0

Introduction to Synthetic Aperture Sonar

Roy Edgar Hansen
Norwegian Defence Research Establishment
Norway
1. Introduction
SONAR is an acronym for SOund Navigation And Ranging. The basic principle of sonar is
to use sound to detect or locate objects, typically in the ocean. Sonar technology is similar to
other technologies such as: RADAR = RAdio Detection And Ranging; ultrasound, which
typically is used with higher frequencies in medical applications; and seismic processing,
which typically uses lower frequencies in the sediments. There are many good books that
cover the topic of sonar (Burdic, 1984; Lurton, 2010; Urick, 1983). There are also a large number
of books that cover the theory of underwater acoustics more thoroughly (Brekhovskikh &
Lysanov, 1982; Medwin & Clay, 1998).
The principle of Synthetic aperture sonar (SAS) is to combine successive pings coherently
along a known track in order to increase the azimuth (along-track) resolution. A typical data
collection geometry is illustrated in Fig. 1. SAS has the potential to produce high resolution
images down to centimeter resolution up to hundreds of meters range. This makes SAS a
suitable technique for imaging of the seaoor for applications such as search for small objects,
imaging of wrecks, underwater archaeology and pipeline inspection.
SAS has a very close resemblance with synthetic aperture radar (SAR). While SAS technology
is maturing fast, it is still relatively new compared to SAR. There is a large amount of SAR
literature (Carrara et al., 1995; Cumming & Wong, 2005; Curlander & McDonough, 1991;
Franceschetti & Lanari, 1999; Jakowatz et al., 1996; Massonnet & Souyris, 2008).
This chapter gives an updated introduction to SAS. The intended reader is familiar with
sonar but not SAS. The only difference between traditional sonar and synthetic aperture is
Fig. 1. Data acquisition geometry for synthetic aperture sonar.
1
www.intechopen.com
2 Will-be-set-by-IN-TECH
Fig. 2. Basic imaging geometry.
the construction of the aperture (or array). We start by giving a review of sonar imaging and
describe the backprojection method. Then, we derive the angular resolution. We describe the
multi-element receiver concept and calculate the area coverage rate for SAS. We explain the
frequency dependence in SAS, and discuss some of the choices and trade-offs in SAS design.
We list some of the specic challenges in SAS, and suggest possible solutions. We give an
overview of the signal processing involved in a SAS processor, and we discuss properties of
SAS images. Finally, we show numerous examples of SAS images from a particular system,
the HISAS 1030 interferometric SAS.
2. Sonar Imaging
Assume a transmitter that insonies a scene with acoustic reecting material represented by
a reectivity function (x, y). A part of the scattered acoustic eld is recorded by one or more
receiver hydrophones. This is illustrated in Fig. 2. Sonar imaging is the inverse problem, namely
to estimate the reectivity function from the received data from one or more transmitted
pulses.
Sonar imaging can be separated into range processing of the data and angular processing
(beamforming) of the data. Beamforming is dened as the method or processing algorithmthat
focuses the signal fromseveral receivers in a particular direction. Beamforming can be applied
to all types of multi-receiver sonars: active, passive, towed array, bistatic, multistatic, and
synthetic aperture. Sonar beamforming is well covered in (Burdic, 1984; Johnson & Dudgeon,
1993; Nielsen, 1991).
Range processing is signal processing applied to each time series individually. The processing
is a function of transmit waveform. There are several types of signals that are used as transmit
waveforms. Classical sonar often uses gated continuous-wave (CW) pulses, sometimes
referred to as pings. Modern sonar, and virtually all SAS systems, use phase coded transmit
signals where the signal bandwidth is determined by the phase coding (or frequency spread of
the signal modulation). The reason to use phase coded waveforms is to increase the transmit
signal energy while maintaining large signal bandwidth. The range processing done on phase
coded signals are referred to as pulse compression or matched ltering. This is covered in a
number of excellent books such as (Franceschetti & Lanari, 1999; Levanon, 1988).
2.1 Imaging by backprojection
The simplest and most intuitive type of beamforming is time-domain beamforming by
backprojection. This is done by back propagating the received signal via each pixel in the scene
to be imaged and into the transmitter. This is illustrated in Fig. 3. Backprojection is also
known as Delay-And-Sum (DAS) (Johnson & Dudgeon, 1993, pages 117-119). Functionally,
4 Sonar Systems
www.intechopen.com
Introduction to Synthetic Aperture Sonar 3
Fig. 3. Backprojection geometry.
backprojection is also closely related to Kirchhoff migration in seismic processing (Claerbout,
1995). Formally, backprojection is straightforward and can be done as follows. First the signals
in each receiver are ltered and pulse compressed. Then the signal from each receiver is
delayed to the pixel of interest
t
n
(x, y) = (r
t
+ r
n
)/c. (1)
where r
t
is the distance from the transmitter to the pixel, r
n
is the distance from the pixel to
receiver n, and c is the sound velocity. Then the recorded, pulse compressed signals s
n
are
summed coherently over all receivers to produce an estimate of the reectivity function
(x, y) =
1
N
N

n=1
A
n
s
n
(t
n
(x, y)). (2)
The amplitude A
n
is generally a function of geometry and frequency A
n
= A
n
(r
t
, r
r
, ).
In addition, the direction dependent sensitivity of the receivers can be taken into account.
This amplitude factor can also be used to control the sidelobe suppression vs the angular
resolution (Van Trees, 2002). For sampled time series s
n
(t
n
(x, y)) may have to be interpolated
to obtain accurate values in the image. A simple algorithmic description of the backprojection
algorithm is shown in Listing 1. We immediately realise why this method is called
Delay-And-Sum. The signals are delayed to the correct pixel, and then summed.
2.2 Angular resolution
The angular resolution can be dened as the minimum angle for which two reectors can be
separated in the sonar image. A very simple and intuitive way to derive this is as follows.
Assume a phased array receiver of length L consisting of a number of elements as illustrated
in Fig. 4. The angular resolution for this receiver is the angle difference for which the echo from
two reectors gives destructive interference in the receivers. Consider a reector at broadside
and a reector at an angle /2. As the maximum range difference will always be on the ends
of the array, we only consider the center element and the end element displaced L/2 apart.
The distance difference between these two reectors is
R = R
0
R
1
. (3)
5 Introduction to Synthetic Aperture Sonar
www.intechopen.com
4 Will-be-set-by-IN-TECH
f or a l l di r e c t i ons
f or a l l ranges
f or a l l r e c e i ve r s
Cal cul at e t he time del ay
I nt e r pol at e t he recei ved time s e r i e s
Apply appropri at e amplitude f a c t or
end
sum over r e c e i ve r s and s t or e i n gamma( x , y)
end
end
Listing 1. Backprojection Pseudo-code
For range differences larger than R = /4 destructive interference will start to occur. The
difference in range is
R = R
0
R
1
= L/2 sin(/2). (4)
We dene as the angular resolution and solve for when destructive interference will start
R = L/2 sin(/2) = /4. (5)
For small angles, we can approximate the sin-function
L/2/2 = /4 =

L
. (6)
Hence, the angular resolution (in radians) equals the inverse length of the array measured in
wavelengths. A longer array or a higher frequency (shorter wavelength) gives better angular
resolution. More details and a more accurate derivation of angular resolution can be found in
(Van Trees, 2002).
