorithm to dual constellation tracking and successfully demonstrated its feasibility to track
GPS L1 and BeiDou B1 signals under strong scintillation conditions.
Xu and Morton [2015] further applied the approach to ground-based receivers with known
surveyed positions. This so-called Fixed-Position-Feedback (FPF) algorithm applies extended
integration time and accurate receiver and satellite position information to reveal carrier phase
structures during deep signal fading. Figure 5 shows example results obtained by applying the
FPF algorithms to Ascension Island IF data. Carrier phases on GPS PRN 24 L1 and L5 signals
both showed half-cycle changes
(upper panel) during their deep
amplitude fading (lower panel).
Our investigation showed that
phase reversal during deep fading
is not uncommon in equatorial
scintillation [Xu and Morton,
2015]. Does this phenomenon
occur with RO measurements in
the ionosphere and in the lower
troposphere? What are the
underlying ionosphere and
atmosphere properties that lead
to such phenomena? Accurate
carrier phase estimations hold the
key to reliable retrievals of
atmospheric and ionospheric
profiles from RO measurements.
Being able to uncover phase
structures during deep fading is
critical for sensing ionospheric
irregularities and troposphere
water vapor.
The algorithms discussed above exploit spatial diversity of ionospheric irregularities, with
the expectation that some GNSS satellite signals will arrive at the receiver without penetrating
irregularities. These healthy signals are used to derive receiver PVT solutions to generate
feedback parameters for the scintillating channel. As the number of multi-GNSS signals
increases, this strategy will continue to improve its performance for ground-based receivers.
On LEO satellites such as COSMIC, a frequency model is used as the reference to enable
open loop (OL) tracking for low altitude occultation, while closed loop (CL) is used to track high
altitude occultation signals for ionosphere profiling. The OL frequency model is based on
!
Figure 5. GPS L1 and L5 carrier phase reversals during deep
amplitude fading. The plots are generated by applying the Fixed-
Position-Feedback (FPF) algorithm [Xu and Morton, 2015] to
Ascension Island IF data collected on March 8, 2013.
!
ROSES 2014/A.26 GNSS Remote Sensing Science Team
NRA NNH14ZDA001N-GNSS Multi-GNSS Radio Occultation Algorithms
1-­‐
7
satellite PVT solutions determined by a separate receiver and other onboard navigation sensors
[Tseng et al., 2014]. The VTL and FPF described above are in essence adaptive mixed mode
tracking techniques as they continuously evaluate the signal conditions and tracking loop stress
to decide the mode of operation at each integration period. OL tracking is only evoked when the
signal intensity drops below a certain threshold value. CL tracking will take over after the signal
power returns to above the threshold. Instead of blanking the entire flight of signals through
strong scintillation regions with OL tracking, these algorithms evoke OL tracking only during the
short intervals when deep fading occurs.
Our studies of ground-based ionospheric scintillation indicate that the average deep fading
duration for equatorial scintillations is 60~100 ms with average time between consecutive deep
fading being 5~10 seconds [Morton et al., 2015]. Therefore, it is possible for OL tracking to
operate for ~100 ms before handing over to CL tracking. Unlike the current RO receiver OL
tracking method, the VTL and FPF algorithms have knowledge of the most recent signal
parameters, and therefore can generate a more accurate frequency and code reference model for
the short time period during deep signal fading. By adopting these algorithms for RO receivers,
we are effectively exploiting temporal diversity of scintillation signals. More studies are needed
to characterize the deep fading durations and consecutive fading time for lower troposphere
scintillations to have a better understanding of the benefit of the approach at lower troposphere.
Adaptive multi-frequency
(AMF) tracking is another
approach that can be applied to
RO receivers [Yin et al., 2014].
Figure 6 shows an example of
GPS L1, L2C, and L5 signal
intensity during an intense
scintillation event over Ascension
Island. The plot shows that deep
fading on L1, L2C, and L5 do not
always occur simultaneously.
Statistical analysis of thousands of
triple-frequency GPS signal deep
fading based on data collected on
Ascension Island, and in
Singapore and Hong Kong show
that the probability of having
simultaneous fading across all
three GPS bands is less than 4%
[Morton et al., 2015].
