DCS Interference IssuesBackup Pilot Interference

The 00-06z Problem

Presented by

Microcom Design, Inc.

May 2012

Microcom Design, Inc. 2

Backup Pilot Interference

For several months, Microcom and NOAA have been observing and analyzing interference on the Backup Pilot. Backup Pilot uplinked from WBU at 401.700 MHz.

Erratic platform transmitting near the Backup Pilot. Originally at approximately 45 minutes after the hour; now

seems to have moved to around 47 minutes after the hour. Frequency not consistent; sometimes affecting Pilot, but not

always. Believe Platform ID has been captured and owner notified.

Continuous low-level carrier that drifts around the Pilot. Confirmed it is not related to uplink equipment. Frequency and Amplitude changing. Generally Interferer is strongest on GOES-West. On February 22, 2012: Microcom observed Interferer drift from

~ 40 Hz above Pilot to ~ 40 Hz below pilot in about 2 ½ hours – went right through Backup Pilot.

Microcom Design, Inc. 3

Backup Pilot Interferer – GOES-East

Impact on DCS Receiving Equipment: Worst is when interferer is within approximate 10 Hz of Pilot:

• Can cause Pilot Module to unlock.• If tracking Backup Pilot, will affect message reception.

From 10-100 Hz Pilot appears noisy. Outside 100 Hz little to no affect.

Microcom Design, Inc. 4

Backup Pilot Interferer – GOES-West

Believe the Interferer is in West. Stronger amplitude on GOES-West. Drift occurs when Western US/Eastern Pacific is in sunrise.

May be an errant DCP.

Microcom Design, Inc. 5

The 00-06z Story Begins …

In Mid-October 2011 NOAA was contacted by a representative of the Chilean Tsunami Warning Center: “During the last couples of months, GOES transmissions have

been very unstable. I can't detect any pattern but from time to time all our transmission suffer a lot of interruptions during the first 6 hours of the day.”

The user further noted that they did not believe the outage was related to their DCS receive system.

Microcom was also able to independently confirm the outages across the various receive sites (Wallops, NSOF & EDDN) .

Messages were either missing altogether or were garbled toward end of message.

In Early November 2011 NOAA authorized Microcom to investigate further …

Microcom Design, Inc. 6

Outage Graph from Chilean User

Microcom Design, Inc. 7

Chilean Tsunami Platform 1

Platform ADC22526 - Channel 215 @ 00:01:55 Every 00:05:00

0

2

4

6

8

10

12

14

16

18

20

Nov 4-12:00 Nov 5-00:00 Nov 5-12:00 Nov 6-00:00 Nov 6-12:00 Nov 7-00:00 Nov 7-12:00

UTC

Ph

ase

No

ise

0

5

10

15

20

25

30

35

40

45

50

Sig

nal

Str

eng

th

Signal Strength Phase Noise

Microcom Design, Inc. 8

Chilean Tsunami Platform 2

Platform 1401F372 - Channel 215 @ 00:01:50 Every 00:05:00

0

2

4

6

8

10

12

14

16

18

20

Nov 4-12:00 Nov 5-00:00 Nov 5-12:00 Nov 6-00:00 Nov 6-12:00 Nov 7-00:00 Nov 7-12:00

UTC

Ph

ase

No

ise

0

5

10

15

20

25

30

35

40

45

50

Sig

nal

Str

eng

th

Signal Strength Phase Noise

Microcom Design, Inc. 9

Hawaiian Tsunami Platform

Platform 354041A2 - Channel 215 @ 00:02:00 Every 00:05:00

0

2

4

6

8

10

12

14

16

18

20

Nov 4-12:00 Nov 5-00:00 Nov 5-12:00 Nov 6-00:00 Nov 6-12:00 Nov 7-00:00 Nov 7-12:00

UTC

Ph

ase

No

ise

0

5

10

15

20

25

30

35

40

45

50

Sig

nal

Str

eng

th

Signal Strength Phase Noise

Microcom Design, Inc. 10

Initial Conclusions – Platform Problem?

Daily pattern of signal strength and phase noise variations clearly evident from 00-06 UTC in platform ADC22526.