2.3 Imaging techniques for phased arrays
Fig. 5 shows the basic imaging geometry for an array of receivers (a phased array). The eld
of view is dened by the beamwidth of each of the receivers and their look direction. The
Fig. 4. Geometrical derivation of angular resolution.
6 Sonar Systems
www.intechopen.com
Introduction to Synthetic Aperture Sonar 5
Fig. 5. Field of view and resolution in imaging sonar.
angular resolution is given by the array length and the wavelength. If the individual elements
are more than /2 apart, the directivity of each element must be small enough to mitigate alias
lobes (see section 3.1). For an active system, the range resolution is dened by the bandwidth
of the system. The maximum range is determined by the pulse repetition interval and/or the
sensitivity of the receiver elements. For a linear array it can be fruitful and efcient for the
imaging to dene a polar coordinate system xed to the sonar.
In underwater applications, beamforming can be used to produce different types of products.
For high frequency sonar imaging, this can be categorized into three different types, as
illustrated in Fig. 6:
Sectorscan sonar produces a two-dimensional image for each pulse. These images are
usually shown on a display pulse by pulse. Sectorscanning sonar is often used as
hull mounted sonars for forward looking imaging or wide swath imaging. In sheries
acoustics, some cylindrical arrays actually produce full 360 degrees view.
Sidescan sonar is a particular type of sonar that uses the platform motion to cover different
parts of the seaoor. A sidescan sonar produces one or a few beams around broadside,
and an image of the seaoor is produced by moving the sonar and using repeated pulses.
This is a very popular technology, it has fairly low hardware complexity and can therefore
be more affordable. There are several books covering sidescan sonar very thoroughly
(Blondel, 2009; Fish & Carr, 2001).
Synthetic aperture sonar uses multiple pulses to create a large synthetic array (or aperture).
From this, an image of the seaoor is produced such that the information from multiple
pulses goes into each pixel on the seaoor. It is, from a certain point of view, the
combination of sidescan sonar and sectorscan sonar.
3. SAS sampling and coverage rate
In all types of array signal processing, the sampling of the array, or the spacing of the elements
and their directivity, is critical. If the array is too sparsely sampled, grating or alias lobes will
7 Introduction to Synthetic Aperture Sonar
www.intechopen.com
6 Will-be-set-by-IN-TECH
Fig. 6. Phased array imaging concepts for sonar.
occur, and the image quality will be reduced. In this section we describe the sampling criterion
and show how this affects SAS design.
3.1 Undersampling of arrays
Consider a simple array with two elements displaced D apart, as illustrated in Fig. 7. Assume
an inbound plane wave at broadside = 0 with wavelength from a reector in the far eld.
Any plane wave coming in the direction
sin
n
= n

2D
(7)
where n is an integer, will be perfectly in phase with the plane wave at broadside. Note that the
factor two, i.e. /2, comes fromassuming two-way propagation (transmission and reception).
This ambiguity will cause grating lobes or alias lobes in the beampattern (Manolakis et al.,
2000; Van Trees, 2002). To avoid grating lobes, the beamwidth of each individual element
must be smaller than the distance to the rst grating lobe. Assuming that each element is of
size d with a beamwidth of /d, we get the sampling criterion
D d/2. (8)
Hence, for a well sampled array, the effective distance between each element must be half the
size of each element. This is fullled for a densely populated array without gaps between
elements (see below).
Grating lobes are such that the problem they cause cannot be xed in signal processing
afterwards. For linear uniform arrays (ULAs), there is a deterministic and simple one-to-one
Fig. 7. Along track sampling and grating (alias) lobes.
8 Sonar Systems
www.intechopen.com
Introduction to Synthetic Aperture Sonar 7
Fig. 8. Multi-element receiver array and the Phase Center Approximation.
relation between the sampling of the array and the angular position of the grating lobes.
Grating lobes are frequency dependent. For large bandwidth systems, the grating lobes will
be less dened due to the fact that their positions change with frequency.
3.2 Multi-element receiver arrays in SAS
In SAS, the transmitter-receiver pair is moved along-track and pulsed repeatedly to form a
long synthetic antenna. The sampling criterion (8) imposes a maximumallowed displacement
between pings, which again, transforms into a maximum range for a given speed. The
maximum range R
max
is given by
2R
max
/c = T
pri
= D/v R
max
=
cD
2v
(9)
where v is the platform speed and T
pri
is the pulse repetition interval. This relation imposes
a very strong limitation to SAS. Consider a platform speed of v = 2 m/s. For a displacement
of D = 5 cm, the maximum range then becomes R
max
= 18.75 m for a sound velocity of
c = 1500 m/s. This is an unacceptable small range for most applications. To overcome
this, a receiver array of multiple elements can be used (Bruce, 1992; Cutrona, 1975). The
multi-element receiver array is used in almost all existing SAS systems today.
A simple description of how a single transmitter multiple receiver system can be used to
form a synthetic aperture is as follows. Consider a single transmitter Tx and N receivers
of size d in a linear array Rx, as illustration shown in Fig. 8. In stead of treating this as a
bistatic systemwith one transmitter and many receivers, we assume a virtual array where each
element is placed at the middle position between each transmitter-receiver pair (indicated
in light green). In this array, each element is a transmitter-receiver (monostatic). This is the
Phase Center Approximation (PCA) (Bellettini &Pinto, 2002). The virtual array (or PCAarray)
is half the length of the receiver array, with element size d/2, and each of the elements in
the virtual array can be placed directly in a synthetic array (similar to SAR). The maximum
displacement between two pulses becomes the length of the PCA array D = Nd/2 = L/2,
and the maximum range becomes
R
max
=
cL
4v
(10)
where L is the length of the receiver array. The maximum range can then be increased by
increasing the length of the receiver array. Fig. 9 shows the maximum range as function of
vehicle speed for three different receiver array lengths. Note that the area coverage rate (speed
times range) is a constant proportional to the array length.
In airborne SAR, aperture undersampling is not a problem, since the phase velocity for
electromagnetic waves in air is relatively high compared to the platform speed. In large
9 Introduction to Synthetic Aperture Sonar
www.intechopen.com
8 Will-be-set-by-IN-TECH
Fig. 9. Maximum range for given platform speed for three different receiver array lengths.
swath high resolution space-borne SAR, there is a limitation related to the sampling criterion.
