Space Weather

Lab

GNSS Reflectometry for Studies of Surface Ice And

Water Conditions

Roohollah Parvizi, Boris Pervan and Seebany Datta-Barua.

Illinois Institute of Technology

November 7, 2018 1

Supported by: NASA award NNX15AV01G

Space Weather

LabOutline

• Motivation ‒ Why GNSS-R

• Objective‒ Detecting GNSS-R

• Background‒ Delay Doppler Map

• Method ‒ What are our sensors

‒ Data Campaigns

‒ Signal Processing

• Results

• Conclusion

November 7, 2018 2

Space Weather

LabMotivation

November 7, 2018 3

Icy road

[2]: https://www.news-journal.com/news/police/icy-roads-run-out-of-wreckers

[1] [2]

[1]: https://www.bikebandit.com/blog/black-ice-the-invisible-winter-threat

Space Weather

LabMotivation

November 7, 2018 4[3]: https://www.wsj.com/articles/laser-eyes

[4]: https://www.teslarati.com/tesla-lidar-sensors-spotted-testing-palo-alto/

[5]

[3]

[4]

[5]: https://www.detroitnews.com/story/business/autos

Space Weather

LabMotivation

LiDAR disadvantages:

▪ Expensive, costly.

▪ Unreliable for water surface and breaking waves.

▪ It is affected by rain.

▪ Low operating range (500-2000m).

November 7, 2018 5

Space Weather

LabMotivation

Can we do better with GNSS?

▪ Lower cost.

▪Widespread coverage and availability.

▪Unaffected by precipitation and cloud cover.

▪Potentially better resolution.

November 7, 2018 6

Space Weather

LabGNSS-Reflectometry

November 7, 2018

GNSS-R:

• The study of the

characteristics of

a reflected signal

7

Space Weather

LabObjective

November 7, 2018 8

Satellite

𝑃𝑅𝑁𝑖

Satellite

𝑃𝑅𝑁𝑗

Power and Electronics

Verification sensors

• Detect reflected GNSS signals from the water/ice surface.

• Verify GNSS-R.

Space Weather

LabGNSS Reflection Signal

November 7, 2018

SP

𝛼𝑟 𝛽𝑖

Glistening Zone

Sensor

suite Satellite

𝑃𝑅𝑁𝑖

𝛼𝑟 = 𝛽𝑖

Specular Point(SP) :The point on

the surface where the incident

and reflected angles are equal

9

Verification sensors

Space Weather

LabGlistening Zone

November 7, 2018 10

Receiver

Glistening Zone

Receiver

Glistening Zone

Receiver

Space Weather

LabDelay Doppler Domain

November 7, 2018

Specular Point(SP)

Sensor

suite

Satellite

𝑃𝑅𝑁𝑖

East

North

11M. Martin-Neirea, A Passive Reflectometry and Interferometry System.

Iso-Range line

Iso-Doppler line

Space Weather

Lab

Spatial Domain andDelay Doppler domain

November 7, 2018 12

C/A

code

chips

delay

Doppler

frequency

y

x

SP

Iso-Range line

Iso-Doppler line

delay

Scattered Power

Space Weather

LabSensor Suite

November 7, 2018 13

Reflect

antenna

Direct

antenna

Universal software

radio peripheral

(USRP)

