Synthetic Aperture and Radar Imaging

Introduction and Context

Synthetic aperture radar (SAR) and inverse synthetic aperture radar (ISAR) are coherent imaging techniques that exploit variation in the radar-target geometry to obtain resolution finer than that provided by the physical antenna beamwidth. Unlike an optical camera, radar can image through darkness, cloud, haze, and, at suitable frequencies, some vegetation and soil. SAR is principally an active sensor carried by a moving platform and used to image stationary terrain. ISAR generally uses target motion, such as rotation or translation of a ship or aircraft, to synthesize an aperture while the radar may remain stationary or move independently.

A radar image is not merely a map of reflected power. It is the result of a measurement chain involving waveform design, coherent sampling, calibration, motion compensation, range compression, azimuth processing, geometric correction, and interpretation. Errors in phase or geometry can defocus an image even when the signal-to-noise ratio is high.

Learning Objectives

After completing this chapter, the reader should be able to:

1. Explain SAR and ISAR geometry and distinguish real-aperture from synthetic-aperture resolution.

2. Derive the principal range and azimuth resolution relationships.

3. Describe range migration and the major steps of SAR image formation and focusing.

4. Explain ISAR motion compensation and Doppler-based cross-range imaging.

5. Discuss polarimetric SAR, interferometric SAR, and moving-target indication.

6. Assess image quality using resolution, sidelobes, radiometric accuracy, phase quality, and geometric fidelity.

7. Select appropriate processing and validation practices for airborne, spaceborne, and ground-based systems.

1. Radar Imaging Foundations

A coherent radar records both amplitude and phase of the received complex signal. For a transmitted linear-FM pulse with bandwidth \(B\), ideal matched filtering gives approximate slant-range resolution

\[

\delta_r=\frac{c}{2B},

\]

where \(c\) is the speed of light. This expression assumes sufficient signal bandwidth, a point target, matched filtering, and negligible range-walk or sampling limitations. In a medium of refractive index close to one, electromagnetic propagation speed is approximately \(c\).

Ground-range resolution differs from slant-range resolution because of incidence geometry. If \(\theta_i\) is the incidence angle measured from the local vertical, then for a locally flat surface

\[

\delta_{gr}=\frac{\delta_r}{\sin\theta_i}.

\]

Thus, ground-range resolution becomes poorer at small incidence angles. Terrain slope and topography can invalidate this simple projection.

For a real-aperture antenna of azimuth length \(L_a\), wavelength \(\lambda\), and slant range \(R\), the beamwidth is approximately \(\lambda/L_a\), giving an azimuth resolution proportional to \(R\lambda/L_a\). Consequently, real-aperture resolution degrades with range. SAR removes this range dependence by coherently combining echoes collected over platform motion.

1.1 SAR Geometry

Consider a platform moving at velocity \(v\) along the azimuth direction. A stationary point target has closest-approach, or broadside, slant range \(R_0\). At slow time \(\eta\), measured relative to closest approach, the ideal broadside geometry is

\[

R(\eta)=\sqrt{R_0^2+v^2\eta^2}.

\]

For \(|v\eta|\ll R_0\), a second-order approximation is

\[

R(\eta)\approx R_0+\frac{v^2\eta^2}{2R_0}.

\]

The corresponding phase history is

\[

\phi(\eta)=-\frac{4\pi}{\lambda}R(\eta),

\]

for a monostatic radar, where the factor four arises from two-way propagation and phase expressed in radians. The changing range produces a Doppler history, approximately linear near broadside. This predictable phase curvature is the basis of azimuth focusing.

For an ideal stripmap SAR with uniform illumination and adequate coherent aperture, azimuth resolution is approximately

\[

\delta_a\approx\frac{L_a}{2},

\]

where \(L_a\) is the physical antenna dimension in azimuth. This result assumes broadside stripmap operation and ideal processing; actual resolution is affected by weighting, squint, bandwidth, motion errors, and processing aperture length. In spotlight SAR, beam steering keeps a scene patch illuminated for a longer interval, increasing synthetic-aperture angle and generally improving azimuth resolution, at the cost of more complex geometry and processing.

1.2 ISAR Geometry

In ISAR, target motion creates the synthetic aperture. A translating target produces a common Doppler shift, while rotation creates differential Doppler frequencies across scatterers. For a target rotating with angular rate \(\omega\), a scatterer at cross-range coordinate \(x\), under a small-angle and narrow-angle approximation, its Doppler frequency is approximately

\[

f_D\approx \frac{2\omega x}{\lambda}.