Advanced SAR systems have 2D phased arrays and use electronic steering of the transmitter
and receiver arrays for different SAR modes. ScanSAR (Franceschetti & Lanari, 1999) and
Terrain Observation by Progressive Scans (TOPS) (Gebert et al., 2010) are used to increase the
area coverage rate while lowering the resolution. Spotlight mode (Jakowatz et al., 1996) is
used to increase the resolution while lowering the area coverage rate.
4. Design of Synthetic Aperture Sonar
SAS is different from traditional sonar in several ways. In this section we describe some of the
signicant differences that affect the performance and application areas of SAS.
4.1 Resolution
While traditional imaging sonar has constant angular resolution, and thereby range
dependent along-track resolution, SAS produces range independent along-track resolution.
This is done by increasing the length of the synthetic array as function of range. Consider the
left illustration of Fig. 10. The maximum length of the synthetic aperture is given by the eld
of viewof each transmitter/receiver element. At range R
1
, the length of the synthetic aperture
becomes
L
1
R
1
(11)
where = /d is the eld of view. The along-track resolution is given by the synthetic
antenna
x R
1

2L
1
(12)
where the factor 2 comes from the fact that both the transmitter and receiver is moved along
the synthetic aperture (creating a focused receiver and transmitter). Inserting for L
1
and we
get a resolution of
x R
1

2L
1
= R
1

2R
1
= R
1

2/dR
1
=
d
2
. (13)
10 Sonar Systems
www.intechopen.com
Introduction to Synthetic Aperture Sonar 9
Fig. 10. Left: Range independent along-track resolution in SAS. Right: Frequency
independent along-track resolution in SAS.
Hence, the along-track resolution for full length synthetic apertures is given by the element
size alone, independent of range. This is a well known result, and there are numerous
books covering this (Curlander & McDonough, 1991; Franceschetti & Lanari, 1999). This
gives a degree of freedom not available in traditional sonar, where the cross-range resolution
inevitably decreases with increasing range.
The length of the synthetic aperture is range dependent. Consider again the left drawing
in Fig. 10. At range R
2
, the length of the synthetic aperture becomes L
2
to obtain the same
along-track resolution as for range R
1
. The image quality becomes range dependent, since
the transmission loss (Lurton, 2010) and thereby the Signal to Noise Ratio (SNR) is range
dependent. Since the synthetic aperture is longer at longer range, the required accuracy in
navigation, sound velocity and topography becomes thougher (see section 5). Hence, SAS at
long range is inherently more difcult than SAS at short range.
4.2 Frequency dependence
Another intriguing fact is that SAS provides frequency independent along-track resolution.
This is achieved by increasing the length of the synthetic aperture for decreasing frequency.
The right drawing in Fig. 10 illustrates this. The lower frequency (red curve) requires a longer
synthetic aperture than the higher frequency (blue curve) for a xed resolution. The angular
spread, however, becomes frequency dependent. For lower frequency SAS, larger angular
spread is needed for a given along-track resolution. There is a lower limit to the resolution,
where (13) does not hold. The best possible along-track resolution that can be achieved is
x = /4, for 180

eld of view.
The frequency independence gives a degree of freedom compared to traditional sonar.
Seawater is a dissipative mediumfor acoustic waves through viscosity and chemical processes
(Brekhovskikh & Lysanov, 1982, pages 9-11), (Lurton, 2010, pages 23-27), (Medwin & Clay,
1998, pages 104-110). Acoustic absorption in seawater is frequency dependent, such that lower
frequencies reach longer than higher frequencies. Fig. 11 shows the two-way transmission
loss for some typical frequencies used in high frequency sonar imaging. For a given reception
threshold (sensitivity), the maximum achievable range changes substantially. In traditional
sonar, a typical design criterion is to choose the highest possible frequency for the desired
range. Then design the transmitter and receiver arrays to obtain best possible azimuth
resolution. In SAS, the along-track resolution is independent of frequency. This allows the
center frequency to be chosen for other reasons than resolution.
11 Introduction to Synthetic Aperture Sonar
www.intechopen.com
10 Will-be-set-by-IN-TECH
Fig. 11. Two-way transmission loss for typical frequencies in SAS.
4.3 Design parameters
There are many different SAS designs based on different philosophies. Here, we list some
systems to illustrate the variety in existing designs. In (Bellettini & Pinto, 2009), the sonar
has a center frequency of 300 kHz, and a receiver array length of 1.2 m with 36 elements.
This sonar was designed to have a maximum range of 200 m at 2 m/s, good shallow water
performance and as little angular spread as possible on the targets. In (Larsen et al., 2010), the
center frequency is 60 kHz, and the receiver array is 4.3 m with 32 elements and a maximum
range is 1500 m at 1 m/s speed. This system was specically designed for deep tow with very
large swath. The systemdescribed in (Jean, 2008) has a center frequency of 100 kHz, a receiver
array of length 2 mwith 24 elements, giving a maximumrange of 300 mat 2.5 m/s speed. The
system described in (Glover & Campell, 2010) is a dual frequency SAS at 17.5 kHz and 150
kHz, with a receiver array of 0.6 m length and 24 receiver elements. This gives a maximum
range of 100 m at 2 m/s speed. The system described in section 8.1 has a center frequency of
100 kHz and a receiver length of 1.2 m with 32 elements, giving 200 m range at 2 m/s speed.
In Table 1 we list some design parameters (left column) and what controls these (right
column). The general rule of thumb is that higher performance in resolution and area coverage
rate requires more hardware, more processing capability and thereby also more space and
power consumption (as expected). A slightly less obvious conclusion is that a high resolution
system inherently is more robust. The future will surely bring new designs as the frequency
agility of transducers improves, and the ability for large number of channels increases.
5. Challenges in SAS
The principle of SAS has been known in more than 40 years (Gough & Hawkins, 1998; Hayes
& Gough, 2009), but it is only the last few years SAS has become commercially available. The
main reason for this is the specic challenges that has to be solved for successful SAS imagery.
In this section, we list some fundamental challenges and indicate how these can be solved.
More details can be found in (Hagen & Hansen, 2008; 2009; Hansen, 2010; Hansen, Callow,
Sb & Synnes, 2010).
12 Sonar Systems
www.intechopen.com
Introduction to Synthetic Aperture Sonar 11
Design Parameter Relation
Range resolution System Bandwidth
Along-track resolution Element size
Area coverage rate Receiver array length
Maximum range Frequency, receiver array length and speed
Signal to Noise Ratio Frequency, geometry and maximum range
Robustness Relative bandwidth and redundancy
Complexity Beamwidth, number of elements and relative bandwidth
Throughput Bandwidth times number of receiver elements
Table 1. Design parameters and their relation in SAS.
