NEW THREATS
                  NEW THINKING. ®

  1.    The ABC’s of Radar
  2.    Radar Terminology for the Layman
  3.    Ground Surveillance Radar i...
THE ABC’s OF RADAR                                                                                  TECHNICAL
Highly reflective targets will return more energy back to the radar and will be more easily
detected. The reflectivity or ...
In radars that are pulsed, the pulse width is the duration of the “on” time of the pulse (in
micro or nano sec...
RADAR SPECTRAL BANDS                                                                                  TECHNICAL
The table below presents the various frequency bands used for radar with their most common

Frequency Band  ...
FREQUENCY MODULATED CONTINUOUS                                                                     TECHNICAL
each radar, and therefore, can walk at a higher speed than the velocity threshold, and
still not be detected. This deficie...
RADAR PERFORMANCE IN RAIN                                                                                                 ...
Figure 2 and Figure 3 show a comparison of the backscatter for two representative ground
based radar systems. Figure 2 for...
RADAR PERFORMANCE PARAMETERS                                                                                   TECHNICAL
Radar revisit time is the time it takes for the radar to complete its search for targets and return to
begin ...
EARLY HISTORY OF RADAR                                                                                       TECHNICAL
WHAT TO CONSIDER WHEN DESIGNING                                                                                 TECHNICAL
The table below presents the various frequency bands used for radar with their most common

Frequency Band  ...
SOME COMMON IR APPLICATIONS                                                                                 TECHNICAL
At these long wavelengths, infrared radiation be-
haves differently from visible light. Glass is opaque
in the LWIR band, ...
MRTD is measured by also looking at a four bar targets of varying spatial...
RANGE PERFORMANCE MODELING                                                                                                ...
UNCOOLED IR DETECTORS                                                                                      TECHNICAL
ASi detectors also employ a TE device to stabilize their temperature. Similar to vanadium oxide, design-
ers have develope...
I Cx Surveillance Tech Notes
I Cx Surveillance Tech Notes
Upcoming SlideShare
Loading in...5

I Cx Surveillance Tech Notes


Published on

A collection of white papers and tech notes on integrated surveillance systems

  • Be the first to comment

No Downloads
Total Views
On Slideshare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

I Cx Surveillance Tech Notes

  2. 2. TABLE OF CONTENTS 1. The ABC’s of Radar 2. Radar Terminology for the Layman 3. Ground Surveillance Radar in Security Applications 4. Radar Spectral Bands 5. Frequency Modulated Continous Wave (FMCW) Radars 6. Radar Performance in Rain 7. Radar Performance Parameters 8. Early History of Radar 9. What to Consider When Designing a Security System with Radars 10. The Power of Slew-to-Cue Surveillance Capability 11. Multilayered Perimeter Surveillance for Airports 12. Common IR Applications 13. IR Spectral Bands and Performance 14. IR Technology Parameters and Tradeoffs 15. Range Performance Modeling 16. Uncooled IR Detectors 17. Advantages of Technologies vs. Border Fences
  3. 3. THE ABC’s OF RADAR TECHNICAL NOTES The word radar is an acronym taken from RAdio Detection And Ranging. A radar is a device that transmits and receives electromagnetic energy as radio waves to detect and esti- mate distance to and/or velocity of an object. It generally includes a transmitter emitting ra- dio waves toward an object and a receiver detecting the radio energy reflected by the object. Distance is estimated by measuring the propagation time to and from the object. Velocity 1 is estimated by measuring the Doppler frequency shift. This so-called Doppler effect occurs Pierre when the frequency of a wave changes and results from the relative velocity between an Poitevin object and the radar. The velocity can also be estimated by observing the rate at which the range of the target changes. Transmitting and receiving radio waves requires the use of an antenna that acts as an in- terface between the electrical circuits and free space where the radio waves are radiated towards potential targets. Antenna characteristics are critical in shaping performance of a radar system, and high gain antennas yield more sensitive systems. The angular position of targets can be found with radar systems using directive antennas where the antenna scan angle is recorded for each target detected. Radar receivers have to deal with very low level signals because of the power density de- creasing with the distance traveled by the radio waves. When received, signals are amplified, conditioned, digitally sampled and fed to a radar processor that applies specific processing techniques to optimally extract targets. To be detected by the radar, targets need to generate strong enough signals (by reflection of transmitted signal) to rise above the noise. Signal to Noise Ratio (SNR) is therefore an important measurement that determines whether targets can be seen at all by the radar. In general, SNRs in excess of 10 ~12 dB are required to assure reliable detection with an acceptable false alarm rate. False alarms are undesired detections that can occur when no target is present. The SNR for a received signal is given by the classic radar equation: SNR = Pt Gt Gr λ2 σ (4π)3 r4 k T Bn Where • SNR = Signal to Noise Ratio • Pt = transmitter power • Gt = gain of the transmitting antenna • Gr = gain of the receiving antenna • λ = wavelength of radio wave • σ = radar cross section of an object • k =Boltzmann’s constant • T = receiver noise equivalent temperature • Bn = processing noise bandwith • r = distance from the transmitter to the target As can be seen from the above equation, more transmitter power and more antenna gain directly improve signal to noise ratio and capability to detect targets. The detectability of an object by the radar also greatly depends on the distance between the object and the radar. The further away the object is, the smaller is the SNR and the more difficult it is for the radar to detect it. We can also see that the signal returning from a target varies as the 4th power of the range. For example, if a target moves from 1km to 2km from the radar, the strength of the target’s signal return is 16 times less.
