Operational Electromagnetic Compatibility Study between NonGeostationary EarthExploration Satellite

1
Operational Electromagnetic Compatibility Study
between Non-Geostationary Earth-Exploration
Satellite Networks (Interference/Sharing
Analysis)

• Robert Bowen, Ali Shoamanesh
• (Telesat Canada), and
• Arvind Bastikar (CSA)
• SPACEOPS 2004
• May 17-21, Montreal

2
The Interference Environment of CSAs NGSO
Satellite Networks at 2 GHz and at 8 GHz

• CSAs NGSO satellite networks such as RADARSAT,
  SCISAT, and MOST operate in limited frequency
  ranges at both 8 GHz and at 2 GHz.
• The large number of such NGSO satellites
  worldwide prevents dedicated frequency bands for
  each operating satellite network.
• Because of this shared use of the available
  spectrum, inter-network interference is a
  necessary consideration in the design and
  operation of such networks.
• This paper discusses the interference environment
  of such satellite networks, and describes a
  method of evaluating the seriousness of that
  interference.

3
Differences in the Characteristics of
Interference in GSO and in NGSO Satellite Networks

• The interference between two GSO satellite
  networks is static or relatively
  time-invariant, because the relative position of
  the satellites is stationary.
• In contrast, the interference between two NGSO
  satellites, or between an NGSO and a GSO
  satellite network is a series of short
  high-interference bursts separated by long
  intervals of no appreciable interference.

4
Characteristics of the Interference at an NGSO
Earth Station or Space Station Receiver
5
Characteristics of Interference in NGSO Satellite
Networks (continued)

• Recognized characteristics of interference in
  NGSO include
• Maximum levels of interference
• Interval between interference bursts, and
• Probability that the interference exceeds a
  specific magnitude.
• Criteria of interference into EESS NGSO networks
  in ITU-R recommendations is based on this third
  criteria, the probability that the interference
  from another network exceeds a specified level a
  specific percentage of time.

6
Interference Between NGSO Downlink Networks

• The work that is described in detail in this
  paper is the interference between the RADARSAT
  and another satellite network at 8 GHz.
• ITU-R Recommendation SA.1026-3 specifies that the
  interference into such a network should not
  exceed -117 dBW per 100 MHz bandwidth more than
  0.025 of the time.

7
Ways to Determine and Meet the Conditions in
SA.1026-3

• Determination whether the limit of -117 dBW per
  100 MHz bandwidth more than 0.025 of the time
  limit is met is not easy.
• This determination may be necessary as part of
  the ITU coordination between two networks.
• Determination can be by either analysis or by
  simulation.
• Use of the Information obtained, to meet the
  required interference limit where necessary, can
  be either through either system design changes or
  by implementation of operational measures.

8
Method Described Here to Estimate the
Interference Between Two Networks

• The method described here to determine the
  interference probability is analytic, rather than
  a simulation tool.
• Result is in terms of the sum of a finite number
  of terms, determined by modeling the interference
  as a stochastic process.
• The resulting algorithm can be as accurate as is
  desired by increasing the number of terms in the
  finite sums involved.
• The algorithm has been computerized in an EXCEL
  program. This EXCEL program is being modified to
  be faster and more flexible.

9
The Cause of an Interference BurstBetween Two
Downlink Networks

• The interference into the earth station receiver
  of an 8 GHz satellite network exceeds the ITU
  limit only when both the desired transmitting
  satellite and the interfering satellite are
  simultaneously in the main beam or the near
  sidelobe of the antenna of the receiving earth
  station
• This condition occurs with very low probability,
  resulting in the interference limit being
  exceeded with a similarly very low probability.
• The task is to determine as accurately as
  possible the magnitude of that very low
  probability.

10
Diagram of the Interference Condition (Figure 10
of ITU-R Rec. S.1325-2)

11
Interference Estimation Process

• The basis for determining the probability that
  the interference exceeds a specified level is
  determination of the probability that the
  interfering satellite is in the main beam of the
  tracking interfered-with earth station.
• This has to be determined for every location of
  the interfered-with and the interfering
  satellites, in the region visible from the
  interfered-with earth station antenna beam.

12
Key to Determining the Probability of Significant
Interference

• The key to determining the probability of
  significant interference is determining the
  probability that a satellite is in a specified
  small segment of its orbital shell.
• That key is found in ITU-R Rec. S.1257. It
  specifies the probability that a satellite in a
  circular LEO orbit is in a small solid angle of A
  steradians.

13
Key (continued)

• Rec. S.1257 specifies that the probability that a
  satellite in a LEO circular orbit is in an area
  defined by its solid angle of A steradians is
• P (A/2p2)sin2(i) - sin2(L)-1/2 .. (1)
• where i is the latitude of the area of interest,
  and L is the maximum latitude of the satellites
  orbit.