5.1 Navigation
The sonar has to be positioned with accuracy better than a fraction of a wavelength along the
synthetic aperture. This is the same requirement as for any other array sensor, but inherently
more difcult to obtain since a synthetic antenna is formed by a moving platform. A sharp
image depends on accurate positioning of the elements in the array. How much error is
tolerated is dependent on the scale of the error and the required image quality (Carrara et al.,
1995). Navigation of underwater platforms is more difcult than navigation of airborne and
terrestrial platforms because GPS is not available. This challenge is the most important, and
the rst that has to be overcome for successful SAS imagery.
One solution is to use the sonar data itself for navigation (referred to as micronavigation).
Compare the illustration in Fig. 12 with Fig. 8. By moving the SAS system a distance
D = Md/2 such that M < N, there will be redundancy (or overlap) in the synthetic aperture.
The overlap in the phase center antenna (PCA) is indicated in orange. Estimating which
channels overlap can be used to estimate the displacement along-track and cross-track. This
can be done by cross correlating the data recorded by elements fromping p with data recorded
by other elements from ping p + 1. The method is named displaced phase center antenna
(DPCA) or redundant phase center (RPC). DPCA was rst used in Moving Target Indicator
(MTI) radars in the 1950s (Dickey Jr et al., 1991). In SAS, early descriptions of DPCA can be
found in (Pinto et al., 1997; Sheriff, 1992). A very good overview of the method is given in
(Pinto, 2002), and a detailed study of the performance is given in (Bellettini & Pinto, 2002). In
(Hansen et al., 2003) different strategies to combine DPCA with traditional inertial navigation
is described.
Micronavigation using DPCA causes a trade-off between navigation performance and
efciency (area coverage rate). This is due to the fact that the displacement between pings
has to be less than the maximum displacement L/2. More overlap gives better navigation but
lower area coverage rate. An exception to this is if several transmitters, specic bandwidth,
and a specic pulse regime is used (Billon & Fohanno, 2002).
5.2 Topographic errors
When running a vehicle on a non-straight track, a non-straight synthetic aperture is formed
and the imaging geometry becomes dependent on the full three-dimensional geometry. This
means that the position of the sonar has to be known and the topography (or bathymetry)
of the scene to be imaged has to be known (Jakowatz et al., 1996, pages 187-197). Only
then, successful SAS processing can be done. This is critical for robust SAS, and a signicant
problem since the topographic changes in rough terrain may impose severe non-linear tracks
(Hansen et al., 2009). There are two solutions to this challenge: either run on a straight line or
13 Introduction to Synthetic Aperture Sonar
www.intechopen.com
12 Will-be-set-by-IN-TECH
Fig. 12. Multi-element receiver array and the principle for Displaced Phase Center Antenna.
obtain a map of the area before synthetic aperture processing. The former is impractical (or
impossible) on small platforms such as autonomous underwater vehicles (AUVs) or towsh
systems. The latter can be obtained by using an interferometric sonar to map the scene prior
to SAS processing.
5.3 Ocean environment
SAS is near-eld imaging. This implies that the sound velocity has to be accurately estimated
for well focused imaging. The sound velocity in the ocean varies with depth (Brekhovskikh
& Lysanov, 1982, pages 1-9), (Lurton, 2010, chapter 2.6). There might also be local horizontal
and temporal variations. This can cause variation in the sound velocity of up to 2% along
the acoustic path. The effect of incorrect sound velocity on SAS is described in (Hansen
et al., 2007; Hansen, Callow, Sb & Synnes, 2010). To overcome this, the sound velocity
can be measured directly or calculated using a Conductivity, Temperature, Depth (CTD)
sensor. Another possibility is to auto correct for incorrect sound velocity by using an autofocus
technique (Hansen et al., 2007; Jakowatz et al., 1996).
5.4 Vehicle stability
SAS systems are generally multi-element receiver systems (see section 3.1). This affects
synthetic aperture processing in several ways. For vehicles operated in a crabbing
environment (where the heading is not aligned with the track), a baseline occurs between
overlapping elements. Even if the track is perfectly linear, the synthetic aperture becomes
non-linear and the image quality in SAS processing becomes dependent on the topography
accuracy (see section 5.2). Large crab-induced baselines is challenging for imaging and also
micronavigation (Callow, 2010).
5.5 Multipath environment
When operating in shallow waters, multiple reections (or multipath) via the sea surface
might affect the performance of sonar (Lurton, 2010, chapter 2.4), (Bellettini et al., 2003).
This can affect the SAS data threefold: Multipath can potentially lower the temporal
coherence between pings in the micronavigation; multipath can reduce the spatial coherence
in interferometry; and multipath can add unwanted signal to the SAS images, causing loss of
shadow contrast and delity. The latter two problems applies to all sonars in shallow waters,
not only SAS. How much the shallow water environment affects SAS is actually dependent
of the seaoor conditions, the sound velocity prole and the sea surface roughness (Synnes
et al., 2009).
14 Sonar Systems
www.intechopen.com
Introduction to Synthetic Aperture Sonar 13
Fig. 13. Overview of SAS signal processing ow.
6. Signal processing of SAS data
Signal processing of SAS data can be done in many different ways. Fig. 13 shows the shcematic
overview of a processing ow. Signal processing of SAS data is similar to SAR processing,
with a few exceptions. Accurate knowledge of the vehicle navigation (track position and
sensor orientation), the ocean environment and in particular the sound velocity, and the
seaoor topography must be obtained for successful SAS imagery (see section 5).
The rst step (blocking) in the processing is to divide the data into portions suitable for
synthetic aperture processing. The blocking selects which pings to put into one synthetic
aperture, and decides what to do with the data. If micronavigation is required, this
must take place before imaging. There are two classes of image formation algorithms:
time domain imaging and frequency domain imaging. In general, the frequency domain
algorithms are more efcient but require more controlled vehicle behavior (close to straight
line tracks). Candidates are the wavenumber algorithm (Soumekh, 1994),(Cumming & Wong,
2005, chapter 8), (Carrara et al., 1995, chapter 10) (also referred to as range migration algorithm
or Omega-K algorithm), and the chirp scaling algorithm (Cumming & Wong, 2005, chapter 7),
(Franceschetti &Lanari, 1999). Alternatively, the image formation can be done in time domain
(see section 2.1). An overview of different algorithms used in SAS imaging can be found in
(Gough & Hawkins, 1997).The benet of time domain beamforming is that the imaging grid
can be arbitrary and the data acquisition (the synthetic aperture) can be severely distorted.
This gives an added exibility not available in frequency domain imaging (Massonnet &
Souyris, 2008, chapter 2.4). The disadvantage of time domain imaging is the computational
load. After image formation, blind correction for residual errors (known as autofocus) in
the image can be performed (Jakowatz et al., 1996, chapter 4), (Callow, 2003). It should be
noted that most autofocusing techniques are local methods - they have a certain success in
correcting local errors, but do not perform well on a large image with different sources for
errors in different locations in the image. For interferometric systems, the nal stage in the
SAS processing ow is bathymetry estimation using interferometry (Hanssen, 2001; Sb,
2010).