  4. 4. Highly reflective targets will return more energy back to the radar and will be more easily detected. The reflectivity or scattering from an object depends on factors such as the ma- terial the object is made of, its dimension, its shape and the angle at which the signal hits the object. This is defined as the Radar Cross Section or RCS. The RCS is a measure of the target’s ability to reflect radar signals back in the direction of the radar receiver. In a sur- veillance context, a potential target or intruder could be a pedestrian. A pedestrian is often characterized in literature as having a RCS of 1.0 m2. This does not mean that the pedestrian is physically 1 m2 in cross section, it means that its apparent size as observed by the radar is equivalent to 1 m2. The table below provides examples of RCS for various targets that may be encountered in a security or surveillance environment. TARGET RCS (m2) Crawler 0.03 ~ 0.1 Pedestrian 0.3 ~ 1.5 Light ground vehicle 5 ~ 50 Kayak 1~5 Small boat 5 ~ 100 Heavy ground vehicle 20 ~ 1000 Ship or vessel 50 ~ 10000 Small airplane 5 ~ 20 Helicopter 10 ~ 500 In attempting to detect targets, radars are faced with reflections coming from undesired objects, landscape and precipitation from the environment. Those unwanted reflections are called clutter and compete with the targets of interest. A target can only be detected pro- vided it rises above the system noise and clutter. 2100 Crystal Drive Suite 650 Arlington, VA 22202 T+ 1.866.458.ICXT
  5. 5. RADAR TERMINOLOGY TECHNICAL FOR THE LAYMAN NOTES The radar concept is a very simple one: throw packets of energy (traveling at the speed of light) in a known direction and time how long it takes for echoes of those packets to get back. Where no echo comes back there is nothing; where echoes come back there was something. That something can be a target of interest or an echo from the ground, a building, aircraft, person or other objects of no interest. Radar returns can be plotted (on a screen) or directly 2 interpreted by a computer to yield target data. Alan Browne RADAR is an acronym for “Radio Detection and Ranging” but has become a word in its own right used and misused for various things such as “They flew under the radar and won the contract.” Here’s some terminology decoded for non-rocket (or radar) scientists. APERTURE The size of the antenna. In general, for a given frequency, larger antennas are more effi- cient, sensitive and form tighter, more accurate beams. Bearing: The angle from north (usually true north) to the target from the radar. CLUTTER Radar returns not important to a radar’s function. If a radar is looking for walking people, competing returns from trees, fences, buildings, rainfall, etc. are called clutter. FALSE ALARM Also known as a “False Positive”, occurs when an alarm is generated but there is no target present. Most likely causes of false alarms are receiver noise and antenna sidelobes. FREQUENCY As radar is a radio device, each radar type has a frequency (cycles per second) that it trans- mits at. This is like tuning the radio in your car, but usually at much higher frequency. GHz Giga (billions) Hz (cycle per second). Your car radio FM tuner has numbers like 97.7. Those are in Mega (millions) of Hz (MHz). Radar operates at higher frequencies, typically between 1 GHz and up to 100 GHz in narrow to wide channels of about 100 to 2000 MHz. The higher the frequency the more resolution, however, the maximum detection range is typically reduced by rainfall and, in extreme cases, humidity. JAMMING Radars can be jammed (interfered with) by strong sources of radio energy. Some radars are designed to also act as jammers against other radars or radio receivers. MICROWAVE A “catchall” word to describe radio frequencies in the range of 1,000 MHz to 30,000 MHz. (1 GHz to 30 GHz). Most radars operate in some part of the microwave spectrum as do microwave ovens. MILLIMETER-WAVE Radio frequencies in the range of 30 GHz to 300 GHz. Frequencies in this range have wave- lengths ranging from 1 to 10 millimeters, hence ‘millimeter-wave’. PULSE REPETITION FREQUENCY How often (per second) that pulses are “fired” at the target space.
