STATISTICAL STUDY OF SPACE REMOTE SENSORS
Z. Ghadyani a, , S. Afshar Naseri b, S. Adham khiabani c
a
University of Tehran, Faculty of Science, Department of Physics, ghadiany@khayam.ut.ac.ir
b
IEEE member, Siamak_Afshar@ieee.org
c
K. N. Toosi University of Technology, Faculty of Geodesy and Geomatics, Department of Remote Sensing;
Adham@ieee.org
WG I/4 - Airborne Digital Photogrammetric Sensor Systems
KEY WORDS: remote sensor, payload, statistical study, multispectral imagers, spectrometers, radiometers.
ABSTRACT:
Since the beginning of the space remote sensing, many instruments have been invented. The number of remote sensors and payloads
are growing and becoming various increasingly. A mission designer usually has many options and different combinations of remote
sensors to achieve their objectives. However there is a lack of elaborated investigation and statistical study of remote sensors. Thus it
motivated us to make an extensive survey of remote sensors that are most used in earth observation and classify them in main
categories. Then we made a descriptive statistical study that showed some useful relationship between some parameters. There are
some good reasons and interpretations behind these relations that are investigated in this paper. It is expected this paper can help
mission designers to make the best decision in selecting the appropriate sensors and also to predict their requirements in the early
stages of the conceptual design.
their mass [3]. Table 2 shows one of the most popular ways for
classifying satellites with respect to their mass.
1. INTRODUCTION
Statistical study plays an important role in early design stages.
Payload sizing is indeed one of the major tasks in mission
design that makes analogy with other existing remote sensors to
estimate their own payload parameters. Thus a table of remote
sensors, a purposeful classification and a statistical study of
them is always the main requirement for mission designers.
Satellite
Classification
Large Satellites
Medium Satellites
Mini Satellites
Micro Satellites
Nano Satellites
Pico Satellites
Femto satellite
To carry out the statistical study, a table of sample data is
prepared which is mainly obtained from references [1] and [2].
Due to shortage of space, we have included all large tables at
the end of the paper in appendix section. In order to validate the
nature of the selected sample data, general features of the
satellites and their sensors are studied in section 2. Then three
most used sensors -that is multispectral imagers, radiometers
and spectrometers- are investigated statistically in more detail
in sections 3, 4 and 5.
above 1000kg
500 to 1000kg
100 to 500kg
10 to 100kg
1 to 10kg
0.1 to 1kg
Below 100g
Small Satellites
Table 2: satellite classification based on their mass [3]
Most of the satellites in table 1 are weighting from 1000 to
2000 which means they are mainly large satellites in aforesaid
classification.
2. SAMPLE DATA
The last general feature of this sample data that may worth to
consider is the orbit height namely Perigee and Apogee which
predominantly differ from 700 to 900km.
The list of satellites used in this study can be seen in
table1 1 .This sample data contains just 50 remote sensing
satellites out of around 600 total remote sensing satellites.
2.2 Payload Features
2.1 General Features
Payloads of satellites in table1 have been extracted and
classified in main categories and the population of each
category has been computed. Figure1 shows ratio of each
category in the study. As can be seen from this figure the
maximum population belongs to multispectral imagers (22%),
spectrometers (11%) and visible IR radiometers (10%).
Therefore the special features of these most popular payloads
will be considered statistically in the following sections.
Before going through any detail study let us consider general
characteristics of sample data that may affect the final results.
The lunch years of satellites in table1 extend between 1975 and
2005 but most of them have lunched during 1995 to 2005. One
of major ways to classify the satellites is categorizing them by
1
Wet Mass
Due to shortage of space, we have included large tables at the
end of the paper.
567
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B1. Beijing 2008
Unfortunately there was lack of information for the imager size
while collecting the sample data. Nonetheless the mean value of
the size of imagers have been obtained to be 0.96m by
computing the geometric mean value of height, width and
length of each imager and then computing the arithmetic mean
value of them.
Eventually let us consider the last parameter of table3 that is
data rate the plot of which is illustrated in figure3. Based on this
plot the most probable value of data rate is 4Mbit/sec and the
mean value is 22.7Mbit/sec.
Data Rate Histogram
6
Frequency
5
4
3
2
1
0
Figure 1: Ratio of each category of payloads
0.5
3. MULTISPECTRAL IMAGERS
1
2
4
8
16
32
Data Rate Mbit/s
64
128 More
Figure 3: Data rate histogram of multispectral imagers
Multispectral instruments acquire their images from the earth in
a few specific regions of the electromagnetic spectrum which
are called bands.[4] The increased spatial resolution and the
addition of a spectral dimension allows for the discrimination of
materials with this type of imagery. Although multispectral
instruments can discriminate materials, a hyperspectral imager
or spectrometer is required to actually identify materials. [4]
The most significant parameters of these instruments are spatial
resolution, size, weight, power and cost, as miniaturizing and
introduction of new technologies have been of great interest
over last century.