15 Introduction to Synthetic Aperture Sonar
www.intechopen.com
14 Will-be-set-by-IN-TECH
7. Properties of SAS images
The goal of sonar imaging is to estimate the acoustic reectivity in the best possible manner,
given the sensor and geometry (see section 2). (Oliver & Quegan, 1998) gives an excellent
overview of the properties of SAR images (highly relevant for SAS), related to the signal
processing and the scene content. In the following, we list common measures of system
performance, applicable to any imaging system.
Geometrical resolution or detail resolution. This is the minimum distance between two
reectors where they can be resolved in the image. The theoretical geometric resolution is
given by the bandwidth for the range dimension and the element size for the along-track
dimension (see section 4.1). The true geometrical resolution is also dependent on the image
quality. Defocus will reduce the geometrical resolution (Oliver & Quegan, 1998, chapter
3.2).
Radiometric resolution or contrast / value resolution, echogenicity or target strength
accuracy. This is the accuracy of the estimated value in each pixel. All coherent imaging
systems suffers from speckle (Goodman, 2007; Oliver & Quegan, 1998). Speckle is
randomvariability caused by constructive and destructive interference between individual
scatterers in each geometrical resolution cell. Speckle causes a variance in the pixel value
and thereby a reduced radiometric resolution. There are different methods to despeckle
images and to estimate the scattering cross section. The traditional approach in SAR
has been to apply multilook processing to reduce speckle (Jakowatz et al., 1996, chapter
3.3). There also exist more advanced methods to estimate the scattering cross section
(Massonnet & Souyris, 2008, chapter 3.11), (Oliver & Quegan, 1998, chapter 6). The
radiometric resolution strictly depends on a fully calibrated system, where the whole
system has to be energy preserving. This is non-trivial to obtain, and it implies that all
the terms except the Target Strength in the sonar equation has to be accounted for (Ainslie,
2010; Lurton, 2010), (Curlander & McDonough, 1991, chapter 7).
Dynamic range or resolvability of small targets in the presence of large targets. This is a
function of the sidelobe levels or the shape of the point spread function. There is a trade-off
between geometrical resolution and dynamic range. Large dynamic range requires large
sidelobe suppression which causes poorer geometrical resolution (Franceschetti & Lanari,
1999, chapter 3.1), (Carrara et al., 1995, chapter 8). In addition to the sidelobes, the alias
lobes (or grating lobes) must also be controlled in order to obtain the desired dynamic
range. This can only be achieved by oversampling the synthetic aperture (Bellettini &
Pinto, 2009; Gough & Hawkins, 1997).
Sensitivity or detection ability of lowlevel targets. This is determined by several of the terms
in the sonar equation (Lurton, 2010; Urick, 1983), such as systemself noise, transmit power,
transmission loss (which is a function of acoustic frequency) and processing gain. See
(Ainslie, 2010) for a detailed description of the terms in the sonar equation. A system with
larger pulse compression gain will have improved sensitivity.
Temporal resolution or framerate. This is the number of independent images on the scene
per unit time. Since SAS is based on space-time processing (spatial movement uses time to
generate aperture), SAS only has one frame on the scene when using full aperture length.
Multi-aspect imaging (Hansen et al., 2008) can be applied to produce multiple aspects from
different apertures and thereby different time intervals. There is a trade-off between looks
and along-track resolution (since the aperture becomes shorter).
16 Sonar Systems
www.intechopen.com
Introduction to Synthetic Aperture Sonar 15
Fig. 14. The ability to retrieve relevant information from a SAS image is dependent on a
number of different factors.
The ability to extract the relevant information from a SAS image depends on a number of
different factors, as illustrated in Fig. 14. Observation geometry given by range and elevation
angle is important for interpretation of the highlight structure and shadow. Backscattered
target signals can contain elements of specular reections, diffuse scattering, transparency,
resonant scattering and multiple scattering, all of which complicate the process of retrieving
the relevant information. Finally, image resolution is critical in resolving or classifying objects,
shadow shape and the surrounding scatterers on the seaoor.
8. Applications of SAS
There are many applications where SAS is suitable. Very high resolution acoustic imaging
can be achieved on traditional sonar using very high frequencies, and there are sonar systems
today using up to 2 MHz frequency. They are, hovewer, very limited in range. When large
area coverage and very high resolution is needed at the same time, SAS is really the only
technology that can provide a solution. In this section, we describe a particular SAS system,
the HISAS 1030, and show example images from different applications.
8.1 The HISAS 1030 interferometric SAS
HISAS 1030 is a wideband widebeam interferometric SAS developed by Kongsberg Maritime
and FFI (Fossum et al., 2008; Hagen et al., 2008). The sonar contains two along-track receiver
arrays of 1.2 m length with 32 elements in each array, and a vertical baseline approximately
30 cm which equals 20 wavelengths. The transmitter is a vertical phased array with 16
elements, and the transmit beam can be electronically steered and shaped to obtain the best
possible performance in shallowwaters (see Section 5.5). The transmitter can also be used as a
receiver, giving 16 individual receiver channels along a vertical array. Fig. 15 shows the sonar
mounted on a HUGIN 1000-MR AUV. Typical HISAS 1030 specications are listed in Table
2. The SAS processing is done in a software suite named FOCUS Toolbox (Hansen et al., 2005;
2003).
Fig. 16 shows an example SAS image made by the HISAS 1030 on the HUGIN 1000 AUV. The
image shows the wreck of the 1500 dwt oil tanker Holmengraa lying on a slanted seabed at
17 Introduction to Synthetic Aperture Sonar
www.intechopen.com
16 Will-be-set-by-IN-TECH
Fig. 15. The HISAS 1030 interferometric SAS on the HUGIN autonomous underwater
vehicle. The picture was taken just before launch during a scientic mission on board the
research vessel H U Sverdrup II in Norwegian waters in April 2010.
Center frequency 100 kHz
Wavelength 1.5 cm
Bandwidth 30 kHz
Total frequency range 50-120 kHz
Along-track resolution 3 cm
Cross-track resolution 3 cm
Maximum range @ 2 m/s 200 m
Area coverage rate 2 km
2
/h
Table 2. Typical system specications for the HISAS 1030 interferometric SAS.
depth 77 m. The distance to the center of the image is about 95 m. The length of the wreck is
about 68 m and width about 9 m.