  6. 6. PULSE WIDTH In radars that are pulsed, the pulse width is the duration of the “on” time of the pulse (in micro or nano seconds). RADAR CROSS SECTION (RCS) The apparent (to the radar) size of a target. Typically in square meters. Is not directly re- lated to actual size. A hot air balloon has a low RCS. A small airplane can have a very large RCS. It depends on how much of the radar signal the target can reflect. RADAR POSITION Where the radar is located. Usually expressed in Latitude and Longitude. Some radar is mobile such as on ships, aircraft and vehicles. RADAR SCAN Since radars typically have a narrow beam, they have to scan the area of interest. This is done by turning the radar, antenna, or scanning back and forth or in other patterns. RADAR TRACKER The part of the radar (electronics, software or usually both) that sifts through the radar sig- nals to identify potential targets and to keep track of them. RANGE Distance, for example: how far a target is, how far the radar can detect a type of object, etc. SIDELOBES The antennas that form the beam cannot be made economically or technically large enough to make a perfect beam so smaller “extra” beams may be formed beside, above and below the main beam. Targets that get detected in the sidelobes may be rejected in special soft- ware algorithms. SNR (Signal to noiSe Ratio) A measure of how clear a signal is received. The higher the SNR the better. TRACKING THRESHOLD In the presence of clutter and noise a criteria is needed to separate targets from the environ- ment around it. The tracking threshold (usually in “dB” or decibels) is the level required for a signal to be “above the noise” to be a candidate to be a target. TARGET POSITION Where a target is located. This can be expressed as a Range+Bearing from the radar or as a latitude/longitude (or other position reference). Knowing where the radar is located one (or usually a computer) can compute the latitude and longitude of a target. 2100 Crystal Drive TRANSCEIVER Transmitter + Receiver packaged as one unit and typically exchanging signals for timing Suite 650 and comparison. Arlington, VA 22202 WAVEGUIDE T+ 1.866.458.ICXT A component of some radars that typically connects the antenna(s) to the transmitter and receiver. WAVELENGTH The length of one cycle of the signal (usually in cm or meters). It is the speed of light divided by the radar frequency. As the frequency goes up, the wavelength gets shorter.
  7. 7. GROUND SURVEILLANCE RADAR TECHNICAL IN SECURITY APPLICATIONS NOTES THE PROBLEM Traditionally perimeter security is provided by a physical barrier that may or may not be sensored and monitored. If configured as a sensor platform and monitored for intrusions a traditional security fence can be prohibitively expensive to install and maintain as well as prone to false alarms. If properly maintained a security fence can only provide information 3 Howard about where an intrusion “was”. Once the perimeter is breached, security personnel are forced to rely on other alarms and sensors to track the progress of intruders. Unfortuneatly, Borst at this point an intruder may have already reached their objective. Physical fences serve well as boundary markers and barriers; they do not make good sensor platforms. The solution to this dilemma is to use sensor systems which give security personnel knowledge of the entire security space they are required to protect. The only sensor system that can reliably provide this information is ground surveillance radar. Ground surveillance radars give operators insight into not only their perimeter but the area both within the perimeter and beyond. Using ground surveillance radar, paired with integrated surveillance cameras and real time command and control will give operators superior ability to “See first, Understand first and Act first.” This capability is not reliably available with any other security system. SEE FIRST The ICx ground surveillance radars are used to detect ground targets at up to a 12,000 me- ter radius surrounding the radar. The radar uses its beam to detect changes in the distance to everything in the radar’s surroundings. Thus, the radar detects moving as well as newly introduced stationary objects. As the radar sweeps, it detects range and bearing to targets and displays that information as needed. The radar combats false alarms with erroneous target detection during events such as heavy rain by having the ability to filter out environ- mental noise. Target range and bearing information is sent to the command and control system. The C2 system then slews various imaging systems to the target, and displays the target image, position, and range and azimuth information on a central display for security forces. UNDERSTAND FIRST Understanding the nature of a threat is critical to intercepting and defeating the threat. Knowing where the threat is coming from, where it is now, and where it is going is a key objective of an effective security system. Traditional perimeter sensors give very little real time information to responders. For example, a sensor fence surrounding a critical facility with an 8,000 meter perimeter notionally has knowledge of 8,000 square meters of ground space. A single radar “watching” the same perimeter has knowledge of over 6 million square meters of ground space. Now you know where the intruders are anywhere inside the “view” of the radar. ACT FIRST Ground surveillance radar, paired with slew to queue infrared video cameras, and the latest 2100 Crystal Drive command and control technology allows security personnel to operate inside the adversary Suite 650 decision loop. Ground surveillance radars allow you to control your security zone by inter- Arlington, VA 22202 cepting threats before they can complete their mission. Your ability to control your security zone is enhanced dramatically by using properly integrated security technologies such as ground surveillance radar. This technology is available now to solve the toughest security T+ 866.458.ICXT challenges.