To analyze quality factors and parameters of imagers consider
table4 in which the number of bands, resolution and swath
width of each of mentioned imagers have been given.
As it is evidence at the first glance, most of the imagers work in
visible and NIR region of spectrum. Plot in figure4 is resolution
histogram for visible, NIR, SWIR and TIR. Noticing in diagram,
it can be seen that the most probable resolutions are 50m, 50m,
100m and 1km for Visible, NIR, SWIR and TIR respectively.
Table 3 shows a list of multispectral imagers belong to satellites
listed in table1, with their mass, size, power and data rate.
As can be infer from the table, the values of mass and power for
each imager are at the same range. This sample data covers
light imagers from 0.14kg to heavy imagers of 270 kg with
powers vary between 1.4W and 345W. Figure2 displays a
histogram of mass and power. Considering this plot the most
probable mass and power are 50kg and 50W respectively with
the mean value of 91kg and 118W.
Figure 4: Resolution histogram of multispectral imagers
4. SPECTROMETERS
In remote sensing “spectrometry” or “spectroscopy” refers to
the detection and measurement of radiation spectra of a target
(area or volume) in many bands of the medium (generally the
atmosphere).[1] In general spectroscopy is the science of
measuring the spectral distribution of photon energies (as
wavelengths or frequencies) associated with radiation that may
be transmitted, reflected, emitted, or absorbed upon passing
from one medium (vacuum or air) to another material objects.
Figure 2: Power and mass histogram of multispectral imagers
568
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B1. Beijing 2008
The term radiometry is also often used to include the detection
of the quantity, quality, and effects of such radiation.
In literature, the terms “imaging spectroscopy”, “imaging
spectrometry” and “hyperspectral imaging” are often used
interchangeably in remote sensing. Even though semantic
differences might exist, a common definition is: simultaneous
acquisition of spatially coregistered images, in many, spectrally
contiguous bands, measured in calibrated radiance units, from
a remotely operated platform.[1]
Radiometer is an instrument that quantitatively measures the
intensity of electromagnetic radiation in some bands within the
spectrum. Usually, a radiometer is further identified by the
portion of the spectrum it covers; for example, visible, infrared,
or microwave. [1]
There are usually three “image capture” technologies of
imaging spectrometers in use: Whiskbroom line array,
Pushbroom area array and Framing camera.
Looking at figure1, it can be seen radiometers particularly in
Visible and IR region are the most common payload used in
space missions after spectrometers. Table 6 shows a list of
radiometers belong to satellites listed in table1, with their mass,
size, power and data rate. As can be infer from the table, like
multispectral imagers and spectrometers, the values of mass and
power for each radiometer are at the same range. This sample
data covers light radiometers of 33kg to heavy radiometers of
360kg with powers vary between 27W and 315W. Figure7
displays a histogram of mass and power. Considering this plot
the most probable mass and power are 100kg and 100W and
their mean values are 143kg and 93W respectively.
As mentioned in last sections spectrometers are the most used
payloads in space missions after multispectral imagers. Table 5
shows a list of spectrometers belong to satellites listed in table1,
with their mass, power and data rate. Just like multispectral
imagers, the values of mass and power for each spectrometer
are at the same range. Figure 6 displays a histogram of mass
and power in which the most probable mass and power are
200kg and 200W and the mean values of them are 140kg and
125W respectively.
Figure 5: Mass and Power histogram of spectrometers
The histogram curve of data rate is illustrated in figure 6 in
which it is evident that the most frequent data rate is 1000kbit/s.
Figure 7: Power and mass histogram of radiometers imagers
To analyze quality factors and parameters of imagers consider
table7 in which the number of bands, resolution and swath
width of each of mentioned imagers have been given.
As it is evident at the first glance, most of the imagers work in
visible and NIR region of spectrum. Investigation shows most
of radiometers have used 8 bands in their instrument. Plot in
figure 8 is resolution histogram for visible, NIR, SWIR and
TIR. Noticing in diagram, you can see the most probable
resolution is 1500m.
Figure 6: Data rate histogram of spectrometers
5. RADIOMETERS
Radiometry is the science of radiation measurement in any
portion of the electromagnetic spectrum, i.e. the study of
creation, transport, and absorption of electromagnetic energy,
and the wavelength-dependent properties of these processes.
Figure 8: Resolution histogram of radiometers imagers
569
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B1. Beijing 2008
REFERENCE
3- Website (http://centaur.sstl.co.uk/SSHP) last accessed:
May2008
1-H.G.Kramer,2002, Observation of the Earth and Its
Environment,Springer.