Problem platform bounded by two good platforms. Virtually identical results were seen on another set of

platforms. Signal strength and phase noise information was not

receive site specific. No indications of an interfering platform. Initial conclusion was that this was a platform issue. However, just before sharing this preliminary conclusion

with NOAA, Microcom received an e-mail from a Colombian user complaining of a similar problem … “For the past several weeks the DCP of ISAGEN and

SOPO(Bogota) have been missing transmissions in the hours between 20:00 – 23:00 local time 01:00-04:00 UTC.”

Microcom Design, Inc. 11

Colombian Platform

Platform 574096E0 - Channel 207 @ 00:55:40 Every 01:00:00

0

5

10

15

20

25

30

35

40

45

50

Oct 19-00:00 Oct 24-00:00 Oct 29-00:00 Nov 3-00:00 Nov 8-00:00 Nov 13-00:00

UTC

Sig

nal

Str

eng

th

0

1

2

3

4

5

6

7

8

9

10

Ph

ase

No

ise

Signal Strength Phase Noise

Microcom Design, Inc. 12

Wallops Parity Errors - First 10-Days of Nov

WCDA Hourly Parity Errs (10 Day)

0

100

200

300

400

500

600

11/01/11 11/02/11 11/03/11 11/04/11 11/05/11 11/06/11 11/07/11 11/08/11 11/09/11 11/10/11 11/11/11

Date

Me

ss

ag

es

GOES East GOES West

Microcom Design, Inc. 13

NSOF Parity Errors - First 10-Days of Nov

NSOF Hourly Parity Errs (10 Day)

0

100

200

300

400

500

600

11/01/11 11/02/11 11/03/11 11/04/11 11/05/11 11/06/11 11/07/11 11/08/11 11/09/11 11/10/11 11/11/11

Date

Me

ss

ag

es

GOES East GOES West

Microcom Design, Inc. 14

Wallops Parity Errors - 30-Days

WCDA Hourly Parity Errs (30 Day)

0

100

200

300

400

500

600

700

800

10

/12

/11

10

/13

/11

10

/14

/11

10

/15

/11

10

/16

/11

10

/17

/11

10

/18

/11

10

/19

/11

10

/20

/11

10

/21

/11

10

/22

/11

10

/23

/11

10

/24

/11

10

/25

/11

10

/26

/11

10

/27

/11

10

/28

/11

10

/29

/11

10

/30

/11

10

/31

/11

11

/01

/11

11

/02

/11

11

/03

/11

11

/04

/11

11

/05

/11

11

/06

/11

11

/07

/11

11

/08

/11

11

/09

/11

11

/10

/11

11

/11

/11

Date

Me

ss

ag

es

GOES East GOES West

Microcom Design, Inc. 15

Wallops Missing Messages - 30-Days

WCDA Hourly Missing Msgs (30 Day)

0

100

200

300

400

500

600

700

800

10

/11

/11

10

/12

/11

10

/13

/11

10

/14

/11

10

/15

/11

10

/16

/11

10

/17

/11

10

/18

/11

10

/19

/11

10

/20

/11

10

/21

/11

10

/22

/11

10

/23

/11

10

/24

/11

10

/25

/11

10

/26

/11

10

/27

/11

10

/28

/11

10

/29

/11

10

/30

/11

10

/31

/11

11

/01

/11

11

/02

/11

11

/03

/11

11

/04

/11

11

/05

/11

11

/06

/11

11

/07

/11

11

/08

/11

11

/09

/11

11

/10

/11

11

/11

/11

Date

Me

ss

ag

es

GOES East GOES West

Microcom Design, Inc. 16

Revised Conclusions – Systemic Problem!

Daily pattern of missed messages and messages with parity errors clearly evident from 00-06 UTC.

Issue was not receive site specific (WCDA vs. NSOF).

Certain channels appeared to be affected more so than others, but no real pattern.

Seemed to be affecting southern platforms more so than northern ones.

Dominant affect was on GOES-East (GOES-13). Could this be a spacecraft issue? GOES-13 was only operational GOES-N series satellite

at the time.