Yin et al. [2014] presented the architecture of the AMF algorithm and tested its performance
by tracking real triple-frequency GPS scintillation data. Figure 7 shows the carrier-to-noisedensity
ratio (C/N0) of GPS PRN 25 L1, L2C, and L5 signals on March 7, 2013 on Ascension
Island generated by the AMF algorithm. The dashed green rectangles indicate when deep fading
occurred on at lease one frequency and adaptive frequency aiding was automatically evoked by
the AMF. Conventional tracking algorithms lost lock on signals during this period.
Figure 6. Signal intensity for GPS PRN 25 L1, L2C, and L5
signals over Ascension Island on March 5, 2013. The three
marked fading events show that fading do not occur
simultaneously on the same satellite signals at different carriers.
!!
ROSES 2014/A.26 GNSS Remote Sensing Science Team
NRA NNH14ZDA001N-GNSS Multi-GNSS Radio Occultation Algorithms
1-­‐
8
Figure 7. C/N0 from AMF tracking for PRN 25 L1, L2C, and L5 signals during strong scintillation on
March 7, 2013, over Ascension Island. The AMF algorithm was able to maintain lock on all three signals
throughout this very challenging time period. The dashed green rectangles indicate where deep fading
occurred on at lease one frequency and adaptive frequency aiding was automatically evoked by the AMF.
The AMF method can be directly applied to RO receivers to track ionospheric scintillation
signals, as similar dispersive behavior should occur in RO signals. For lower troposphere
scintillation, the applicability of this method will be determined by the outcome of investigations
of multi-frequency fading properties of water vapor scintillation. Such an investigation requires
wideband IF RO data propagating through
the lower troposphere with rich water
vapor contents. The PI has obtained funds
from Air Force Research Laboratory to
collect multi-frequency troposphere
scintillation data [Morton, 2015a]. The
experiment will be conducted on April 15-
26, 2015 using wideband RF front ends
and a high gain antenna set up on
Haleakala, a mountain peak with 3000 m
elevation on Maui in the Hawaiian
Islands. While the mountain top data is
intended for airborne scintillation
research, the results will also be used to
support the proposed studies.
We propose an adaptive open loop
(AOL) architecture that exploits both time
and frequency diversities by integrating
mixed PLL and VTL with AMF to further
improve the robustness and accuracy of
RO receiver tracking for both ionosphere
and lower troposphere. In this proposed
new architecture as depicted in Figure 8,
75 80 85 90 95 100
0
5
10
15
20
25
30
35
40
45
Time (s)
C/N0(dB-Hz)
L1:AMF
L2:AMF
L5:AMF
L5
aids
L1
&
L2
L1
aids
L2
&
L5
L5
&
L1
aid
L2
1
3
4
5
6
2
Figure 8. Proposed adaptive open loop (AOL) tracking
method based on integrated VTL and AMF tracking
algorithms concepts.
!
Carrier∧&code&
tracking&process
Wideband&Digital&Input&Samples& Compute&tracking&loop&
stress&indicator(s):&
; & Signal&intensity
; & Phase&error
; & Frequency&error
Stress&indicator(s)&
>&Threshold?
Construct&optimized&carrier/code&models&using:&
; & Predication&from&recent&same&channel&
parameters&–&temporal&diversity
; & Tracking&loop&outputs&from&other&frequency&
band(s)&on&the&same&occultation&satellite&–&
frequency&diversity
; & Platform&PVT&solutions∧&GNSS&ephemeris&
–&traditional&open&loop&approach
Close&Loop!using!
same!channel!estimation!
as!feedback
Open&Loop&using!
optimized!carrier/code!models!
as!feedback Yes
No
Figure 8. Proposed adaptive open loop (AOL)
tracking algorithm based on integrated VTL and
AMF algorithms concepts
ROSES 2014/A.26 GNSS Remote Sensing Science Team
NRA NNH14ZDA001N-GNSS Multi-GNSS Radio Occultation Algorithms
1-­‐
9
the default operation is the conventional CL PLL for signals at each carrier frequency. The
tracking loop outputs are used to compute stress indicators such as signal intensity, carrier phase
and Doppler frequency errors. Note that C/N0 is not in the proposed scheme because our studies
have found that it is not suitable as a sensitive indicator for amplitude fading [Jiao et al., 2014].