14
Conversion Factors for Different Area Latitudes
and Inclination (Fig. 4, ITU-R Rec. S.1257-3)

15
Assumptions Made in Determining the Probability
of Significant Interference

• Assumption 1 The two satellites have
  asynchronous orbits, so the probability that
  Satellite 1 is in small area Ai is statistically
  independent of the probability that satellite 2
  is in the same area Ai.
• Assumption 2 Interference times are only
  counted when the interfered-with satellite is
  visible from its earth station.
• Assumption 3 Higher-order statistics such as
  clusters of interference bursts are not taken
  into account.

16
Interference Probability of Interest

• The probability of interest is P(I/K), given in
  Equation 2 of the written paper, is
• P (I/K) ? j P (Kj ) / ? i P (Ki ) P
  (Rj ) .. (2)
• where Kj is the event that the interfered-with
  satellite is in the small area Aj,
• K is the event that the satellite is in area K,
• P (Kj ) is the probability of event Kj,
• Rj is the area over which the interfering
  satellite causes significant interference when
  the interfered-with satellite is in the sub-area
  Aj, and
• P (Rj ) is the probability that the
  interfering satellite is in the sub-area Aj.

17
Calculation of P (I/K)

• The first step is to divide the area A that is
  visible from the interfered-with earth station
  into a large number of sub-areas Aj, j
  1,2,N. The Aj must be small enough that the
  function sin2(i) - sin2(L)-1/2 does not vary
  significantly over Aj. These Aj must be smaller
  at higher latitudes, and must also be such that
  the power-flux-density from the interfering
  satellite does not vary significantly over Aj.
  The Aj are determined using Equation 1.

18
Calculation of P (I/K) (continued)

• The Rj, j 1,2,N are then determined, taking
  into account the earth station antenna pattern,
  the EIRP of the satellite, and the distance from
  the satellite to the earth station. Note that Rj,
  like Aj, is measured in steradians.
• In the example calculations described below, the
  area A was divided into 202 sub-areas Aj, with Aj
  being smaller at higher latitudes, as shown in
  the following figure.

19
Diagram of the Area A and Sub-Areas Aj for One
Calculation
20
Example Calculation (continued)

• In the example calculation done using a
  rudimentary EXCEL computer program
• The latitude of the earth station was that of
  Fairbanks Alaska, a worst-case high-latitude
  location
• Both interfering and interfered-with satellites
  had an 81.4 maximum latitude and a 790 km
  altitude
• The transmitting satellite had the
  characteristics of RADARSAT-2A or RADARSAT-2B
• In the hypothetical worst-case example studied,
  both satellites had receiving earth stations in
  Fairbanks and
• The receiving earth stations had the
  characteristics of the RADARSAT earth station in
  Prince Albert.

21
Example Calculation Results

• Results of the calculations based on a 202
  sub-area quantization, the interference
  probability was
• 5.9 of the 0.025 ITU limit when the
  interfering satellite was RADARSAT-2A and
• 28.3 of the 0.025 ITU limit when the
  interfering satellite was RADARSAT-2B.

22
Example Calculation Results (continued)

• In an earlier simpler calculation with a smaller
  area-quantization, with only 80 sub-areas Aj, the
  results were
• 5.9 of the 0.025 ITU limit when the
  interfering satellite was RADARSAT-2A and
• 28.3 of the 0.025 ITU limit when the
  interfering satellite was RADARSAT-2B

23
Discussion of Results

• The above results indicate that
• 1. Choice of how the area visible from the
  interfered-with earth station is divided into
  Aj has a significant effect on the results
  obtained. This is because of the of the detail of
  the function sin2(i) - sin2(L)-1/2
• 2. The probability estimate for RADARSAT-2B is
  greater than that for RADARSAT-2A, simply because
  of its 3 dB higher EIRP, but neither satellite
  exceeds the ITU-R recommended limit when 10 meter
  earth stations are used and
• 3. The interference-analysis tool described here
  can be used to better understand the capabilities
  and limitations of frequency reuse of the 8 GHz
  EESS space-to-Earth allocation.

24
Extension to 2 GHz of the 8 GHz Analysis Tool
Described Here

• As described above, the analysis tool described
  here is useful in exploring the capabilities and
  limitations of the use of the 8 GHz EESS band.
• The tool can be readily extended to exploration
  of the capabilities and limitations of the use of
  the 2 GHz EESS bands 2025-2110 MHz and 2200-2290
  MHz.
• The major differences in extending the 8 GHz
  analysis tool to 2 GHz is the relevant ITU-R
  interference-probability limitations. At 2 GHz
  recommendations SA.5214-3 SA.1160-2 and
  SA.1163.2 apply rather than SA.SA.1026-3.

25
Operational Electromagnetic Compatibility Study
between Non-Geostationary Earth-Exploration
Satellite Networks (Interference/Sharing
Analysis)

• Robert Bowen, Ali Shoamanesh
• (Telesat Canada), and
• Arvind Bastikar (CSA)
• SPACEOPS 2004
• May 17-21, Montreal