Microcom Design, Inc. 17

A Satellite Problem?

Satellite Comparison: GOES-11 (West):

• I-M Series• Not exhibiting the same phenomenon.

GOES-12 (South America): • I-M Series • Using Microcom’s DRGS and NSOF, Microcom took a look at the reception

of DCS messages from GOES-12 located at 60º.• While the results seemed to indicate a similar problem on GOES-12, the

results were inconclusive due to excessive satellite drift. GOES-13 (East):

• N-P Series GOES-15 (West):

• N-P Series• Came online in mid-December. • West messages still showed no significant impact.

Problem did not appear to be tied to the satellite series. Could still have been an isolated problem on GOES-13.

Needed to look deeper.

Microcom Design, Inc. 18

GOES DCS Signal Analyzer (GDSA)

Developed by Microcom specifically for NOAA/NESDIS. Deployed at WCDA in summer of 2010.

Functions: GOES DCS Message Reception and Capture Spectrum Analyzer (SA) Time Domain Signal Viewer IF Signal Capture

• On definable schedule.• Capturing of the raw IF DCS signal allows more detailed analysis.

Analysis can identify or eliminate: Potential interferers that may not be readily visible on SA. Excessive frequency drift. Phase and/or frequency transients. Amplitude variations.

Microcom Design, Inc. 19

Chilean Tsunami Platforms – A Deeper Look

Chilean Tsunami platforms did not show: Corruption from outside interferers. Phase or frequency issues.

Chilean Tsunami platforms did show Extreme amplitude fluctuations.

• Red Trace: Signal Power• Blue Trace: Signal-to-Noise Ratio (SNR)

Rapid signal level drops in excess of 10 dB.

Microcom Design, Inc. 20

Chilean Tsunami Platforms – More Examples

Microcom Design, Inc. 21

Brazilian Platforms – A Land Based Example

Concerned Tsunami Platforms may have been Buoy installations: Susceptible to wave affects and rough seas. Amplitude variations could be caused to uplink antenna motion. Later found out Chilean Platforms were Pier Tide Gauge installations.

Brazilian Platforms: Identified through DADDS database as experiencing similar problems. Known to be land-based deployments.

Similar if not worse amplitude variations.

Microcom Design, Inc. 22

Brazilian Platforms – Not the Pilot …

Message amplitude variations did not coincide with any Pilot variations.

Microcom Design, Inc. 23

Not the Spacecraft … Not the Platform …

Nothing in Satellite can … produce short time duration amplitude fluctuations. produce time of day affects. produce frequency selective effects (i.e. on isolated channels).

Nothing Common to Platform Different users – Chile, Colombia, and Brazil – albeit all in South America. Not a single transmitter manufacturer. Same platforms showed normal signal characteristics outside 00-06z.

Microcom Design, Inc. 24

Brazilian Platforms – The Smoking Gun

Signals received simultaneously from GOES-13 (top) and GOES-12 (bottom) conclusively show amplitude variation is travel path specific.

Microcom Design, Inc. 25

Summarizing What We Know

Phenomenon Is Not Specific to or produced by platform. Specific to or produced by satellite. Specific to or produced by receive site.

Phenomenon Is Time varying both on diurnal and sub-second scales. Travel path dependent. Primarily affecting platforms located in South America. Causing extreme amplitude fluctuations in DCS messages.

Varying Amplitude On short time scales can only be caused by coherent affects, e.g.

multipath interference. Early enough in transmission can prevent carrier lock resulting in

Missed Message. Later in transmission can cause loss of phase reference which will

most likely produced Parity Errors. Conclusion …

Microcom Design, Inc. 26

Ionospheric Scintillation

Ionosphere: Part of the upper atmosphere. Consists of multiple layers beginning at 85 km and ranging to 600 km;

scintillation affects occur around 350 km. Consists of electrons and charged atoms/molecules ionized by

ultraviolet radiation from the Sun. Creates refraction and diffraction of radio signals.

Scintillation: Refraction: Creates unexpected phase shifts. Diffraction: Creates amplitude and phase variations due to multipath

summing and cancellation.