Predetermined stress threshold values can be established based on prior experiments or
simulation studies of RO data. The stress indicators are compared with the thresholds values at
each integration period. CL tracking will use its own channel outputs as feedback, as long as the
stress indicators remain below their corresponding thresholds values. If the threshold values are
exceeded, an optimization process will determine and construct the signal carrier and code
models using information from three sources: prediction from recent same-channel CL tracking
outputs, outputs mapped from other healthy carriers transmitted from the same occultation
satellites, and platform PVT solutions and GNSS ephemerides. The first source introduces
temporal diversity by using recent same channel prior estimations. The second source utilizes
frequency diversity by incorporating aiding information from other frequency channels. And the
last source is the current OL implementation on the RO platform. We envision that for future RO
systems, the proposed AOL architecture could replace the current dedicated OL and CL
operations to yield optimized performances at all altitudes.
To evaluate the performances of the algorithms, we will use real scintillation IF data
collected at our high latitude and equatorial stations, Haleakala mountain top RO IF data, and
simulated RO scintillation data. The RO simulation data generation is part of Task 2 to be
discussed next.
1.3.2 Task 2 - Data-Driven Physics-Based Ionospheric Irregularities Localization
The presence of ionosphere F layer irregularities renders invalid the spherical asymmetry
assumption in the Abel inversion, resulting in inaccurate retrieval of Ne profiles as well as
incorrect location of the irregularities. The Ne profile errors will propagate down in altitude to
impact lower ionosphere and troposphere profiles retrieval [Hysell, 2007; Nicolls et al., 2009;
Yue et al., 2011]. We propose to extend the phase-screen approach presented by Bernhardt et al.
[2006], Carrano et al. [2012], and Sokolovskiy et al. [2002] to back-propagate measurements
from both ground-based GNSS receivers and LEO RO receivers to localize high altitude
irregularities in a common volume intercepted by the signal paths to both kinds of receivers.
With the proliferation of multi-GNSS signals and advancement of LEO RO technologies,
more RO profiles and ground-based measurements with ionospheric scintillation signatures are
becoming available. Figure 9 shows an example of RO and ground-based observations of
ionospheric scintillation effects near our GNSS array in Gakona, Alaska. Two COSMIC RO
profiles with scintillation structures (as contrasted by a background 2012 IRI model profile)
corresponding to 16:49 and 17:04 UTC on March 15, 2012, are shown in the left panel. The large
Ne peaks in the D and lower E regions are most likely due to propagation of large F region
retrieval errors in the presence of irregularities. The middle panel shows 9 GPS and 7 GLONASS
satellite tracks from 15:56 to 17:56 UTC, within a +/-15o longitude and +/-10o latitude window
centered over the ground receiver array. The gray trajectories in the middle panel correspond to
tangent points of the RO profiles; with darker colors indicate occultation at lower altitudes. The
GPS and GLONASS satellite tracks are color-coded according to their carrier phase scintillation
index as indicated by the color scheme below the middle panel. Because of the higher phase
noise associated with GLONASS signals, two different color scales are used for GPS and
GLONASS measurements. The right panel shows the phase scintillation index for all satellites
during the two-hour span.
ROSES 2014/A.26 GNSS Remote Sensing Science Team
NRA NNH14ZDA001N-GNSS Multi-GNSS Radio Occultation Algorithms
1
-­‐1
0
Figure 9. Ionospheric irregularities and scintillation effects observed simultaneously from COSMIC RO
Ne profiles (left panel) and our ground-based GNSS array at Gakona, Alaska (middle and right panel) on
March 15, 2012. Two COSMIC RO profiles with scintillation structures retrieved at 16:49 and 17:04
UTC and a background IRI (2012) model profile at 16:56 UTC are shown in the left panel. The RO
profile tangent point tracks are shown as the gray trajectories in the middle panel with the darker colors
corresponding to occultation at lower altitudes. The tracks in the middle panel are color-coded with GPS
and GLONASS signal carrier phase scintillation index values. Each satellite track is identified by its
constellation (G: GPS; R: GLONASS), followed by the satellite PRN (GPS) or slot number (GLONASS).
The right panel shows the phase scintillation index for the satellites during the two-hour span.
Although the two occultation profiles shown in Figure 9 intercept multiple GPS and
GLONASS satellite signal paths tainted with strong phase fluctuations, the irregularity structures
shown in the COSMIC profiles do not necessary correspond to ionospheric irregularities at the
tangent point. Figure 10 illustrates the geometrical relationship among a GNSS-LEO signal
path, the tangent point of a retrieved Ne profile from the LEO satellite, potential location of a
plasma bubble, and a ground-based receiver reception of a different GNSS satellite signal
through the same plasma bubble. We propose an algorithm that uses joint LEO and ground-based
multi-GNSS receivers measurements to localize ionospheric irregularities and validate the
results. LEO receivers will include the ones on COSMIC, COSMIC-2, and the Canadian
CASSIOPE satellite which generates up to 100Hz high rate data [Kim and Langley, 2010; Shume
et al., 2015]. Figure 11 is the block diagram of this proposed method.