  8. 8. RADAR SPECTRAL BANDS TECHNICAL NOTES 4 The selection of a radar’s operating frequency is generally the result of a trade-off analysis that considers desired detection range, weather and clutter environment, available aperture size, properties of targets of interest, and cost of RF components. Radar performance must first be quantified on the basis of propagation efficiency in various media (air, foliage and ground) with respect to the required detection range. The general prin- Pierre ciple is that the lower the frequency, the more efficient the propagation of radio waves through Poitevin the medium. In the presence of obscurants, propagation is most favorable when the RF wave- length is much larger than the particle size composing the propagation medium. This is why radars tend to have much better performance than optical systems through smoke, dust, fog and rain. However, using a lower frequency dictates a larger antenna for a given angular resolution. As a rule of thumb, the physical dimensions of an antenna are related to the required resolution by the following equation. θ ≈ 70° λ / w where θ is the half-power antenna beam width (resolution) in degrees, λ the wavelength and w the antenna longitudinal dimension. Using the above equation, an imaging radar at Ka-band (35 GHz) with a two-degree angular resolution would require a 30 cm antenna while a foliage penetrating radar working at UHF (900 MHz) would have a 12-meter antenna to achieve comparable resolution. Applications dictate the radar detection range. A radar intended for airborne surveillance may require a detection range in excess of 1000 km to provide an adequate response time to coun- ter an incoming threat. To assure range performance out to 1000 km under rain conditions, propagation properties would require operating around S-band (2~3 GHz) or even L-band (1~2 GHz). In another application, a police radar used for measuring speed of vehicles would only require a maximum range of 500 m. Given the shorter range required, practically any fre- quency up to W-band (110 GHz) could be used. In this case, antenna size and component costs are most likely to influence the choice of operating frequency. Properties of targets must also be considered in selecting the operating frequency as optimal de- tection is achieved when radar resolution matches the target size. As radar resolution depends on antenna beam widths (azimuth and elevation) and range resolution, the higher frequencies are best suited because they yield small antennas and little fractional bandwidth to achieve range resolution. The fractional bandwidth may be expressed as the percentage of radar signal bandwidth with respect to transmit frequency. In an application to protect flight lines from in- trusions by pedestrians and crawling persons, a Ka-band radar is ideal because it allows small size for easy deployment and high range resolution with little fractional bandwidth. Cost of RF components is another factor to consider in selecting the operating frequency. The higher the frequency, the more expensive are the components. With the recent developments in the telecommunications industry, very affordable components are now found up to 5.8 GHz (C-band). Components at X-band (~ 9 GHz) are now manufactured with high yields and prices have been declining steadily over the last few decades. Components at V and W bands are still expensive because of the manufacturing tolerances required for the short wavelengths (mil- limeter wave) and of limited demand.