\]

Hence, the cross-range coordinate estimated from Doppler is

\[

x\approx\frac{\lambda f_D}{2\omega}.

\]

The relation assumes monostatic operation, known or estimated rotation rate, small rotation over the coherent processing interval, and a geometry in which the relevant motion is rotational. If the target motion is not modeled correctly, ISAR images may be shifted, smeared, or geometrically distorted.

Unlike SAR, ISAR generally has no fixed ground coordinate system. Its image coordinates are target-relative and depend on motion compensation. The same target can produce different images at different aspect angles or times.

2. Range Migration and the SAR Signal Model

As the platform passes a target, the target's echo moves through range cells because \(R(\eta)\) changes. This is range cell migration (RCM). It occurs even when the target is stationary and is not an error in the radar; it is a geometric effect. If uncorrected, energy from one scatterer is spread across several range bins and cannot be coherently integrated in azimuth.

For a sampled range-compressed data matrix \(s(\tau,\eta)\), \(\tau\) denotes fast time and \(\eta\) slow time. A simplified point-target model is

\[

s(\tau,\eta)=A\,p\left(\tau-\frac{2R(\eta)}{c}\right)

\exp\left[-j\frac{4\pi}{\lambda}R(\eta)\right],

\]

where \(A\) is a complex scattering coefficient and \(p(\cdot)\) is the range-compressed pulse response. This model neglects antenna pattern, propagation loss, noise, range-spreading loss, and target scintillation.

RCM can be corrected in several ways:

• Range migration algorithm (RMA): uses a two-dimensional frequency-domain transformation and Stolt interpolation to map the measured spectrum to a uniform wavenumber grid.
• Range-Doppler algorithm (RDA): estimates migration as a function of Doppler and interpolates or shifts data before azimuth compression.
• Chirp scaling: applies phase operations and scaling to correct migration and range-dependent azimuth modulation without explicit interpolation in some implementations.
• Backprojection: evaluates the measured data directly at hypothesized pixel locations using the predicted propagation delay. It is flexible for wide-angle, spotlight, and irregular-motion systems but computationally expensive.

The selected method depends on geometry, bandwidth, squint angle, motion accuracy, computational resources, and required image fidelity.

3. SAR Image Formation and Focusing

A typical SAR processing chain contains the following stages:

1. Data conditioning: remove corrupt records, correct timing, apply antenna and receiver calibration, and organize pulses by precise navigation time.

2. Range compression: match-filter each pulse to obtain fine delay resolution. Windowing reduces sidelobes but broadens the mainlobe.

3. Motion compensation: correct deviations from the assumed flight path. This includes phase errors, range errors, and sometimes antenna-pointing errors.

4. Range migration correction: align the target's energy with its changing range history.

5. Azimuth compression: apply the conjugate of the expected phase history or an equivalent matched filter.

6. Radiometric and geometric correction: compensate for known system gains, antenna patterns, spreading loss, terrain projection, and map geometry.

7. Image formation and product generation: create complex, detected, multilooked, geocoded, or interferometric products.

3.1 Range Compression

For a transmitted waveform \(s_t(t)\), the ideal matched filter has impulse response proportional to \(s_t^*(-t)\). Matched filtering maximizes peak signal-to-noise ratio in additive white noise when the signal model is correct. A weighting window may reduce peak sidelobe ratio (PSLR) and integrated sidelobe ratio (ISLR), but it sacrifices some resolution and signal-to-noise ratio.

For a rectangular spectrum, the compressed response has a sinc-like mainlobe. Practical systems use window functions such as Taylor, Kaiser, or raised-cosine families to control sidelobes. The window must be documented because resolution values without a specified criterion—such as 3-dB width or null-to-null width—are not unambiguous.

3.2 Azimuth Compression

A stationary point target produces a quadratic phase history near broadside. Azimuth focusing applies a matched filter to that history. In frequency-domain processors, the data are often transformed in slow time, corrected for range dependence, and interpolated to a wavenumber grid. In time-domain backprojection, each pixel is formed by delaying and phase-aligning samples according to its predicted range:

\[

I(\mathbf{x})=\sum_n w_n\,s_n\left(\frac{2R_n(\mathbf{x})}{c}\right)

\exp\left(j\frac{4\pi}{\lambda}R_n(\mathbf{x})\right),

\]

where \(I(\mathbf{x})\) is the complex image value at pixel position \(\mathbf{x}\), \(R_n(\mathbf{x})\) is the transmitter-to-target-to-receiver equivalent range for pulse \(n\), \(w_n\) is a weighting factor, and interpolation is implied in evaluating the signal at the required delay. For a monostatic radar, \(R_n\) is the one-way geometric range and the phase term represents two-way propagation. Backprojection assumes a sufficiently accurate waveform and motion model.