The algorithm starts with identification of a geographic-and-time window based on
ionospheric scintillations observed from ground-based multi-GNSS network. Within this
geographic and time window, we search for potential COSMIC, CASSIOPE, and in future,
COSMIC-2 RO occultation events. Ne profiles will be obtained using the Abel inversion
algorithm for identified occultation events. If scintillation structures exist on a profile, complex
EM fields will be constructed using high rate RO receiver tracking loop amplitude and carrier
phase outputs and back-propagated along the projected LEO signal path. The location at which
minimum phase fluctuation occurs corresponds to the equivalent phase screen location of the
irregularities. The altitude and horizontal location of the irregularities can be mapped from the
phase screen location along the occultation signal path.
ROSES 2014/A.26 GNSS Remote Sensing Science Team
NRA NNH14ZDA001N-GNSS Multi-GNSS Radio Occultation Algorithms
1
-­‐1
1
Figure 10. Illustration of geometrical relationship among a GNSS-LEO satellite signal path, the tangent
point of retrieved ionosphere profile from the LEO satellite, potential location of a plasma bubble, and a
ground-based receiver reception of a different GNSS satellite signal through the same plasma bubble.
Figure 11. Block diagram of the proposed ionospheric irregularity localization algorithm using joint LEO
and ground-based multi-GNSS receiver measurements.
The identified altitude and horizontal location of the irregularities will be validated using
local ground-based GNSS receiver measurements. With the known coordinates of the groundbased
GNSS receivers and GNSS ephemeris, we can determine which receiver-satellite signal
paths traverse the identified irregularities. Scintillation indices can be computed during the time
window near the occultation event to qualitatively validate the existence of the irregularities.
To quantitatively evaluate the irregularities’ locations, the same LEO signal backpropagation
procedure can be applied to ground-based receivers to locate equivalent phase
screens along their signal paths, leading to new estimations of the irregularity altitude and
horizontal locations. These new estimations can be used to evaluate the accuracy of the
estimation obtained from LEO measurements. As the number of GNSS satellites increases and
ROSES 2014/A.26 GNSS Remote Sensing Science Team
NRA NNH14ZDA001N-GNSS Multi-GNSS Radio Occultation Algorithms
1
-­‐1
2
GNSS receiver tracking algorithms improve, the number of LEO and ground observations within
a defined geographic area will also increase. It is therefore possible for the same irregularities to
be intercepted by multiple signals received by LEO and ground GNSS receivers. Scintillations
observed on the LEO platform and by the ground-based network corresponding to these different
signal paths will reveal characteristics of the same irregularities along different dimensions.
There are a number of potential error sources that impact the algorithm performance and
accuracy. The phase screen equivalence assumption for ionization irregularity is affected by the
level of amplitude modulation inside the irregularity volume. The proposed studies will explore
means to improve the physics-based approach by investigating back-propagations through
multiple phase screens. The projected signal propagation path is determined by the background
ionosphere. Geographical locations, tangent point altitude, and background geomagnetic field
vector estimation error will all contribute to the error budget of the localization results. Detailed
analysis of these potential error impact factors and observations along different dimensions of
the same irregularities will be conducted in this study.
The reverse process of back-propagation is the forward propagation of incident GNSS EM
waves through phase screens. Forward propagation will allow us to validate the backpropagation
results and simulate scintillation RO signals at a receiver [Sokolovskiy et al., 2002;
Carrano et al., 2011]. Task 2 will implement forward propagation of multi-frequency GNSS
signals through phase screens identified by the back-propagation algorithm to simulate RO
signals at LEO satellite for testing and evaluations of the AOL algorithm proposed in Task 1.
1.3.3 Task 3 - Joint RO and Ground-Based Network Ionosphere Interferometry
We propose to utilize the network of space- and ground-based GNSS receivers, which
continuously track all GNSS satellites in view, for interferometric imaging of the spatial and
temporal evolution of the polar
ionosphere. The availability of such
a large network of GNSS receivers
creates the opportunity for multiple
baseline interferometry to form
ionospheric images. Figure 12 shows
a GNSS receiver array, which is
convenient to form manifold
baselines in Greenland, and a
closely-spaced small array in Poker
Flat, AK. An additional closelyspaced
array will be established in
Resolute Bay by the summer of 2016
[Morton, 2015b]. These networks,
along with the Canadian High Arctic
Ionospheric Network (CHAIN)
[Jayachandran et al., 2009], will be
used to support this proposed study.
