High resolution radar

1
High resolution radar technology
for environmental applications
Dr Richard Holliday Dr Duncan Wynn Matt-Rhys
Roberts
2
Content

• Introduction
• Survey methods - manual - LIDAR/hyper-spectral
  , thermal imaging - RADAR/ real-beam
  mapping/SAR, interferometric SAR
• Overview of high resolution radar
• Environmental applications - mapping
• - remote sensing and surveillance - pollution
  monitoring

3
Introduction

• A major threat to global stability is the
  change in the Earths climate
• Extremes of heat and drought, storms, wind,
  rain and more intense cold
• Unpredictable environmental behaviour -
  Temperature rises are likely to be non-uniform
  across the globe
• Uncertainty of the impact is incorporated into
  long-term national and international
  decision-making - reflected in environmental
  standards and targets including
• - protection of people, homes and business from
  risk of flood - ensure availability of suitable
  water for drinking and bathing - prevention of
  destruction of natural habitats and extinction of
  animal species - There is a very wide breadth
  of environmental issues ..

4
Climate change .. evidence
Winter and Summer rainfall rates 1766 to 1991
5
Introduction

• Better understanding of the complex interaction
  between the Earths surface and atmosphere is
  essential
• Accurate descriptions of local and regional
  surface features and associated phenomena with
  timely monitoring are vital

6
Survey methods

• The quality of surface maps is gauged primarily
  by the ability of the survey method and sensor
  to resolve closely spaced features . generally
  defined in terms of resolution Historically,
  surface feature (map or topographic) data has
  been obtained by manual survey methods
• Autonomous - surface based - airborne

7
Survey methods Manual

• Wide variety of methods - traditional survey,
  laser based, GPS based - passive samplers,
  ultrasonic gauges etc
• Labour intensive
• Time consuming untimely data/latency ?
• Poor spatial coverage
• Low spatial resolution point
  sensors/measurements
• Invasive perturbation of measurement
  environment
• Limited by accessibility of environment -
  ad-hoc methods such as General Quality Assessment
  (GQA for rivers)
• Inaccurate and inconsistent requires
  accreditation and standardisation
• Autonomous survey solutions are required .

8
Survey methods Automatic flow monitoring

• Time consuming untimely data/latency ?
• Poor spatial coverage
• Low spatial resolution point
  sensors/measurements
• Invasive perturbation of measurement
  environment
• Limited by accessibility of environment
• Inconsistent requires accreditation and
  standardisation

9
Survey methods Airborne

• LIght Detection And Ranging (LIDAR)
• Wide area coverage typically 600 m swathe
  width at 800m altitude
• Measurements every 2m with resolution between
  1m and 10m

• Compact Airborne Spectral Imager (CASI)
• Thermal imager / daylight camera

10
Disadvantages of LIDAR

• High operating costs (gt 10k / hour)
• Not all-weather performance - ineffective
  during heavy rain and/or low cloud/mist -
  degraded at high Sun angles and reflections
• Latency data not processed locally (Coventry
  airport/OS)
• Unreliable for water depth (lt 2m) and
  breaking/turbulent waves
• Lack of foliage/vegetation penetration

11
Survey methods Real-beam mapping radar

• Wide area coverage surface, airborne /
  satellite borne
• All-weather capability

12
Real-beam mapping radar
Ludlow
X-band (10 GHz) 1 beamwidth 100m range resolution
Hereford
13
Real-beam mapping radar
X-band (10 GHz) radar and video based
measurements from a traffic scene (circa 1968)
77 GHz radar and video based measurements from a
traffic scene (circa 1998)
14
Survey methods Synthetic Aperture Radar (SAR)

• Wide area coverage airborne / satellite borne
  
• All-weather capability
• Costs - Capital costs typically gt100 k
  excl. aircraft installation and maintenance gt
  1m for satellite payloads - Operating costs
  typically gt 10k / hour

15
Synthetic aperture radar (SAR) Airborne
X-band (10 GHz) 1 beamwidth 0.3m range resolution
16
Synthetic aperture radar (SAR) Satellite-borne
TOPEX/POSEIDON ocean topography project
17
Disadvantages of microwave real-beam and SAR radar

• SAR imagery is prone to distortions,
  obscuration and RF interference
• Lack of foliage/vegetation and surface
  penetration
• Speckle cancellation is required

• Latency data not processed locally - Time
  dependent phenomena are not imaged
• Moving object detection with MTI radar modes
• Lack of availability
• Ownership of data
• Cost

18
Distortion of SAR imagery
Optical and SAR view of same airfield
19
Obscuration of SAR imagery
20
Effect of RF interference upon SAR imagery
21
High resolution radar for environmental
applications
22
High resolution radar operating modes

• High resolution mapping
• Velocimetry
• Target classification - Polarimetry -
  Non-cooperative Target Recognition (NCTR)
• Bathymetry

23
Outline radar specification (target)
Modes High resolution surface mapping (2D and
3D) Velocimetry max. velocity lt 15 m/s (33
mph / 54 km/hr) Bathymetry lt 2m water
depth Classification / non-cooperative target
recognition surface texture, birds, insects,
humans Coverage 360 azimuth, 40
elevation Resolution spatial lt 0.03m
(range), lt 0.05m (azimuth) at 300m
velocity lt 0.003 m/s or 0.01 max. velocity
Sensitivity gt 10 dB SNR at 5 km against 1m2
non-fluctuating, stationary target Polarisation
Fully polarimetric Physical weight
under 80 kg maximum size 1m x 1m x
1m Integral navigation unit including GPS and
INS
24
High resolution radar