Space Weather

LabSensor Suite

November 7, 2018 14

Reflect

antenna

LiDAR

Camera

Weather

Station

Space Weather

LabTest Locations

November 7, 2018

Test 1,3 Locations

IIT

15

Test 4, 5

Locations

Test 2 Location

Space Weather

LabTest 2 Location

November 7, 2018 16

𝑆𝑒𝑛𝑠𝑜𝑟𝑑𝑖𝑟𝑒𝑐𝑡𝑖𝑜𝑛

𝑁𝑜𝑟𝑡ℎ

𝜙 = 2080

Space Weather

LabTest 5 Location

November 7, 2018 17

𝜙 = 600

𝑆𝑒𝑛𝑠𝑜𝑟𝑑𝑖𝑟𝑒𝑐𝑡𝑖𝑜𝑛

𝑁𝑜𝑟𝑡ℎ

Space Weather

LabData Campaigns Info

November 7, 2018 18

Data campaign info Test 02 Test 05

Date Tue Jan 23 14:35:48 2018 Thu Jul 05 12:13:28 2018

Sensor Location

Latitude: 41.8388924° 𝑁 Latitude: 41.84066 ° 𝑁

Longitude: 87.603140° W Longitude: 87.60698° W

Height:3 𝑚 Height: 3 𝑚

Sample rate 5 𝑀𝐻𝑧 5 𝑀𝐻𝑧

USRP_dir RF gain 31 𝑑𝐵 31 𝑑𝐵

USRP_ref RF gain 31 𝑑𝐵 31 𝑑𝐵

USRP_dir inline gain 0 𝑑𝐵 30 𝑑𝐵

USRP_ref inline gain 30 𝑑𝐵 40 𝑑𝐵

Look direction(azimuth angle) 2080 (southwest) 60𝑂 (northeast)

Condition Ice Water

Space Weather

LabPost-Processing

November 7, 2018 19

• Verification:

– Evaluate surface condition with weather station.

– Determine where SPs are.

– Compare SPs with LiDAR.

• Detection:

– DDM for SP on the water/ice.

Space Weather

LabPost-Processing

November 7, 2018 20

Data Field

campaign

Approximate sensor location

Almanac info

LiDAR range intensity

Compute SPENU SP-LiDAR

Map

Select SV whose DDM is

desiredDDM

Verification

Detection

Space Weather

Lab

November 7, 2018 21

Sky Plot and SP-LiDAR, Test 2 (ice)

Space Weather

LabSky Plot and SP, Test 5 (water)

November 7, 2018 22

(m)(m

)

Space Weather

LabSP-LiDAR Map

November 7, 2018 23

Θ = 450 LiDAR

Parallel to the dock

Perpendicular to the

dock

LiDAR direction

Perpendicular to the

LiDAR direction

LiDAR Field of vision

(± 15° Vertical FOV)

“PRN 28 is reflecting off the water”

(m)

(m)

Space Weather

Lab

Incoherent DDM for PRN 28 Test 5 (water),Method 1

November 7, 2018 24

Reflected Signal Direct Signal

[7]:R.Parvizi, H. S. Zadeh, L. Pan, B. Pervan, S. Datta-Barua, ” Multi-sensor Study of Lake Michigan’s Surface using GNSS-Reflectometry”, Proceedings of the 29th International Technical Meeting of The Satellite Division of the Institute

of Navigation (ION GNSS+ 2018), Miami, Florida, September 2018

Space Weather

LabDDM-Processing

Problem:

‐Weak and noisy signal: direct signal not acquired by SDR, reflected signal band unexplained.

Solution:

• Use auto-correlation peak to test the direct signal for acquisition.

‐Method 2: Incoherent integration.

‐20 ms, 100 ms, 500 ms and 1s

‐Method 3: Coherent integration

‐Coherently 10 ms

‐Method 4: Differentially coherent integration

• Generate reflected DDMs for the Method (4) that successfully acquires direct

signals.

November 7, 2018 25

Space Weather

LabMethod 2: Incoherent DDM Processing

November 7, 2018 26

Incoming signal

𝑒−𝑗𝑤𝑡

900

𝐼

𝑄

Correlator

Correlator

Coherent

integration

Coherent

integration

2

2

Incoherent

integration

Space Weather

LabMethod 3: Coherent DDM Processing

November 7, 2018 27

20 msOf

data

Second10msOf

data

First 10msOf

data

Results for D1 = 𝑃1

Results for D2= 𝑃2

D1

D2

Pick data set that has the

maximum power.

Coherent integration

Coherent integration

Space Weather

LabMethod 4: Coherent differential integration

November 7, 2018M. H. Zarrabizadeh and E. S. Sousa, “A Differentially Coherent PN Code Acquisition

Receiver for CDMA Systems”, IEEE TRANSACTIONS ON COMMUNICATIONS, VOL. 45, NO. 11, NOVEMBER 199728

Incoming signal

𝑒−𝑗𝑤𝑡

900

𝐼

𝑄

Correlator

Correlator

Coherent

integration

Coherent

integration

2

2

𝑌𝑘

𝑌𝑘−1∗

𝑍−1

. 2

𝑍𝑘

Space Weather

LabEvaluate Methods

for Direct signal at high and low elevations, Test 5

November 7, 2018 29

SKY plot

based on almanac infoSKY plot

based on real GPS data

Space Weather

Lab

Autocorrelation of Direct Signal, PRN 28

November 7, 2018 30

Incoherent integration,

method 2

Coherent integration,

method 3

Result: Methods 2 and 3 each result in acquiring the high elevation

direct signal.