2- Website
May2008
(http://directory.eoportal.org/)
last
4- Website: (http://www.es.ucsc.edu/%7Ehyperwww/chevron)
last accessed: May2008
accessed:
APENDIX
NO
year
Satellite
Country
Perigee
(km)
Apogee
(km)
Dry
Mass(kg)
Payload
Mass(kg)
Satellite
Power(W)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
1996
2002
1991
2003
1994
2002
1991
1988
1999
2000
2007
1987
1994
1995
2002
1982
1992
2003
1996
1999
2005
2003
1991
1992
1978
1982
1999
1997
1975
1985
2001
2002
2002
1995
ADEOS1
ADEOS2
Almaz1
CBERS2
Electro
Envisat 1
ERS 1
FY-1B
FY-1D
FY-2B
FY-3
GEOS 7
GEOS8
GMS-5
HY-1
Insat 1A
Insat 2A
Insat 3A
IRS P3
IRS P4
IRS P5
IRS P6
IRS1B
JERS
Landsat3
Landsat4
Landsat7
Lewis
Meteor2
Meteor3
Meteor3M
Meteosat-8
Metsat-1
MicroLab
Microlabsat
MOS
NOAA11
NOAA15
Okean-E
Okean-O
Okean-O1_N3
OrbView2
OrbView3
RADARSAT1
Resurs-0 N4
Resurs-F3
Sich-1M
SPOT-1/2
SPOT-3
Japan
Japan
Russia
China & Brazil
Russia
Europe
Europe
China
China
China
China
US
US
Japan
China
India
India
India
India
India
India
India
India
Japan
US
US
US
US
Russia
Russia
Russia
Europe
India
US
Japan
Japan
US
US
USSR/Ukrainian
Ukraine
Ukraine
US
US
804.6
802.9
270
778
36000
800
782
900
863
36000
836.4
36000
36000
36000
795
36000
36000
36000
819
720
618
817
867
568
901
705
705
523
950
1200
925
36000
36000
733
800
909
848
833
500
670
630
705
470
789
802.9
380
778
36000
800
785
900
863
36000
836.4
36000
36000
36000
795
36000
36000
36000
821
720
618
817
913
568
920
705
705
523
950
1200
925
36000
36000
749
800
909
865
833
660
670
660
705
470
3560
3500
18550
1450
2580
8140
2384
750
954
600
2200
399
1140
746
365
650
911
1348
922
1036
1560
1360
975
1340
891
1941
2020
385.6
1300
2215
2500
2040
1055
68
54
738
1700
1454
1950
6500
1950
309
304
1300
1200
3420
4500
5300
2400
1100
1500
6500
2650
800
250
280
2480
450
1057
300
450
Canadian
798
798
1540
Russia
Russia
Ukraine
France
France
663
260
285
832
832
691
275
650
832
832
1950
6300
2223
1907
1907
44
45
46
47
48
49
1987
1988
1998
1980
1999
1988
1997
2003
1995
1998
1999
2004
1986
1993
570
900
2150
888.2
247
87
497
58
303
288
500
700
900
1180
3100
873
800
1100
1250
700
2000
515
990
1550
740
500
500
1000
600
550
55
149
386
100
2000
500
50
1500
5000
1100
3400
550
500
790
1000
1000
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B1. Beijing 2008
NO
year
Satellite
Country
Perigee
(km)
Apogee
(km)
Dry
Mass(kg)
50
51
1998
2002
SPOT-4
SPOT-5
France
France
820
822
820
822
2550
3030
Payload
Mass(kg)
Satellite
Power(W)
2200
2400
Table 1:List of Satellites used in the study [1] [2]
Payload
AWiFs
CCD Camera
CMR
COCTS
CZI
ISR
LISS 3
LISS 4
LISS I
LISS I
LISS II A/B
LISS II M
LISS III
MSS
MSS
MSU-E
MSU-K
MSU-V
SEVIRI
TM
Visible/IR TV
VMI
VMI
WFI
WFI
WiFs
WiFs
Sat
IRS P6
Insat 2E/3A
Microlabsat
HY-1
HY-1
CBERS-3,4
IRS P6
IRS P6
IRS 1A/1B
IRS 1E
IRS 1A/1B
IRS P2
IRS 1C/1D
Landsat1,2,3
Landsat4,5
Resurs-0 N3&N4
Okean-O1_N3
Okean-O
Meteosat-8
Landsat4,5
Electro
SPOT-4
SPOT-5