Solar Cycle: Approximately every 11 years the Sun enters a period of increased

solar activity, known as the solar maximum. During this time the ultraviolet radiation increases, which increases

ionospheric scintillation. Next maximum is predicted to be May 2013.

Microcom Design, Inc. 27

Ionospheric Scintillation Impact

Affects frequencies from HF (3 MHz) to L-Band (2 GHz). DCS Uplink UHF (402 MHz) - DCS Downlink L-Band (1.694 GHz)

Typically begins after sunset and last several hours. Generally is most significant around the equinoxes. Will affect tropical latitudes the most; within ±20º of the magnetic

equator, which dips through South America.

Reproduced with permission from Reference 1

Microcom Design, Inc. 28

Why Now?

Approaching solar maximum (May 2013). Growth in DCS usage since last solar maximum (2000).

Especially growth in South America. Transition to HDR since last maximum.

100 bps modulation is not impervious to scintillation affects. However, ±60º bi-phase modulation has more phase margin

than 8-phase modulation. Also, HDR also led to more frequent transmissions.

Deployment of additional large receive sites. EDDN and NSOF have provided ability to do receive site

comparisons. Improvements in reception equipment (DAMS-NT) and

database software (DADDS). Beginning to see/address message issues on much smaller

percentage scale. Database provides historical research tool.

Microcom Design, Inc. 29

What Can Be Done in the Short Term?

Advise GOES DCS Users of the problem. GOES-East platforms in South America, southern Central

America, and the southeastern Caribbean will be most affected.

GOES West platforms located within 20o of the magnetic equator will be most affected.

Receive sites located in South America and in the equatorial anomaly region could experience time-of-day outages in the GOES DCS downlink.

Focus on Data Loss Mitigation Since there is no immediate solution to minimizing

transmission impact from ionospheric scintillation. Send prior data in each transmission.

• Repeat data in 2 or even 3 transmissions where possible.• Use Pseudo-Binary to reduce message length to allow for prior

data.

Microcom Design, Inc. 30

What Can Be Done in the Long Term?

Implementation of a Binary Message Format to better allow more redundant information in each transmission.

Reception from Multiple Paths? Receive East channels from GOES-West and vice versa. Not all locations have visibility to both satellites.

More Frequent/Redundant Transmissions? Resurrection of the Interleaver?

Since scintillation affects individual messages on a sub-second time scale, interleaving may help reduce message data loss.

Use of Linear Polarized Antenna? Since scintillation can affect the horizontal and vertical components of

a circularly polarized transmission differently, using linear polarized antenna may improve transmission throughput.

Using linear polarization will require 3 dB increase in uplink power. Wait it Out?

Ionospheric Scintillation will never completely go away, but it’s affects will gradually subside after the solar maximum.

However, it will return on the next solar cycle.

Microcom Design, Inc. 31

References

1. Paul M. Kintner, Jr., Todd E. Humphreys, and Joanna C Hinks, “GNSS and Ionospheric Scintillation How to Survive the Next Solar Maximum”, Inside GNSS, July/August 2009. www.insidegnss.com

2. Dave Anderson and Tim Fuller-Rowell, “The Ionosphere”, Space Environment Topics, NOAA, SE-14 1999, http://sec.noaa.gov.

3. Guozhu Li, Baiqi Ning, and Hong Yuan,“Analysis of ionospheric scintillation spectra and TEC in the Chinese low latitude region”, Earth Planets Space, 59, 279–285, 2007.

4. Stephen M. Hunt, Sigrid Close, Anthea J. Coster, Eric Stevens, Linda M. Schuett, and Anthony Vardaro, “Equatorial Atmospheric and Ionospheric Modeling at Kwajalein Missile Range”, Lincoln Laboratory Journal, Volume 12, Number 1, 2000.

5. Darrel Emerson, “Elliptical Polarization in the Ionosphere”, 1998 http://www.tuc.nrao.edu/~demerson/ionosphere/ionopol.html

END OF PRESENTATION “THANK YOU” FOR YOUR ATTENTION