Space Weather

Lab

Low elevation satellite, PRN 22, Direct signal

November 7, 2018 31

Incoherent integration,

method 2

Coherent integration,

method 3

Result: Methods 3 does NOT result in acquiring the low elevation

direct signal.

Space Weather

Lab

Low elevation satellite, PRN 22, Direct signal

November 7, 2018 32

Coherent differential integration,

method 4Result: Method 4

results in acquiring

the low elevation

direct signal and high

elevation direct signal

(not shown).

Next: Use Method 4

to generate DDM for

the reflected signal.

Space Weather

Lab

Method 4: Coherent differential DDM for PRN 28 Test 5 (water)

November 7, 2018 33

Reflected Signal Direct SignalAcquired in

C/A and

Doppler.Band in C/A

space partly

mitigated.

Space Weather

Lab

Method 4: Coherent differential DDM for PRN 22 Test 2 (ice)

November 7, 2018 34

Reflected Signal Direct Signal

Space Weather

LabConclusion

• Coherent, non-coherent, and coherent differential methods were studied for both direct and reflected GNSS signals.• Coherent differential method did a good job acquiring the low elevation satellite direct

signal.

• We generated coherent differential DDMs for the reflected water/ice surface.

• Future work:• Further analysis of coherent differential DDMs.

• Synchronization of multiple sensors.

• Combination of signal processing methods.

• Using accurate clock for USRPs, such as GPSDO.

• Additional data campaigns throughout 2018!

November 7, 2018 35

Space Weather

LabAcknowledgment

November 7, 2018 36

Houshine Sabbagh Zadeh,

Li Pan,

Yang Su and

Ningchao Wang

for their advice and

technical support.

November 7, 2018 37

Thank YOU.

Extra Slides

November 7, 2018 38

Space Weather

Lab

Method 4: Coherent differential DDM for PRN 28 Test 5 (water), K=2

November 7, 2018 39

Reflected Signal Direct SignalAcquired in C/A and Doppler.However, reflected

peak chip and Doppler change with k.

Space Weather

LabMethod 2 DDM for PRN 28 Test 5 (water)

November 7, 2018 40

Reflected Signal,500 ms Reflected Signal,1000 ms

Result: Reflected signal assumed comparable in noise to low-elevation direct signal. Methods 2 and 3 do NOT result

in a correct DDM (i.e., peak is not at the specular point).

Space Weather

Lab

Method 3 Coherent DDM for PRN 28 Test 5 (water), 10 ms

November 7, 2018 41

Reflected Signal Direct Signal

Space Weather

Lab

Method 4: Coherent differential DDM for PRN 28Test 5 (water)

November 7, 2018 42

Reflected Signal Direct Signal

Result: Method 3 results in a DDM at the specular point (C/A chip 867). However, band in C/A still exists in the reflected DDM only.

Space Weather

Lab

Low elevation satellite, PRN 3,Direct signal

November 7, 2018 43

Coherent

autocorrelation, 1ms

Coherent

autocorrelation, 10ms

Space Weather

LabLow elevation satellite, PRN 3

November 7, 2018 44

Coherent differential

Autocorrelation , 1ms , K=1

Coherent differential

Autocorrelation , 1ms , K=2

Space Weather

Lab

Incoherent DDM Processing, previous work: “Method 1”

November 7, 2018

20 msOf

data

Second10msOf

data

First 10msOf

data

Coherent integration

Results from 1st ms

10th

ms

1st

ms

Coherent integration

Results from 10th ms

Coherent integration

Results from 1st ms

10th

ms

1st

ms

Coherent integration

Results from 10th ms

Results from1st ms

Results from 2nd ms

Results from 10th ms

𝑓1

45

D1

D2

𝑃𝑖𝑛 =

𝑖=1

10

𝑃𝑖