In this proposal, the GNSS array is utilized as an interferometer device to measure the spatial
coherence function by coherently integrating the product of GNSS carrier phase measurements
from a pair of GNSS receivers in the array. The integration should be performed over very small
angular separations where the signals are received from same direction and have coherence in the
ionosphere. Similarly, we will estimate the spatial coherence function for several baseline
ROSES 2014/A.26 GNSS Remote Sensing Science Team
NRA NNH14ZDA001N-GNSS Multi-GNSS Radio Occultation Algorithms
1
-­‐1
3
combinations in the GNSS array. Ionospheric images are synthesized by performing the Fourier
transform on the spatial coherence function for all baselines in the interferometry. The GNSS
array in Greenland provides ~1700 interferometric baselines, making it possible to achieve high
resolution images. These synthesized images will reveal high spatial and temporal characteristics
of the ionospheric irregularities with unprecedented details.
In addition, 3D ionospheric image will be constructed by augmenting the ground-based
GNSS observations with RO profiles obtained using e.g., COSMIC-1, CASSIOPE, and the
prospective COSMIC-2 constellations. The number of daily RO measurements over Greenland
and Alaska will be greatly enhanced with the impending launch of the COSMIC-2 constellation,
which, along with CASSIOPE, will provide us with polar coverage to investigate the 3D
evolution of high-latitude plasma irregularity processes—a telltale indicator of solar-terrestrial
relations. These details can be further compared and analyzed with results generated by the
irregularities localization method proposed in Task 2 and by nearby incoherent scatter radars, RF
sounding instruments, and optical imagers.
1.4 Perceived Impact on the State of Knowledge
The proposed research will have a major impact on advancing multi-GNSS RO receiver
technologies for ionosphere and troposphere remote sensing and on improving our understanding
of GNSS signal propagation effects through ionospheric irregularities and the lower troposphere
on RO platforms. The proposed AOL tracking algorithm and the data-driven physics-based
irregularities localization technique will lead to more accurate, efficient, and robust GNSS RO
systems in high latitude and equatorial areas and at lower troposphere. Our unique event-driven
IF data collection capabilities will supply high quality data for characterization of ionospheric
and tropospheric scintillation and provide design guidance and a unique test bed for new RO
algorithms development. The combined ground- and space-based GNSS for high-resolution
images of the ionosphere will advance our understanding of high latitude ionospheric responses
to solar/geomagnetic activities.
1.5 NASA Programmatic Relevance
The proposed research relates directly to NASA objectives outlined in ROSES 2014/A.26 GNSS
Remote Sensing Science Team solicitation, which “seeks innovative approaches to the
development of GNSS remote sensing techniques and algorithms to advance Earth system
science objectives” and “develop occultation techniques with a focus upon the broader utilization
of existing GNSS signals such as GPS (L1 C/A, L2C and L5) and the GLONASS … signals in
preparation for future space borne receiver capabilities such as the TriG Receiver.” The
proposed RO receiver algorithms leverage our proven techniques for ground-based multi-GNSS
signal processing and our unique event-driven wideband multi-GNSS data collection
capabilities. Our proposed strategies based on exploitation of diversities offered by new GNSS
signals beyond the current legacy operations are in direct response to NRC’s Decadal Survey
recommendations. Finally, our proposed high resolution imaging of the polar ionosphere
demonstrates the potential of utilizing multi-scale ground-based multi-GNSS array with
augmentation from space-based RO observations. The outcomes will substantiate NASA’s
vision that new GNSS signal structures will provide unprecedented opportunities for remote
sensing of the Earth system with new ground-based systems and relatively simple and robust
space borne GNSS receivers.
ROSES 2014/A.26 GNSS Remote Sensing Science Team
NRA NNH14ZDA001N-GNSS Multi-GNSS Radio Occultation Algorithms
1
-­‐1
4
1.6 Plan of Work
1.6.1 Key Milestones
We request four years of support for efforts towards GNSS RO receiver tracking and ionospheric
irregularities localization algorithms development at CSU and high resolution polar ionospheric
imaging and analysis at JPL. The proposed efforts will be completed jointly by the participating
senior personnel, a graduate student, and a postdoctoral researcher. The outcomes of the
proposed research will be a combination of methodologies, algorithms, and scientific results.