  9. 9. The table below presents the various frequency bands used for radar with their most common applications. Frequency Band Nominal Range Specific Radar Typical application Designation Frequency Assign- ment (1) HF 3 ~ 30 MHz no specific bands Over the horizon radar VHF 30 ~ 300 MHz 216 ~225 MHz Very long range Ground penetration radar UHF 300 ~ 1000 MHz 420 ~ 450 MHz Very long range 902 ~ 928 MHz Foliage penetration L 1.0 ~ 2.0 GHz 1.215 ~ 1.390 GHz Long range surveillance S 2.0 ~ 4.0 GHz 2.305 ~ 2.385 GHz Long range surveillance 2.417 ~ 2.483 GHz Air traffic control 2.700 ~ 3.650 GHz C 4.0 ~ 8.0 GHz 5.250 ~ 5.85 GHz Air surveillance Air traffic control Airborne altimeter X 8.0 ~ 12.4 GHz 8.500 ~ 10.55 GHz Long range ground surveil- lance Airborne weather radar Weather observation Marine radar Police radar Ku 12.4 ~ 18 GHz 13.4 ~ 14.0 GHz Guidance 15.7 ~ 17.7 GHz Medium range ground surveillance K 18 ~ 27 GHz 24.05 ~ 24.25 GHz Police radar Ka 27 ~ 40 GHz 33.4 ~ 36.0 GHz Short range ground surveil- lance Targeting Imaging V 40 ~ 75 GHz 59 ~ 64 GHz Automotive anti-collision W 75 ~ 110 GHz 76 ~ 81 GHz Imaging 92 ~ 100 GHz Automotive anti-collision Airborne wire detection Fig 1: Radar frequency bands and applications (1) NTIA, Office of Spectrum Management, October 2003 2100 Crystal Drive Suite 650 Arlington, VA 22202 T+ 1.866.458.ICXT
  10. 10. FREQUENCY MODULATED CONTINUOUS TECHNICAL WAVE (FMCW) RADARS HAVE ADVANTAGES NOTES 5 OVER PULSE DOPPLER RADARS FOR GROUND SURVEILLANCE Ground Surveillance Radar can build a virtual wall around facilities or on a border. It pro- Walker vides operators and agents more response time to access, prioritize and apprehend intrud- ers. It provides wide area surveillance and tracking over a large, 360 degree area, direct- Butler ing responders even after an intrusion has occurred. But, all GSR technologies are not the same. There are two primary GSR technologies - Pulsed Doppler radar technology and Frequency Modulated Continuous Wave (FMCW) radar technology. Most Pulsed Doppler radars are derivatives of legacy military battlefield radar being applied for wide area surveillance, while a new generation of FMCW radar technology was developed for wide area surveillance, site security and force protection. It was specifically developed to detect and track walking per- sonnel. ICx Radars use FMCW radar technology. FREQUENCY MODULATED CONTINUOUS WAVE (FMCW) RADARS FMCW radars operate on the imaging principle; that is, they break up the background into small segments, or resolution cells, and then measure changes in the signal return from each cell to detect small targets, such as walking people. Typical resolutions for long range FMCW radars are less than 1 meter in range and less then 1 degree in azimuth. The smaller the cell the easier it is to detect and track a target. FMCW operation is independent of the speed or direction of travel of the target, only its size with respect to the resolution cell in which it is located. Modern FMCW radars can detect people moving at near zero speed and walking in any direction with respect to the radar. PULSE DOPPLER (PD) RADARS Pulse Doppler Radars operate on the Doppler principle, which states that all moving objects will exhibit a frequency shift from the transmitted signal to the received signal, which is proportional to the speed of the target in the direction of the radar. If a target is walking directly toward the radar at 3MPH, the radar will detect a frequency difference in the re- ceived signal and declare that a 3 MPH target has been detected. If the target is walking at 45 degree angle to the radar, the Doppler signal will be 3 MPH times the cosine of the angle, or about 2.1 MPH. However, background clutter like trees and bushes also have some apparent speed when the wind blows. In order not to have a large number of false alarms, that low speed signal return from the clutter must be filtered out. A virtual velocity threshold (blind speed) is created be- low which targets will not be reliably detected. That means that some slowly moving targets could be filtered out along with the clutter. It also means that higher speed targets moving “across” the radar beam may be filtered out because speed only generates a Doppler signal proportional to the incoming or outgoing speed, which is called radial speed (approaching or receding in the beam). IMPLICATIONS OF USING DOPPLER AS THE DETECTION TECHNIQUE A fundamental deficiency exists such that wide area surveillance systems using Pulse Dop- pler radars have large areas where “slow” targets will not be detected. In fact, if an intruder walks at a speed somewhat below the velocity threshold (defined as the “blind speed”) of the radar, it doesn’t matter in what direction the intrusion takes place, the intruder will likely not be detected at all – the intruder can simply walk through the perimeter or across the border and the radar will not detect the target. Alternatively, an intruder can walk between two radars spaced along a border and will be moving across the beams, or tangentially to