3.3 Motion Compensation and Autofocus

A phase error \(\Delta R\) produces approximately

\[

\Delta\phi=\frac{4\pi}{\lambda}\Delta R.

\]

At X-band, even centimetre-scale path errors can be significant. Motion compensation uses inertial navigation, GPS/GNSS, surveyed control points, autofocus, or combinations of these. Autofocus estimates residual phase errors from the radar data itself. Prominent methods include phase-gradient autofocus and map-drift techniques. Autofocus can improve focus when navigation is imperfect, but it may fail in scenes with low contrast, distributed scattering, strong ambiguity, or rapidly changing targets. It can also absorb genuine scene phase into an erroneous correction.

4. ISAR Processing

ISAR processing begins with range alignment, because target translation causes the entire range profile to move. Common alignment methods include correlation-based shifts, prominent-point tracking, and entropy-minimizing registration. After alignment, phase compensation removes residual translational motion. The aligned data are divided into coherent processing intervals (CPIs), and a slow-time Fourier transform or a more advanced time-frequency method produces cross-range information.

The fundamental trade-off is that a longer CPI improves Doppler resolution but assumes more nearly constant motion. If the target rotates non-uniformly, range-Doppler images can smear. Motion compensation may use autofocus, estimated target kinematics, polynomial phase models, or external tracking. For maneuvering aircraft and ships, short-time Fourier transforms, Doppler-rate estimation, and sparse or parametric reconstruction can be useful, although they introduce model dependence and validation requirements.

ISAR images are therefore best interpreted with aspect angle, CPI duration, carrier frequency, bandwidth, and motion-compensation quality recorded as metadata.

5. Polarimetric SAR

A polarimetric SAR transmits and receives multiple polarization states, commonly horizontal and vertical. The scattering response can be represented by a Sinclair scattering matrix

\[

\mathbf{S}=\begin{bmatrix}S_{HH}&S_{HV}\\S_{VH}&S_{VV}\end{bmatrix}.

\]

For reciprocal natural media, often \(S_{HV}\approx S_{VH}\), subject to calibration and measurement conditions. The matrix elements describe complex scattering between transmit and receive polarization channels. Products may be expressed as a scattering vector or covariance/coherency matrix, enabling decomposition into mechanisms such as surface, double-bounce, and volume scattering.

Polarimetric interpretation is not unique: decomposition results depend on assumptions about reflection symmetry, orientation, noise, and model validity. Applications include crop structure estimation, forest characterization, wetland mapping, urban mapping, sea-ice analysis, and soil-moisture retrieval. Polarimetric calibration must account for channel imbalance, cross-talk, phase bias, and antenna distortion.

6. Interferometric SAR

InSAR uses the phase difference between two coherent SAR observations. For two complex pixels \(z_1\) and \(z_2\), the interferometric phase is commonly represented by

\[

\phi_{int}=\arg\left(z_1z_2^*\right).

\]

The phase contains contributions from topography, surface deformation, orbit geometry, atmospheric propagation, and noise. Phase is measured modulo \(2\pi\), so conversion to elevation or displacement requires phase unwrapping and an appropriate geometric model.

The interferometric coherence is a normalized measure of similarity, often estimated over a local window. One common sample estimator is

\[

\hat\gamma=\frac{|\sum_{k=1}^{N}z_{1k}z_{2k}^*|}{\sqrt{\sum_{k=1}^{N}|z_{1k}|^2\sum_{k=1}^{N}|z_{2k}|^2}},

\]

where \(N\) samples are used. Its value lies between zero and one in the ideal normalized case. Coherence decreases with temporal change, thermal noise, geometric decorrelation, volume scattering, processing mismatch, and atmospheric differences.

Differential InSAR removes a reference topographic phase to estimate deformation. Persistent-scatterer and small-baseline methods use multiple acquisitions to improve robustness in suitable environments. InSAR products require careful orbit refinement, coregistration, baseline control, atmospheric screening, phase unwrapping, and uncertainty reporting.