We will report these outcomes throughout the period of performance at appropriate conferences
(e.g., AGU Fall Meetings and Institute of Navigation meetings, and submit publications to
relevant peer-reviewed journals (i.e., Radio Science, IEEE Transactions on Aerospace &
Electronics Systems, IEEE Transactions on Remote Sensing & Geosciences).
Year 1 (2015-16): Algorithm development, RO scintillation simulation.
CSU Task Emphasis: Task 1 - design architecture and component development/testing of AOL
tracking algorithms for multi-frequency GPS signals; establishing stress indicators, selecting
threshold level settings, and optimization of code and carrier models; processing mountaintop IF
data using existing custom algorithms and analyze troposphere scintillation signal temporal and
spectral characteristics. Task 2 - Identify geographic and time windows based on ground-based
and RO scintillation data to support ionosphere irregularity localization algorithm development;
implement RO forward-propagation algorithm to generate simulated wideband IF data at LEO
RO and ground receivers.
JPL Task Emphasis: Task 3 - Develop algorithms to process GNSS carrier phase observations
made by the GNSS arrays for forming multi-baseline interferometry; generate high-resolution
images and animations, which will reveal the spatial and temporal evolutions of the high-latitude
ionosphere. The algorithm development will include new software, which (1) performs
calculation of the spatial coherence function for each baseline (as many as ~1700) using
interferometry, and (2) performs inverse Fourier transform operations on the spatial coherence
function for all baselines.
Year 2 (2016-17): Algorithm development, GNSS scintillation signal characterization.
CSU Task Emphasis: Task 1 - Expand component development/testing of AOL tracking
algorithm to Galileo multi-carrier and GLONASS frequency-division-multiple-access signal
processing, with special attention to mitigate GLONASS signal phase noise to improve its carrier
phase measurement quality; threshold settings, and code/carrier model optimization for Galileo
and GLONASS. Task 2 - Apply Abel inversion to obtain ionosphere Ne profiles within
identified geographic and time window; identify and archive RO scintillation profiles; implement
2D RO back-propagation algorithm to locate equivalent phase screens for identified profiles; test
the back-propagation algorithm using simulated RO data during Year 1 effort.
JPL Task Emphasis: Task 3 - Test and validate computer codes extensively both on simulated
and actual observations using JPL’s supercomputing facilities; develop computer codes for
processing and augmenting the ground-based GNSS observations with RO observations for 3D
image construction; identify key events and start processing ground-based and space borne data
using algorithms developed in Year 1 and part of Year 2.
Year 3 (2017-18): Algorithm integration and validation.
CSU Task Emphasis: Task 1 - AOL tracking algorithm component integration of Year 1 and 2
results; testing and evaluation of integrated algorithms using ground-based scintillation data,
mountaintop RO IF data, and simulated RO scintillation data; provide close-spaced receiver
ROSES 2014/A.26 GNSS Remote Sensing Science Team
NRA NNH14ZDA001N-GNSS Multi-GNSS Radio Occultation Algorithms
1
-­‐1
5
array scintillation tracking results to JPL team. Task 2 - Implement back-propagation algorithm
for ground receiver measurements; apply the algorithm to test/validate/refine locations of
ionospheric irregularities derived from back-propagation of RO signals during year 2.
JPL Task Emphasis: Task 3 – Utilize scientific software developed in Years 1 and 2 for
processing the vast amount of ground-based and space-based GNSS observations; perform indepth
analysis of processed events.
Year 4 (2018-19): Algorithm performance enhancement and error analysis.
CSU Task Emphasis: Task 1 - Refine, optimize, and speed up integrated AOL tracking algorithm
developed in Year 1-3 for real-time applications on LEO platforms. Task 2: Error analysis of
back propagation algorithm derived irregularity locations; comparison with nearby vertical
sounding instruments, incoherent scatter radar, and optical imager measurements.
JPL Task Emphasis: Task 3 - Interpret our results, prepare manuscripts for publication, present
results at national and international scientific conferences; catalog and prepare processed data for
archiving and release products to the scientific community for further scrutiny and analysis.