Fig. 17 captures the essence of SAS: Long range and high resolution at the same time. The large
(middle) image shows a SAS image where the range is 25 m (left) to 325 m (right), and the
dynamic range is 32 dB. The data was collected with HISAS 1030 on HUGIN AUV running at
2.3 knots at 40 m altitude, outside Horten, Norway in approximately 200 m water depth. The
upper image shows a section from 200 m range to 250 m range with the wreck of the German
WWII submarine U735. The lower image shows a section from 260 m range to 290 m range
with a 1 m
3
cube. The length of the synthetic aperture is (see section 4.1)
L
sa
R

d
(14)
18 Sonar Systems
www.intechopen.com
Introduction to Synthetic Aperture Sonar 17
Fig. 16. SAS image of the wreck of the Norwegian tanker Holmengraa that was sunk during
WWII in 1944. Courtesy of Kongsberg Maritime.
where R is the range, is the wavelength at center frequency, d is the along-track element
size in the array. is a programmable parameter controlling the process beamwidth, that is, the
beamwidth actually processed. In this particular case, = 2/3, and the length of the synthetic
aperture at maximum range becomes L
sa
90 m = 6000. The SAS resolution-gain, dened
as the ratio between along-track resolution in real aperture x
ra
and synthetic aperture x
sa
is
Q
sa
=
x
ra
x
sa
=
L
sa
L/2
R
2
Ld
, (15)
where and L is the array length. In Fig. 17, the SAS resolution-gain is Q
sa
150 at maximum
range. This is a considerable resolution improvement, and the equivalent along-track
resolution is very difcult to obtain using real aperture techniques.
8.2 Underwater archaeology
SAS is a candidate technology in searching for wrecks and other objects of historic interest.
Fig. 18 shows a SAS image collected by the Royal Norwegian Navy in a training mission close
to the town of Troms in the winter of 2009. The data was collected at 34 m water depth,
close to shore. The image shows the wreck of a German WWII Heinkel He 115 seaplane. The
original length of the plane is 17 m and the wingspan was 22 m. Note the small part outside
the right wing of the seaplane. This is probably one of the oats. The object in the lower left
part of the image is probably the tail-section of another plane of the same type.
19 Introduction to Synthetic Aperture Sonar
www.intechopen.com
18 Will-be-set-by-IN-TECH
Fig. 17. Middle: SAS image where the range is 25 m (left) to 325 m (right). Courtesy of
Kongsberg Maritime.
20 Sonar Systems
www.intechopen.com
Introduction to Synthetic Aperture Sonar 19
Fig. 18. SAS image of a German WWII Heinkel He 115 seaplane lying on the seaoor.
Courtesy of the Royal Norwegian Navy. Photograph is from wikipedia.
8.3 Search for small objects
When searching for small objects over very large areas, SAS is an excellent tool. Fig. 19 shows
a SAS image from an area with debris. The sonar has traveled along the vertical direction
of the image, and the sonar look direction is towards right. The size of the image is 190 m
(along-track) and 30 m to 165 m cross-track, and the water depth is around 70 m. The yellow
boxes indicates three objects, and the small images below show zoomed images of the objects.
These are two drums of approximate size 0.9 m length and 0.6 m diameter, and a cylinder of
approximate length 2.5 m. The lower row of images shows optical images of the same objects.
The optical images were collected with another HUGIN AUV. The altitude was 5 m on the
data collection of the optical images. We see several interesting features in the objects. The
drumat 73 mhas clear indications of partial transparency. The back end of the barrel is clearly
visible, and there is acoustic pollution in the shadow region. The drum at 112 m has a less
dened back end and deeper shadow contrast, indicating that this object is less transparent.
In the optical images, indeed, we see that the drum at 73 m has severe damage and holes,
while the drum at 112 m looks to be more intact. In the optical image of the cylinder, we see a
small cavity in the lower right end. This appears as a highlight in the SAS image.
21 Introduction to Synthetic Aperture Sonar
www.intechopen.com
20 Will-be-set-by-IN-TECH
Fig. 19. Upper: SAS image of an area with small objects. The range is 30 m (left) to 165 m
(right), and the along-track (vertical) is 190 m. The small cut-outs shows zoomed images of
three different objects. The lower row shows optical images of the same targets. Courtesy of
Kongsberg Maritime / FFI.
22 Sonar Systems
www.intechopen.com
Introduction to Synthetic Aperture Sonar 21
8.4 Inspection of man made constructions
External inspection of underwater constructions such as pipelines is an important task. The
objective of these inspections is to detect burial, exposure, free spans and buckling of the
pipeline, as well as possible damages due to trawling, anchoring and debris near the pipeline.
SAS may be well suited technology for some of these tasks (Hagen et al., 2010), (Hansen,
Sb, Callow & Hagen, 2010). Fig. 20 shows an example SAS image collected by a HUGIN
AUV during a demonstration in San Diego, USA, in 2010. The image shows a sewer pipeline,
and a rope or wire on the seaoor.
Fig. 20. Upper: SAS image of a pipeline outside San Diego. The range in the upper image is
65 m - 145 m. Lower: 20 m times 10 m zoomed area at range 85 m of a rope or wire on the
seaoor. Courtesy of Kongsberg Maritime.
23 Introduction to Synthetic Aperture Sonar
www.intechopen.com
(a) SAS image 1 (b) SAS image 2
9. Conclusion
Synthetic aperture sonar (SAS) is an advanced signal processing technique to improve
resolution in sonar imagery. The main application is detailed documentation of the seaoor,
in areas such as search for small objects, underwater archaeology, detailed seabed mapping
and documentation of underwater installations. Successful SAS is dependent on accurate
knowledge about the sonar position, the ocean environment and the seabed topography.
SAS is substantially more mature now than 10 years ago. To illustrate the maturity of
commercially available systems, we showa nal example of SAS data collected by the HUGIN
AUV carrying the HISAS 1030. Fig. 21 shows SAS images and interferometric SAS relative
bathymetries of a German WWII Focke Wulf 190 A-3 aeroplane that was found by the Royal
Norwegian Navy mine warfare otilla. The length of the plane is 9 m and the wingspan is
10.5 m. The tail was damaged as we see in the SAS images and the bathymetries. The motor
fell off during the recovery. These images was produced at sea by the Royal Norwegian Navy
personnel during the search operation. The images was constructed using micronavigation
and sidescan bathymetry as a preprocessing step, then backprojection in three dimensions
for image formation, and nally bathymetry estimation using a maximum likelihood phase
estimator in the interferometric processing.
10. Acknowledgments
The author thanks the very good colleagues Hayden J Callow, Torstein O Sb, and Stig
A V Synnes at the Norwegian Defence Research Establishment. The author also thank
Kongsberg Maritime, and in particular Per Espen Hagen and Bjrnar Langli, for a long
standing collaboration and providing data for the analysis. Finally the author wish to thank
the Royal Norwegian Navy Mine Warfare Service for the long fruitful collaboration and kind
permission to use data recorded during Navy operations with their HUGIN 1000-MR AUV.