7. Moving-Target Indication and Moving-Target Imaging

A conventional SAR image assumes stationary scatterers. A moving target violates this assumption and may appear displaced, defocused, or split because its range and Doppler histories differ from those expected by the processor. This is useful for detection but problematic for faithful imaging.

Moving-target indication (MTI) exploits temporal or Doppler differences to suppress stationary clutter and detect moving objects. In airborne SAR, displaced phase-centre methods, along-track interferometry, space-time adaptive processing, and multi-channel receivers can estimate radial velocity and suppress clutter. Moving-target imaging attempts to compensate the target's motion and form a focused image, often using sequential estimation of range, velocity, acceleration, and rotational motion.

A key limitation is that radial velocity is not always sufficient to determine the full three-dimensional velocity vector. Detection and imaging performance depend on clutter stationarity, channel calibration, revisit interval, pulse repetition frequency, ambiguities, and target fluctuation.

8. Image-Quality Assessment

Image quality should be evaluated using controlled point targets, distributed scenes, calibration sites, and application-specific measures.

Spatial and Radiometric Measures

• Resolution: usually reported separately in slant range, ground range, and azimuth, with the width criterion specified.
• PSLR: ratio of the largest sidelobe peak to the mainlobe peak, normally reported in decibels.
• ISLR: ratio of sidelobe energy to mainlobe energy; it captures distributed sidelobe contamination.
• Ambiguity ratios: quantify unwanted responses from range and azimuth ambiguities.
• Radiometric accuracy: agreement between measured backscatter and a calibrated reference, often expressed in decibels.
• Equivalent number of looks: indicates speckle reduction and radiometric variance in multilooked products.

For fully developed single-look speckle, intensity has high relative variance; multilooking reduces variance approximately as the number of statistically independent looks increases, although spatial resolution is reduced. Independence is an assumption and may not hold for strongly correlated looks.

Phase and Geometric Measures

Important phase metrics include coherence, phase standard deviation, phase bias, and unwrapping error rate. Geometric metrics include absolute location error, relative geolocation error, terrain-induced layover, foreshortening, and shadow extent. A visually sharp image may still be poorly geolocated or radiometrically biased.

Quality assessment should use a traceable test plan: define the product type, processing parameters, reference data, acceptance thresholds, uncertainty budget, and environmental conditions. Point-target tests should include known corner reflectors or transponders where possible.

9. Airborne, Spaceborne, and Ground-Based Systems

Airborne SAR offers flexible incidence angle, high resolution, and repeat access, but platform vibration, navigation errors, turbulence, and antenna motion complicate focusing. Accurate inertial/GNSS integration and frequent calibration are essential.

Spaceborne SAR provides systematic wide-area coverage and repeat-pass interferometry. It must manage orbital geometry, ionospheric and tropospheric effects, data volume, Doppler centroid variation, Earth curvature, and long-term calibration stability. Acquisition planning strongly influences InSAR coherence and shadow or layover conditions.

Ground-based SAR places the radar on a rail, tower, or fixed site and can monitor slopes, structures, glaciers, or industrial assets. It provides highly repeatable geometry but is vulnerable to atmospheric phase fluctuations, line-of-sight limitations, occlusion, and local vibration. Ground-based systems are particularly valuable for near-real-time deformation monitoring when a stable observation geometry is available.

10. Illustrative Case: Interferometric Monitoring of a Subsiding Urban Basin

Background

Consider an illustrative spaceborne C-band SAR mission monitoring an urban basin containing roads, high-rise buildings, agricultural land, and a groundwater-extraction zone. The objective is to detect millimetre-to-centimetre-scale deformation over several months.

Problem

A single amplitude image shows buildings and roads clearly but cannot reliably distinguish deformation from terrain height, atmospheric delay, or speckle. Agricultural fields lose coherence between acquisitions.

Analysis and Approach

The processing team selects acquisitions with compatible viewing geometry and small perpendicular baselines. Images are precisely coregistered, interferograms are formed using \(z_1z_2^*\), and a digital elevation model is used to remove the topographic contribution. Persistent urban scatterers are identified using amplitude stability and temporal coherence. Atmospheric artefacts are reduced using spatial-temporal filtering, while phase unwrapping is restricted to connected high-coherence regions. GNSS stations provide independent validation.

A line-of-sight displacement time series is then estimated. The result is reported in the radar line-of-sight direction, not automatically as vertical motion. Conversion to vertical displacement requires assumptions about horizontal motion and viewing geometry.