1.6.2 Management Structure
The PI, Dr. Jade Morton, will be responsible for managing all technical and administrative
aspects of the proposed efforts. She will lead the algorithm development, implementation,
testing and validation, and error analysis as described in Task 1 and 2 at CSU. She will
coordinate collaboration and information/data exchange between the CSU graduate student and
postdoctoral researcher, the NASA JPL PI Dr. Attila Komjathy and Co-I Dr. Esayas Shume,
University of New Brunswick (UNB) Co-I Dr. Richard Langley, and Technical University of
Denmark (TUD) collaborator Mr. Tibor Durgonics.
1.6.3 Contributions of PI and Key Personnel
CSU PI Dr. Jade Morton has extensive experience in GNSS receiver algorithm development,
ionospheric irregularity modeling and remote sensing, and radio wave propagation effects
studies. She will develop the architecture for algorithms in Task 1 and 2, and advise a graduate
student on Task 1 and a postdoctoral researcher on Task 2 implementations.
JPL Institutional PI Dr. Attila Komjathy and Co-I Dr. Esayas Shume will be responsible for
Task 3 effort. Dr. Komjathy has led development of a number of innovative techniques in
ground-based and space-based GNSS remote sensing. Dr. Shume is an expert in solar-terrestrial
relations and space weather effects on GNSS.
UNB Co-I Dr. Richard Langley is the PI of the GPS instrument on CASSIOPE and a science
team member of CHAIN, and a world-leading expert in GNSS and ionosphere monitoring. He
will provide support for CASSIOPE and CHAIN data access and support scientific analysis of
the proposed studies.
TUD collaborator Mr. Tibor Durgonics has experience in GPS remote sensing and oversees the
Greenland GNSS array for which he will provide data access and support scientific analysis.
ROSES 2014/A.26 GNSS Remote Sensing Science Team
NRA NNH14ZDA001N-GNSS Multi-GNSS Radio Occultation Algorithms
2-­‐1
2 References and Citations
Ao, C. O., G. A. Hajj, T. K. Meehan, D. Dong, B. A. Iijima, A. J. Mannucci, and E. R.
Kursinski, “Rising and setting GPS occultations by use of open-loop tracking,” J. Geophy. Res.
Atmos., 114(D4), doi:10.1029/2008JD010483, 2009.
Basu, S., K. Groves, Su. Basu, P. Sultana, “Specification and forecasting of scintillations in
communication and navigation links: current status and future plans,” J. Atmo. Solar-Terr.
Phy., 64(16), 1745-1754, 2002.
Bernhardt, P. A., and C. L. Siefring, “New satellite-based systems for ionospheric tomography
and scintillation region imaging,” Radio Sci., 41(5), doi: 10.1029/2005RS003360, 2006.
Betz, J. W., “Signal structures for satellite-based navigation: past, present, and future,” Proc.
ION Pacific PNT meeting, April 2013, Honolulu, HI.
Beyerle, G., M. E. Gorbunov, and C. O. Ao, “Simulation studies of GPS radio occultation
measurements,” Radio Sci., 38(5), doi:10.1029/2002RS002800, 2003.
Carrano, C. S., K. M. Groves, S. H. Delay, and P. H. Doherty, “An inverse diffraction technique
for scaling measurements of ionospheric scintillations on GPS L1, L2, and L5 carriers to other
frequencies,” Proc. ION ITM, San Diego, CA, January, 2014.
Carrano, C. S., K. M. Groves, R. G. Caton, C. L. Rino, and P. R. Straus, “Multiple phase screen
modeling of ionospheric scintillation along radio occultation raypaths,” Radio Sci., 46,
RS0D07, doi:10.1029/2010RS004591, 2011.
Carroll, M., Y. Morton, E. Vinande, “Triple frequency GPS signal tracking during strong
ionospheric scintillation over Ascension Island,” Proc. IEEE PLANs/ION Annual Meeting,
May 2014, Monterey, CA.
Garcia-Fernandez, M., M. Hernandez-Pajares, M. Juan, and J. Sanz, “Improvement of
ionospheric electron density estimation with GPSMET occultations using Abel inversion and
VTEC information,” J. Geophy. Res. Space Phys., 108(A9), doi:10.1029/2003JA009952,
2003.
Hajj, G. A., and L. J. Romans, “Ionospheric electron density profiles obtained with the Global
Positioning System: Results from the GPS/MET experiment,” Radio Sci., 33(1), 175-190,
1998.
Hernández-Pajares, M., J. M. Juan, and J. Sanz, “Improving the Abel inversion by adding ground
GPS data to LEO radio occultations in ionospheric sounding,” Geophy. Res. Lett., 27(16),
2473-2476, 2000.