  11. 11. each radar, and therefore, can walk at a higher speed than the velocity threshold, and still not be detected. This deficiency gives the intruders a major advantage. Those fa- miliar with border operations know that intruders learn to avoid areas where they are apprehended regularly. Thus, holes in coverage inherent to Pulsed Doppler radars will be found and exploited, nullifying the very purpose of the radars. Changing the spacing or offsetting radars in latitude will somewhat change the shape of the non-detect zones, but will not eliminate the deficiency. In summary, PD radars have an inherent flaw when used in ground surveillance ap- plications. There is a conflicting trade off between minimizing clutter returns and the minimum detection speed of the target. Most PD radars will never detect at speeds less than 1.5 miles per hour (a distinct probability with walkers carrying 50 pounds or more of contraband). THE FMCW ADVANTAGE - SUMMARY The STS-12000 radar has the advantage of being designed specifically for perimeter and border surveillance using the most optimum technology for this mission: frequency modulated continuous wave (FMCW). The benefits of FMCW over other technologies such as pulse Doppler (PD) are numerous: • FMCW is less complex, safer and lower cost than PD • FMCW gives low false alarm rates Proven in Government testing - The only radars to pass stringent U.S. Air Force false alarm test Less likely to alarm with wind blown objects --- grass and leaves, rain One FMCW installation has 31 radars netted together using only one operator • FMCW sees a higher percentage of valid targets Won’t miss slower targets or tangential ones – no holes in coverage – no one penetrates • Smaller beamwidth for better pointing of cameras 2100 Crystal Drive Suite 650 Arlington, VA 22202 T+ 1.866.458.ICXT
  12. 12. RADAR PERFORMANCE IN RAIN TECHNICAL NOTES Radar (RAdio Detection And Ranging) devices operate in the Radio Frequency (RF) band typically in UHF (300 MHz) through W-band (110 GHz) or higher. Radar detection devices are affected by water in the atmosphere although radar to a much lesser extent than other detection technologies such as Laser, InfraRed and Video. The biggest detection loss is due to rain in the atmosphere and is based on the droplet size and the radar transmission wave- 6 length as shown in Figure 1. Terry Wilson Radar Attenuation & Backscatter due to Rain 10 1.E+02 1.E+01 backscatter cross section per unit rainfall attenuation (one-way) dB/km 1 1.E+00 1.E-01 volume, cm²/m3 0.1 1.E-02 1.E-03 0.01 1.E-04 0.001 1.E-05 1.E-06 0.0001 1.E-07 0 1 2 3 4 5 6 7 8 9 10 Wavelength, cm drizzle 0.25mm/hr light rain 1mm/hr moderate rain 4mm/hr heavy rain 16 mm/hr excessive rain 40mm/hr drizzle 0.25mm/hr light rain 1mm/hr moderate rain 4mm/hr heavy rain 16 mm/hr excessive rain 40mm/hr Figure 1: Attenuation and backscatter versus frequency at various rain rates1 For ground based radar systems detecting intruders, personnel and vehicles, the radars typ- ically operate at the X-band (8 to 12.5 GHz), Ku-band (12.5 to 18 GHz) and Ka-bands (26.5 to 40 GHz). Attenuation of the RF energy transmitted from the radar, reflected from the target and re- ceived by the radar affects the detection performance. This is considered the two-way ab- sorption loss. For the X and Ku bands there is little difference in absorption performance. For the Ka band there is much more significant absorption loss and this is why these ground based radars are typically used for much shorter range applications. For longer range (>3 kilometers) applications, using radars in the X and Ku bands is preferred. Rain backscatter is transmitted RF energy from the radar that is reflected off the rain drop- lets back to the radar that competes with the signal-to-noise (SNR) ratio of the intruder. If the rain backscatter gets large enough then it can dominate the SNR of the intruder making the intruder undetectable. This rain backscatter is a function of the volume over which the intruder is being detected and is defined by the radar antenna, the range cell and the trans- mit frequency. The antenna defines the detection footprint of the radar and the range cell is the typical range bin or resolution cell of the radar. Radar antennas are inversely propor- tional to the beam width so the larger the antennas the smaller the beam width or detection foot print of the radar and the better the detection performance in rain. The range cell is defined by the pulse width for pulse Doppler type radar systems or the frequency sweep for FMCW radar systems and in general, the smaller the pulse width or larger the frequency sweep the better detection performance the radar will have in rain. M. Skolnik, (1980) Introduction to Radar Systems, pg 501-503. 1