24 Sonar Systems
www.intechopen.com
(c) Bathymetry by SAS interferometry 1 (d) Bathymetry by SAS interferometry 2
Fig. 21. The German WWII Focke Wulf 190 A-3 aircraft. The plane was found by the Royal
Norwegian Navy Mine warfare otilla at 98 m water depth during an underwater
archaeology mission. Courtesy of the Royal Norwegian Navy.
(e) Photograph taken during the recovery (f) Photograph taken during the recovery
25 Introduction to Synthetic Aperture Sonar
www.intechopen.com
24 Will-be-set-by-IN-TECH
11. References
Ainslie, M. (2010). Principles of Sonar Performance Modelling, Springer Verlag.
Bellettini, A. & Pinto, M. A. (2002). Theoretical accuracy of synthetic aperture sonar
micronavigation using a displaced phase-center antenna, IEEE J. Oceanic Eng.
27(4): 780789.
Bellettini, A. & Pinto, M. A. (2009). Design and Experimental Results of a 300-kHz Synthetic
Aperture Sonar Optimized for Shallow-Water Operations, IEEE J. Oceanic Eng.
34(3): 285293.
Bellettini, A., Pinto, M. A. & Wang, L. (2003). Effect of multipath on synthetic aperture sonar,
Proc. 5th World Congr. Ultrasonics, Paris, France, pp. 531534.
Billon, D. & Fohanno, F. (2002). Two improved ping-to-ping cross-correlation methods for
synthetic aperture sonar: theory and sea results, OCEANS02 MTS/IEEE, Vol. 4, IEEE,
pp. 22842293.
Blondel, P. (2009). The Handbook of Sidescan Sonar, Geophysical Sciences, Springer Praxis Books.
Brekhovskikh, L. & Lysanov, Y. (1982). Fundamentals of Ocean Acoustics, Vol. 8 of Springer Series
in Electrophysics, Springer-Verlag, Berlin, Germany.
Bruce, M. P. (1992). A processing requirement and resolution capability comparison of
side-scan and synthetic-aperture sonars, IEEE J. Oceanic Eng. 17(1): 106117.
Burdic, W. S. (1984). Underwater acoustic system analysis, Prentice Hall.
Callow, H. J. (2003). Signal Processing for Synthetic Aperture Sonar Image Enhancement, PhD
thesis, University of Canterbury, Christchurch, New Zealand.
Callow, H. J. (2010). Comparison of SAS processing strategies for crabbing collection
geometries, Proceedings of Oceans 2010 MTS/IEEE, Seattle, US.
Carrara, W. G., Goodman, R. S. & Majewski, R. M. (1995). Spotlight Synthetic Aperture Radar:
Signal Processing Algorithms, Artech House.
Claerbout, J. (1995). Basic earth imaging, Stanford Exploration Project.
URL: http://sepwww.stanford.edu/
Cumming, I. G. & Wong, F. H. (2005). Digital Processing of Synthetic Aperture Radar Data:
Alogirthms and Implementation, Artech House.
Curlander, J. C. & McDonough, R. N. (1991). Synthetic Aperture Radar: Systems and Signal
Processing, John Wiley & Sons, Inc., 605 Third Avenue, New York, NY.
Cutrona, L. J. (1975). Comparison of sonar system performance achievable using
synthetic-aperture techniques with the performance achievable by more
conventional means, J. Acoust. Soc. Am. 58(2): 336348.
Dickey Jr, F., Labitt, M. & Staudaher, F. (1991). Development of airborne moving target radar
for long range surveillance, Aerospace and Electronic Systems, IEEE Transactions on
27(6): 959972.
Fish, J. P. & Carr, H. A. (2001). Sound reections: Advanced Applications of Side Scan Sonar,
LowerCape Publishing.
Fossum, T. G., Hagen, P. E., Langli, B. & Hansen, R. E. (2008). HISAS 1030: High resolution
synthetic aperture sonar with bathymetric capabilities, Shallow survey, Portsmouth,
NH, USA.
Franceschetti, G. & Lanari, R. (1999). Synthetic Aperture Radar Processing, CRC Press.
Gebert, N., Krieger, G. & Moreira, A. (2010). Multichannel azimuth processing in ScanSAR
and TOPS mode operation, IEEE Trans. Geosci. Remote Sensing. 48(7): 29943308.
26 Sonar Systems
www.intechopen.com
Introduction to Synthetic Aperture Sonar 25
Glover, N. P. & Campell, I. (2010). Simultaneous low and high frequency high resolution
SAS and a statistical method of quantifying the resolutions obtained, Proceedings of
Synthetic Aperture Sonar and Radar 2010, Lerici, Italy.
Goodman, J. W. (2007). Speckle Phenomena in Optics: Theory and Applications, Roberts and
Company.
Gough, P. T. & Hawkins, D. W. (1997). Imaging algorithms for a strip-map synthetic aperture
sonar: Minimizing the effects of aperture errors and aperture undersampling, IEEE
J. Oceanic Eng. 22 (1): 2739.
Gough, P. T. & Hawkins, D. W. (1998). A short history of synthetic aperture sonar, Geoscience
and Remote Sensing Symposium Proceedings, 1998. IGARSS98. 1998 IEEE International,
Vol. 2, IEEE, pp. 618620.
Hagen, P. E., Brhaug, E. & Midtgaard, . (2010). Pipeline inspection with interferometric
SAS, Sea Technology pp. 3740.
Hagen, P. E., Fossum, T. G. & Hansen, R. E. (2008). HISAS 1030: The next generation mine
hunting sonar for AUVs, UDT Pacic 2008 Conference Proceedings, Sydney, Australia.
Hagen, P. E. & Hansen, R. E. (2008). Synthetic aperture sonar challenges ... and how to meet
them, Hydro International pp. 2631.
Hagen, P. E. & Hansen, R. E. (2009). Robust synthetic aperture sonar operation for AUVs,
Proceedings of Oceans 09 MTS/IEEE Biloxi, Biloxi, MS, USA.
Hansen, R. E. (2010). Robust synthetic aperture sonar for autonomous underwater vehicles,
Proceedings of Synthetic Aperture Sonar and Radar 2010, Lerici, Italy.
Hansen, R. E., Callow, H. J. & Sb, T. O. (2007). The effect of sound velocity variations on
synthetic aperture sonar, Proceedings of Underwater Acoustic Measurements 2007, Crete,
Greece.
Hansen, R. E., Callow, H. J., Sb, T. O., Hagen, P. E. &Langli, B. (2008). High delity synthetic
aperture sonar products for target analysis, Proceedings of Oceans 08 Quebec, Quebec,
Canada.
Hansen, R. E., Callow, H. J., Sb, T. O., Synnes, S. A., Hagen, P. E., Fossum, T. G. & Langli,
B. (2009). Synthetic aperture sonar in challenging environments: Results from the
HISAS 1030, Proceedings of Underwater Acoustic Measurements 2009, Nafplion, Greece.