Outcome

The analysis identifies a coherent subsidence bowl centred on the extraction zone, while low-coherence agricultural areas are excluded or reported with larger uncertainty. Agreement with GNSS at stable urban sites supports the result, but residual atmospheric error remains a limitation.

Learning Points

The case demonstrates that InSAR is a multi-stage measurement problem rather than a simple image subtraction. Acquisition design, coherence management, orbit accuracy, atmospheric treatment, phase unwrapping, independent validation, and explicit uncertainty are all necessary.

11. Limitations, Governance, and Contemporary Developments

Fundamental limitations include speckle, shadow, layover, foreshortening, temporal decorrelation, atmospheric propagation, model mismatch, limited observation geometry, and ambiguity. Resolution cannot be improved indefinitely by increasing computation; it is constrained by bandwidth, coherent aperture, sampling, motion knowledge, and signal-to-noise ratio.

Radar imagery can reveal sensitive information about infrastructure, individuals, agricultural activity, and military assets. Professional practice therefore requires lawful data access, privacy-aware dissemination, secure storage, export-control compliance where applicable, and clear communication of uncertainty. Automated target recognition should not be treated as infallible: training-data bias, domain shift, adversarial conditions, and false alarms must be evaluated. Human review and audit trails are important in high-consequence decisions.

Contemporary research includes neural and model-based autofocus, learned despeckling, compressed sensing, distributed and multistatic SAR, geosynchronous and low-cost small-satellite constellations, 3-D tomography, cognitive waveform selection, and joint radar–optical fusion. AI methods are most reliable when constrained by radar physics and evaluated on independent scenes. A learned enhancement that changes phase, radiometry, or weak-target statistics without disclosure may be unsuitable for scientific or legal applications.

Best Practices

1. Specify geometry, coordinate conventions, waveform, sampling, windowing, and resolution criteria.

2. Preserve complex data and metadata whenever phase, polarimetry, or future reprocessing matters.

3. Calibrate amplitude and phase with traceable references.

4. Use navigation data and radar-derived autofocus together, while checking that autofocus has not removed real scene information.

5. Select processing algorithms according to squint, bandwidth, aperture, motion, and computational constraints.

6. Report sidelobes, ambiguities, coherence, radiometric uncertainty, and geolocation error—not only visual sharpness.

7. Validate deformation, target motion, or classification products using independent observations.

8. Document limitations, exclusions, and uncertainty in all operational products.

Chapter Summary

SAR forms high-resolution images of stationary scenes by coherently combining echoes collected along a moving aperture; ISAR uses target motion to generate cross-range resolution. Range resolution is governed mainly by waveform bandwidth, whereas SAR azimuth resolution arises from coherent aperture processing. A target's changing slant range causes range migration, which must be corrected before accurate azimuth focusing. Motion compensation and autofocus are central because phase errors measured in fractions of a wavelength can defocus an image.

ISAR requires range alignment and translational phase compensation before Doppler-based cross-range formation. Polarimetric SAR uses multiple polarization channels to infer scattering mechanisms, while InSAR uses interferometric phase to estimate topography and deformation, subject to coherence, atmospheric, orbit, and unwrapping limitations. Moving targets can be displaced or defocused in conventional SAR and require dedicated MTI or moving-target imaging methods. Image quality includes spatial resolution, sidelobes, ambiguities, radiometric accuracy, phase quality, coherence, and geolocation. Airborne, spaceborne, and ground-based systems offer different geometry, stability, coverage, and calibration trade-offs. Physics-constrained AI, multi-sensor systems, and large satellite constellations are important contemporary directions, but transparent validation and uncertainty remain essential.

Key Terms

SAR: Synthetic aperture radar imaging of principally stationary scenes from a moving platform.

ISAR: Inverse synthetic aperture radar imaging using target motion to synthesize an aperture.

Slant range: Direct radar line-of-sight distance to a scattering point.

Ground range: Distance projected onto the reference ground surface.

Azimuth: Along-track or cross-range direction in radar imaging.

Range migration: Movement of a target response through range cells during coherent processing.

Range compression: Matched filtering that converts waveform bandwidth into fine delay resolution.

Azimuth compression: Coherent processing of Doppler or phase history to focus cross-range position.

Motion compensation: Correction of platform or target motion relative to the assumed imaging geometry.

Autofocus: Data-driven estimation of residual phase errors.

Polarimetric SAR: SAR using multiple transmit and receive polarization channels.