  13. 13. Figure 2 and Figure 3 show a comparison of the backscatter for two representative ground based radar systems. Figure 2 for an FMCW radar and Figure 3 for a Pulsed radar. Note the large differences in range cell size cause the difference in back scatter performance. Typical values of effective Radar Cross Section (RCS) for a walking person are on the order of 0.5 to 1.0 m² and for a vehicle on the order of 10 m². If the effective RCS is equal to the intruder size then the SNR is zero and the intruder is undetectable. BACKSCATTER DUE TO RAINFALL 10.00 FREQUENCY: 16 GHZ AZIMUTH BEAMWIDTH: 1 DEGREES 8.00 EFFECTIVE RCS (SQ-M) ELEVATION BEAMWIDTH: 4 DEGREES 6.00 RANGE CELL SIZE: 1 METERS 4.00 2.00 0.00 0 0.6 1.2 1.8 2.4 3 3.6 4.2 4.8 5.4 6 6.6 7.2 7.8 8.4 9 9.6 10.2 10.8 11.4 12 RANGE (KM) 0.25 mm/hr 1 mm/hr 4 mm/hr 10 mm/hr 40 mm/hr Figure 2: Rain Backscatter versus Range for a representative FMCW Radar BACKSCATTER DUE TO RAINFALL 10.00 FREQUENCY: 9 GHZ AZIMUTH BEAMWIDTH: 3 DEGREES 8.00 EFFECTIVE RCS (SQ-M) ELEVATION BEAMWIDTH: 5 DEGREES 6.00 RANGE CELL SIZE: 15 METERS 4.00 2100 Crystal Drive 2.00 Suite 650 Arlington, VA 22202 0.00 0 0.6 1.2 1.8 2.4 3 3.6 4.2 4.8 5.4 6 6.6 7.2 7.8 8.4 9 9.6 10.2 10.8 11.4 12 T+ 1.866.458.ICXT RANGE (KM) 0.25 mm/hr 1 mm/hr 4 mm/hr 10 mm/hr 40 mm/hr Figure 3: Rain Backscatter versus Range for a representative Pulsed Radar Other atmospheric obscurants such as sand, smoke, dust, snow, hail, etc. are either very small in relation to the wavelength or have less water content than rain and their effects over the detection ranges of the ground based radars is negligible.
  14. 14. RADAR PERFORMANCE PARAMETERS TECHNICAL NOTES 7 Radar Performance parameters define what a radar does. Characteristics describe how a radar achieves its performance. Features are desirable characteristics. This tech note discusses the nature of the parameters that define radar performance. MAXIMUM DETECTION RANGE False alarm detection plays a significant role in the effectiveness of a radar and its maximum detection Walker Butler range. A false alarm is an event that erroneously signals the presence of a radar target when there is no legitimate target. The maximum detection range of a radar is the longest distance from the radar to a target at which the radar can reliably declare that the return signal from the target has exceeded a set threshold. The return signal is usually very low in amplitude and must be detected above the thermal noise level in the radar electronics. In well designed radars, the thermal noise is quite low and small signal returns can be detected reliably. This threshold is set based primarily on reducing false positives (false alarms) that would occur on noise energy generated internally to the radar. This noise is present in all radars and is called thermal noise because its amplitude is a function of the radar temperature – higher temperature causes more noise. Many factors influence the ability of a radar to detect a target. The radar design itself, the type of target (person, vehicle), what characteristics of the target determine its measurability (size, speed), the distance from the radar to the target, the size of the target (radar cross section), the environment between the radar and the target (rain, fog) and the environment in the immediate vicinity of the target called clutter, like trees, grass, building – they all compete with the target. Since the target must be detected in a background of noise and/or clutter, the likelihood of detecting a target on any given opportunity (look) is statistical, that is, it varies from look to look. That is because the noise amplitude varies in a random manner from look to look, either adding to the target signal or subtracting from it. The statistical nature of the target creates even more uncertainty. The Probability of Detection on any given look is a measure of the likelihood of detecting the target, or of having the signal from the target cross the previously mentioned detection threshold. The nature of the statistics of the noise, the target and the clutter are all different, but can be described mathematically by complex equations which predict detection range for various false alarm conditions, which are selected by the radar designer. All of the above factors affect the maximum detection range of a radar for a particular target, and any testing for determining detection range, must account for the statistical nature of the process by doing many trials to establish this parameter. FALSE ALARM RATE From the foregoing description, it seems that the detection range could be increased by lowering the detection threshold to “see” lower signal levels. While this is true, the noise environment described means that, on any given look, there is a finite probability that a noise spike could cross the threshold, causing a false alarm. Thus exists the classical battle between sensitivity and false alarms; that is, the desire to increase the Probability of Detection is offset by the resulting increase in the Probability of False Alarms. The latter is generally quantified in time by using the parameter False Alarm Rate (FAR), which expresses the false alarm probability as a function of time. Thus, all comprehensive radar specifications contain a FAR requirement, say 2 or 3 per day, so that a radar operator is not unduly distracted attending to an alarm that doesn’t really exist. This becomes very important in security systems which combine many radars for perimeter or border protection over long distances, because higher false alarm rates require more responders to chase down the cause of an alarm.