• Angular resolution (related to antenna
  beamwidth and physical size) improves at
  millimetric/sub-millimetric wavelengths but
  atmospheric attenuation is worse

3dB beamwidth of 1m antenna is typically 1 at
94 GHz
25
High resolution radar Atmospheric attenuation
150 GHz
94 GHz
windows
26
High resolution radar

• Radar technology at millimetric/sub-millimetric
  wavelengths is more affordable
• Hardware at millimetric/sub-millimetric
  wavelengths is physically small . portable and
  hand-held equipment is convenient

Slot antenna
27
High resolution radar

• Portable high resolution radar is able to
  exploit improved geometry to overcome distortion
  and obscuration
• Multiple radars can be networked for
  simultaneous coverage
• Surface based measurements underneath
  tree-canopies overcome foliage and vegetation
  shielding
• RF electromagnetic spectrum is sparsely
  occupied at millimetric/sub-millimetric
  wavelengths very low probability of RF
  interference

28
High resolution radar

• 10 dB SNR against 1m2 non-fluctuating target at
  5.5 km with RF transmitter input power 50 mW
• gt5 RF bandwidth at 94 GHz for better than 0.5
  m range resolution
• Multi-frequency RF transmission to mitigate
  multipath

Typical radar performance Target RCS 1m2, radar
height 1m
29
Velocimetry

• Determination of velocity vector of scattering
  centre within resolution cell by separate or
  multiple radar measurements

V(n)
Resolution cell n
VD(n)
VE(n)
VA
VN(n)
?A
Radar position B
Reference
?A
Radar position A
Velocity vector of resolution cell, n
Measured radial velocity VA of resolution cell n,
projected at angle ?A ?A
V(n)VN(n)VE(n)VD(n)
30
Classification Polarimetry

• Radar returns are polarisation dependent
  offers best opportunity for classification
• A fully polarimetric high resolution radar will
  record four separate complex reflectivities for
  the same scene
• Radar image texture can be interpreted as
  environmental features such as trees, hedges,
  fields (bare soil and vegetation), hills,
  ditches, bridges, shadows, buildings, roads,
  rivers and fences
• Texture can be measured by RCS of clutter CDF
  and moments of distribution
• Discrimination between type of vegetation is
  possible
• Close interaction with surface with topology
  data - roughness and soil moisture retrieval
  are possible

31
Classification Non-cooperative Target
Recognition (NCTR)

• Radar cross-section is enhanced by large number
  of point scatterers at millimetric wavelengths
• Scattering centres are discernible from
  measurements with high spatial and temporal
  resolution
• RCS is dependent upon aspect angle to radar
• Target motion compensation and RCS database
  collation is effective
• Target classification algorithms are widely
  available, such as template matching
• Target motion can be exploited with Inverse
  Synthetic Aperture (ISAR) imaging

32
Bathymetry

• Water depth can be determined from measurements
  of surface water wavelengths, directions and
  velocities requires high spatial and
  temporal resolution

Y
X
33
Bathymetry

• Combined with velocimetry

34
High resolution radar architecture
35
FMCW radar
Foster scanner antenna
94 GHz FMCW radar
36
High resolution radar
Typical antenna radiation pattern 94 GHz
co-polar 3 dB beamwidth 0.1o
37
Super-resolution

• Resolution limit defined by classical Rayleigh
  criterion can be improved by a factor of
  between 2 and 3
• Facilitated by - rapid update rates -
  stable antenna scanning (gt 400 rpm) - advanced
  signal processing - MUSIC / IMP / MAP
  algorithms

38
Super-resolution
39
Environmental applications Mapping

• Land topography, elevation modelling and height
  contour plots
• Land cover classification, landscape and
  habitat survey - model validation to link land
  use with soil type and erosion prediction - crop
  and animal stock monitoring and grazing
  management for sustainable farming
• Road survey and highway mapping
• Flood plain surveys (with/without flooding)
• Surface water mapping including rivers,
  streams, canals, sewers - water flow rates,
  direction and depth monitoring, bathymetric
  mapping - identification and tracking of
  dissimilar bodies of water - mapping of mixing
  zones, outfalls and rivers
• Inter-tidal vegetation mapping
• Coastal erosion, tidal action and geomorphology

40
Environmental applications Remote sensing and
surveillance

• Monitoring of water abstraction and discharges
  for review of permits
• Meteorological measurements of fog, cloud and
  precipitation including ice particle size,
  internal circulations in fair weather, drop
  size distributions in rain and drizzle in
  shallow continental stratus cloud
• Wind speed and direction, storm direction,
  cloud base and cloud top detection
• Bird, insect and wildlife monitoring as a
  measure of water and environmental quality in
  conjunction with control measures
• Ecological surveys including number and
  distribution of specific wildlife (bird) species
  
• Personnel and livestock location and
  identification
• Fire and smoke monitoring of forested areas

41
Environmental applications Pollution detection

• Water detritus content warning - litter, oil,
  surface scum, foam, sewage fungus, ochre, buoys,
  floats, jetsom/flotsom
• Pollution monitoring of water reservoirs
• Air quality monitoring - Open-path monitoring
  from 10 m to over 1 km - detection and tracking
  of aerosols, airborne particulates (smoke),
  atmospheric absorption - detection of gaseous
  compounds in ambient air