Hansen, R. E., Callow, H. J., Sb, T. O. & Synnes, S. A. V. (2010). Challenges in seaoor
imaging and mapping with synthetic aperture sonar, Proceedings of EUSAR 2010,
Aachen, Germany, pp. 540543.
Hansen, R. E., Sb, T. O., Callow, H. J. & Hagen, P. E. (2010). Interferometric synthetic
aperture sonar in pipeline inspection, Proceedings of Oceans 2010 MTS/IEEE, Sydney,
Australia.
Hansen, R. E., Sb, T. O., Callow, H. J., Hagen, P. E. & Hammerstad, E. (2005). Synthetic
aperture sonar processing for the HUGIN AUV, Proceedings of Oceans 05 Europe,
Vol. 2, Brest, France, pp. 10901094.
Hansen, R. E., Sb, T. O., Gade, K. & Chapman, S. (2003). Signal processing for AUV based
interferometric synthetic aperture sonar, Proceedings of Oceans 2003 MTS/IEEE, San
Diego, CA, USA, pp. 24382444.
Hanssen, R. F. (2001). Radar Interferometry: Data Interpretation and Error Analysis, Kluwer
Academic Publishers.
Hayes, M. P. & Gough, P. T. (2009). Synthetic aperture sonar: A review of current status, IEEE
J. Oceanic Eng. 34(3): 207224.
27 Introduction to Synthetic Aperture Sonar
www.intechopen.com
26 Will-be-set-by-IN-TECH
Jakowatz, J. C. V., Wahl, D. E., Eichel, P. H., Ghiglia, D. C. & Thompson, P. A. (1996).
Spotlight-Mode Synthetic Aperture Radar: A Signal Processing Approach, Kluwer
Academic Publishers.
Jean, F. (2008). Shadows, synthetic aperture sonar and forward looking gap-ller: different
imaging algorithms, Proceedings of OCEANS 2008 - MTS/IEEE Kobe Techno-Ocean,
Kobe, Japan.
Johnson, D. H. &Dudgeon, D. E. (1993). Array signal processing: Concepts and Techniques, Signal
processing series, Prentice Hall, Englewood Cliffs, NJ, USA.
Larsen, L. J., Wilby, A. & Stewart, C. (2010). Deep ocean survey and search using synthetic
aperture sonar, Proceedings of Oceans 2010 MTS/IEEE, Seattle, US.
Levanon, N. (1988). Radar Principles, Wiley Interscience.
Lurton, X. (2010). An Introduction to Underwater Acoustics: Principles and Applications, second
edn, Springer Praxis Publishing.
Manolakis, D. G., Ingle, V. K. & Kogon, S. M. (2000). Statistical and Adaptive Signal Processing,
McGraw-Hill.
Massonnet, D. & Souyris, J. (2008). Imaging with synthetic aperture radar, EFPL Press.
Medwin, H. &Clay, C. S. (1998). Fundamentals of Acoustical Oceanography, Academic Press, San
Diego, CA, USA.
Nielsen, R. O. (1991). Sonar signal processing, Artech House.
Oliver, C. & Quegan, S. (1998). Understanding Synthetic Aperture Radar Images, Artech house,
Inc.
Pinto, M. A. (2002). High resolution seaoor imaging with synthetic aperture sonar, IEEE
Oceanic Eng. Newsletter pp. 1520.
Pinto, M. A., Fohanno, F., Trmois, O. & Guyonic, S. (1997). Autofocusing a synthetic
aperture sonar using the temporal and spatial coherence of seaoor reverberation,
in O. Bergem & A. P. Lyons (eds), High Frequency Acoustics in Shallow Water,
SACLANTCEN Conference Proceedings, NATO SACLANT Undersea Research
Centre, La Spezia, Italy, pp. 417424.
Sb, T. O. (2010). Seaoor Depth Estimation by means of Interferometric Synthetic Aperture Sonar,
PhD thesis, University of Troms, Norway.
Sheriff, R. (1992). Synthetic aperture beamforming with automatic phase compensation for
high frequency sonars, Autonomous Underwater Vehicle Technology, 1992. AUV92.,
Proceedings of the 1992 Symposium on, IEEE, pp. 236245.
Soumekh, M. (1994). Fourier Array Imaging, Prentice Hall, Englewood Cliffs, NJ, USA.
Synnes, S. A., Hansen, R. E. & Sb, T. O. (2009). Assessment of shallow water
performance using interferometric sonar coherence, Proceedings of Underwater
Acoustic Measurements 2009, Nafplion, Greece.
Urick, R. J. (1983). Principles of Underwater Sound, Mcgraw-Hill Book Company.
Van Trees, H. L. (2002). Optimum Array Processing (Detection, Estimation, and Modulation Theory,
Part IV), Wiley-Interscience.
28 Sonar Systems
www.intechopen.com
Sonar Systems
Edited by Prof. Nikolai Kolev
ISBN 978-953-307-345-3
Hard cover, 322 pages
Publisher InTech
Published online 12, September, 2011
Published in print edition September, 2011
InTech Europe
University Campus STeP Ri
Slavka Krautzeka 83/A
51000 Rijeka, Croatia
Phone: +385 (51) 770 447
Fax: +385 (51) 686 166
www.intechopen.com
InTech China
Unit 405, Office Block, Hotel Equatorial Shanghai
No.65, Yan An Road (West), Shanghai, 200040, China
Phone: +86-21-62489820
Fax: +86-21-62489821
The book is an edited collection of research articles covering the current state of sonar systems, the signal
processing methods and their applications prepared by experts in the field. The first section is dedicated to the
theory and applications of innovative synthetic aperture, interferometric, multistatic sonars and modeling and
simulation. Special section in the book is dedicated to sonar signal processing methods covering: passive
sonar array beamforming, direction of arrival estimation, signal detection and classification using DEMON and
LOFAR principles, adaptive matched field signal processing. The image processing techniques include: image
denoising, detection and classification of artificial mine like objects and application of hidden Markov model
and artificial neural networks for signal classification. The biology applications include the analysis of biosonar
capabilities and underwater sound influence on human hearing. The marine science applications include fish
species target strength modeling, identification and discrimination from bottom scattering and pelagic biomass
neural network estimation methods. Marine geology has place in the book with geomorphological parameters
estimation from side scan sonar images. The book will be interesting not only for specialists in the area but
also for readers as a guide in sonar systems principles of operation, signal processing methods and marine
applications.
How to reference
In order to correctly reference this scholarly work, feel free to copy and paste the following:
Roy Edgar Hansen (2011). Introduction to Synthetic Aperture Sonar, Sonar Systems, Prof. Nikolai Kolev (Ed.),
ISBN: 978-953-307-345-3, InTech, Available from: http://www.intechopen.com/books/sonar-
systems/introduction-to-synthetic-aperture-sonar