Scattering matrix: Complex matrix describing polarization-dependent scattering coefficients.

InSAR: Interferometric SAR using phase differences between coherent observations.

Coherence: Normalized similarity of two complex SAR observations.

MTI: Moving-target indication based on temporal, Doppler, or spatial-channel differences.

PSLR: Peak sidelobe ratio relative to the mainlobe peak.

ISLR: Integrated sidelobe ratio based on sidelobe and mainlobe energy.

Layover: Geometric overlap caused when elevated terrain returns before lower terrain.

Speckle: Coherent interference fluctuation in radar imagery.

Geocoding: Transformation of radar image coordinates into map or geographic coordinates.

Review Questions

Short Answer Questions

1. Why does SAR provide better azimuth resolution than a real-aperture radar at long range?

2. Define range cell migration and explain why it must be corrected.

3. State two differences between SAR and ISAR geometry.

4. What physical information can polarimetric SAR provide beyond single-channel intensity?

5. List four factors that reduce InSAR coherence.

Long Answer Questions

1. Derive the broadside SAR range history and explain how its phase curvature enables azimuth focusing.

2. Describe a complete range-Doppler or backprojection SAR processing chain, including assumptions and limitations.

3. Explain ISAR range alignment, translational motion compensation, and cross-range formation.

4. Discuss the principles and practical limitations of polarimetric SAR and InSAR.

5. Develop an image-quality assessment plan for an airborne SAR calibration campaign.

Analytical / Application Questions

1. A SAR uses a 30 MHz transmitted bandwidth. Calculate its ideal slant-range resolution and discuss why the operational value may differ.

2. An ISAR observes a target at wavelength 0.03 m. If a scatterer has Doppler frequency 2 kHz and the target rotation rate is 0.5 rad/s, estimate its cross-range coordinate using the small-angle model and state the model assumptions.

3. Design an acquisition and validation strategy to monitor deformation near a city, considering coherence, baseline, atmospheric effects, phase unwrapping, and independent measurements.

Multiple Choice Questions with Answers

Question 1

For an ideal radar waveform with bandwidth B, the slant-range resolution is approximately:

A. cB/2

B. 2c/B

C. c/(2B)

D. B/(2c)

Correct Answer: C. c/(2B)

Question 2

In broadside SAR geometry, the target range history is approximately:

A. R0 + vη

B. sqrt(R0² + v²η²)

C. R0 − v²η²/(2R0)

D. R0/vη

Correct Answer: B. sqrt(R0² + v²η²)

Question 3

Range cell migration is caused primarily by:

A. Changing target-platform range during coherent integration

B. Receiver thermal noise only

C. Polarization mismatch only

D. Digital image display scaling

Correct Answer: A. Changing target-platform range during coherent integration

Question 4

Which operation primarily produces fine delay resolution from a transmitted chirp?

A. Azimuth FFT only

B. Matched filtering or range compression

C. Phase unwrapping

D. Multilooking

Correct Answer: B. Matched filtering or range compression

Question 5

In the stated ISAR small-angle model, cross-range coordinate is proportional to:

A. λfD/(2ω)

B. 2ω/(λfD)

C. 2λfDω

D. fD/(2λω)

Correct Answer: A. λfD/(2ω)

Question 6

The interferometric phase from two complex pixels z1 and z2 is commonly calculated as:

A. |z1| + |z2|

B. arg(z1z2*)

C. z1/z2 without phase extraction

D. Re(z1)Re(z2)

Correct Answer: B. arg(z1z2*)

Question 7

Which phenomenon causes elevated terrain to overlap lower terrain in radar geometry?

A. Speckle

B. Layover

C. Range compression

D. Cross-talk

Correct Answer: B. Layover

Question 8

A major purpose of SAR autofocus is to:

A. Estimate and correct residual phase errors

B. Increase transmitted carrier wavelength

C. Remove all physical sidelobes without resolution trade-off

D. Convert slant range into optical color

Correct Answer: A. Estimate and correct residual phase errors

Question 9

Which metric directly describes the largest sidelobe peak relative to the mainlobe peak?

A. Coherence

B. Equivalent number of looks

C. PSLR

D. Ground-range scale

Correct Answer: C. PSLR

Question 10

In repeat-pass InSAR, low coherence can result from:

A. Temporal decorrelation and atmospheric differences

B. Using complex data

C. A perfectly stable point target

D. Matched filtering alone

Correct Answer: A. Temporal decorrelation and atmospheric differences