CBERS-1/2
CBERS-3,4
IRS 1C/1D
IRS P3
Mass
(kg)
103.6
55
0.14
50
15
106.1
169.5
38.5
38.5
80.5
80.8
171
64
58
60
6.5
Size
L
0.6
W
H
0.5
Power(W)
Data Rate
(kbit/s)
114
50
1.4
41.7
30
0.3
52500
2670
26000
52500
105000
5200
5200
10400
70
216
34
34
34
343
85
50
81
1.27
0.58
0.53
270
245
2.1
2
2.1
1.1
1
0.7
123
345
152
152
1
1
1
1
0.7
0.7
200
200
41
41
35790
15000
15000
5120
3000
84900
2560
520
520
1100
50000
2060
2060
22
22
Table 3: Technical parameters of multispectral imagers in the study[1] [2]
Payload
AWiFs
CCD Camera
COCTS
CZI
IRS
LISS 3
LISS 3
LISS 4
LISS I
LISS II A/B
LISS II M
LISS III
MR-2000M1
MSS
MSS
MSU-A
MSU-E
MSU-E
MSU-M
MSU-M
MSU-M
MSU-M
Sat
IRS P6
Insat 2E/3A
HY-1
HY-1
CBERS-1/2
IRS-1C
IRS P6
IRS P6
IRS 1E
IRS 1A/1B
IRS P2
IRS 1C/1D
Meteor3M
Landsat1,2,3
Landsat4,5
Okean-O
Resurs-0 N3&N4
Meteor3M
Okean-O
Okean-O1_N3
Sich-1M
Sich-1M
Resolution(m)
Band
4
3
10
4
4
4
4
3
4
4
4
4
1
4
5
3
3
3
4
4
4
4
visible
NIR
SWIR
56
1000
1100
250
40
23.5
23.5
5.8
72.5
36.25
32
23.5
700
80
80
157
45
38
1500
1000
1500
1500
56
1000
56
1000
1100
571
250
40
23.5
23.5
5.8
72.5
36.25
40
70
23.5
23.5
70.5
80
80
30
45
38
1500
1000
1500
1500
80
80
Swath
TIR
1100
80
Width(km)
740
300
1400
500
120
148
140
140
148
74
66
142
3100
185
186
300
45
76
1930
1900
2000
2000
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B1. Beijing 2008
Payload
Sat
MSU-S
MSU-S
MSU-SK
MSU-SK
MSU-V
TM
VMI
VMI
WFI
WFI
WiFs
WiFs
Resolution(m)
Band
Meteor-Priroda
Okean-O1_N3
Okean-O
Okean-O
Okean-O
Landsat4,5
SPOT-5
SPOT-5
CBERS-1/2
CBERS-3,4
IRS-1C
IRS P3
visible
NIR
140
250
157
157
50
30
1116
1116
260
73
188
188
240
250
157
157
50
30
1116
1116
260
73
188
188
2
2
5
5
8
7
4
4
2
4
2
3
SWIR
100
30
Swath
TIR
Width(km)
590
590
250
120
1380
1100
600
600
1380
185
885
866
804
804
188
Table 4: Image quality parameters of multispectral imagers in the study [1] [2]
Payload Name
Satellite
Mass(kg)
GLI
GOME
GOMOS
HIS
IMG
JEA(ILAS2)
MERIS
MIPAS
MOS A
TOMS
XRS
ADEOS2
ERS 2
Envisat 1
Lewis
ADEOS1
ADEOS2
Envisat 1
Envisat 1
IRS P3
Meteor3
Clark
450
55
163
23
150
130
200
320
8.5
35
5.2
Power(W)
Data Rate (kbit/s)
400
32
146
50
150
80
175
195
16000
40
222
327000
600
500
24000
533
1300
25
5
Table 5: Technical parameters of spectrometers in the study[1] [2]
Payload
Sat
Mass(kg)
Data Rate
(kbit/s)
Power(W)
AVHRR
NOAA15
33
27
2000
AVNIR
ADEOS1
230
300
60000
MVISR
FY-1C/1D
95
45
OCM
IRS P4
75
OCTS
ADEOS1
360
315
3000
OPS
JERS
174
250
60000
PLODER
ADEOS1
33
40
900
665.4
20800
Table 6: Technical parameters of radiometers in the study [1] [2]
Payload
AVHRR
AVNIR
MVISR
OCM
OCTS
OPS
PLODER
S-VISSR
VHRR
VIRR
VISSR
Sat
NOAA15
ADEOS1
FY-1C/1D
IRS P4
ADEOS1
JERS
ADEOS1
FY-2C/2D
INSAT-2
FY-3
GMS-5
Resolution(m)
Band
6
5
5
8
12
8
15
10
4
visible
SWIR
TIR
1100
16
1100
236
700
18
6
1250
2500
1100
1250
1100
1.1 km
1100
236
701
5000
702
5000
11000
5000
Table 7: Image quality parameters of radiometers in the study [1] [2]
572