Humphreys, T. E., M. Psiaki, B. Ledvina, A. Cerruti, and P. M. Kintner, Jr., “Data-driven testbed
for evaluating GPS carrier tracking loops in ionospheric scintillation,” IEEE Trans. Aero. &
Elec. Sys., 46(4), 1609-1623, Oct. 2010.
Hysell, D. L. "Inverting ionospheric radio occultation measurements using maximum
entropy." Radio Sci., 42(4), doi:10.1029/2007RS003635, 2007.
ROSES 2014/A.26 GNSS Remote Sensing Science Team
NRA NNH14ZDA001N-GNSS Multi-GNSS Radio Occultation Algorithms
2-­‐2
Jakowski, N., A. Wehrenpfennig, S. Heise, Ch Reigber, H. Lühr, L. Grunwaldt, and T. K.
Meehan, “GPS radio occultation measurements of the ionosphere from CHAMP: Early
results,” Geophy. Res. Lett., 29(10), doi:10.1029/2001GL014364, 2002.
Jayachandran, P. T., R. B. Langley, J. W. MacDougall, S. C. Mushini, D. Pokhotelov, A. M.
Hamza, I. R. Mann et al., “Canadian high arctic ionospheric network (CHAIN),” Radio
Sci., 44(1), doi:10.1029/2008RS004046, 2009.
Jiao, Y., Y. Morton, S. Taylor, W. Pelgrum, “Characterization of high latitude ionospheric
scintillation of GPS signals,” Radio Sci., 48, doi:10.1002/2013RS005259, 2013.
Jiao, Y., Y. Morton, S. Taylor, M. Carroll, “Characteristics of low-latitude signal fading across
the GPS frequency bands,” Proc. ION GNSS+, Tempa, FL, Sept. 2014.
Jiao, Y., Y. Morton, S. Taylor, “Comparative studies of high-latitude and equatorial ionospheric
scintillation,” Proc. IEEE/ION PLANS meeting, Monterey, CA, May 2014.
Kassabian, N., Y. Morton, “Extending integration time for Galileo tracking robustness under
ionosphere scintillation,” Proc. IEEE/ION PLANS meeting, Monterey, CA, May 2014.
Kassabian, N., Y. Morton, “Galileo tracking performance under ionosphere scintillation,” 4th Int.
Colloquium Sci. Fundamental Aspects Galileo Prog., Oct. 2013.
Kim, D., and R. B. Langley, “The GPS attitude, positioning, and profiling experiment for the
enhanced polar outflow probe platform on the Canadian CASSIOPE satellite,” Geomatica,
64(2), 233-243, 2010.
Komjathy, A., B. Wilson, X. Pi, V. Akopian, M. Dumett, B. Iijima, O. Verkhoglyadova, and A.
J. Mannucci, “JPL/USC CGAIM: On the impact of using COSMIC and ground-based GPS
measurements to estimate ionospheric parameters,” J. Geophys. Res., 115, A02307,
doi:10.1029/2009JA014420, 2010.
Mannucci, A. J., B. A. Iijima, U. J. Lindqwister, X. Pi, L. Sparks, B. D. Wilson, “GPS and
Ionosphere,” Revised Submission to URSI Reviews of Radio Science, March 1999.
Mao, X., and Y. Morton, “GNSS receiver carrier tracking loop designs for robust navigation
application and for ionosphere studies,” Proc. ION GNSS+, Nashville, TN., Sept. 2013
Morton, Y., An add-on project to CONRAD: Collaborative research and development program
on navigation and time-keeping with Air Force Research Laboratory, WPAFB, FA8650-08-D-
1451 0001, 2015a.
Morton, Y., ANSWER: Advanced novel spectrum warfare environment research with Air Force
Research Laboratory, WPAFB, FA8650-14-D-1735 0002,
2015b.
Morton, Y., H. Bourne, M. Carroll, Y. Jiao, N. Kassabian, S. Taylor, J. Wang, D. Xu, H. Yin,
“Multi-constellation GNSS observations of equatorial ionospheric scintillation,” Proc. URSI
General Assembly & Sci. Sym., Beijing, China, August, 2014.
Morton, Y., Y. Jiao, F. van Graas, E. Vinande, and N. Pujara, “Analysis of receiver multifrequency
response to ionospheric scintillation in Ascension Island, Hong, and Singapore,”
Proc. ION Pacific PNT, Honolulu, HI, May 2015.