  15. 15. REVISIT TIME Radar revisit time is the time it takes for the radar to complete its search for targets and return to begin another search interval. For example, in a radar that goes around 360 degrees, it is the time for one revolution to be completed. Since the target detection process is statistical, it follows that the more time the radar “looks” at an area where there is a target, the sooner that target will be detected. Of course, the longer the detection range, the longer period of time can be allocated to the detection process. Also, the slower target speeds can be allocated more detection time. For example, a crawling person may need many looks to establish detection, due to the very small target size – but the crawler doesn’t travel very far during the process, so a relatively long detection time is acceptable. A fast moving, vehicle should be detected quickly or it will travel a long distance prior to detection. A short revisit time improves the detection process and also improves target tracking after detection. Since target speed and direction are not controllable, more looks in a given time will result in more accurate tracking of the target. RADAR RESOLUTION The ability of a radar to detect and track a target is affected by the radar’s resolution, that is, how small of a “space” does the radar look at. Radar space is defined in four dimensions – range, azimuth angle, elevation angle and speed. Not all radars measure in all these dimensions. For example, a police radar uses the angle and speed dimensions, and doesn’t measure range to the vehicle. Generally, a maritime navigation radar uses the angle and range dimensions, and doesn’t measure speed because most returns from this radar are not moving. Almost all radars limit the area they look at in angle, because they use an antenna to focus the energy on a suspected targeted area. This angle is two dimensional, horizontal (azimuth) and vertical (elevation), generally expressed in degrees. A “pencil” beam is symmetrical in both planes, but many radars will have a very narrow azimuth beam width, depending on the radar’s function. For example, a maritime navigation radar will have a very narrow azimuth beam width to very accurately trace a shoreline, but will have a very wide elevation beam so the boat can pitch and roll in the waves but the land will still be within the beam. Resolution in range is important to accurately determine the range to a target and to eliminate clutter behind and in front of the target. Some radars measure the radial speed of the target directly using the Doppler principle, which states that the speed of a moving object will affect the frequency of the return signed. This “speed resolution” feature is useful to discriminate moving vs. stationary target. However, because wind-blown vegetation and rain can appear to be moving, much care must be taken not to generate false alarms for ground radars susceptible to windblown clutter motion. Radars which have small resolution cells are called high resolution. High resolution radars provide more clutter background rejection, which radars discriminate better against competing returns from the ground, grass, trees, and rainfall. The drawback of high resolution is that it takes longer to search for a target because more resolution cells must be looked at to find the target. SEARCH VOLUME 2100 Crystal Drive Typically the more “space” a radar can search for targets, the more utility the radar provides to a surveillance system. A 360 degree radar may be more useful than one that merely scans a sector, Suite 650 unless it is known with certainty where potential targets exist. However, there is a trade off between Arlington, VA 22202 available search volume and resolution. It takes longer to search a given volume with a high resolution radar than a radar with less resolution. Therefore, there is a constant battle between revisit time, T+ 1.866.458.ICXT resolution and search volume. SUMMARY This technical note has presented the major performance parameters which describe how well a radar performs. The design of a radar is primarily one of performance tradeoffs, involving range resolution, search volume, clutter rejection, false alarm rate and revisit time. These issues must be balanced against target types, frequency allocation regulations, size, weight, cost, power and environmental considerations, such as rain, snow, operating temperatures, vibration and shock. A well designed radar represents a delicate balance of many seemingly incompatible factors.
  16. 16. EARLY HISTORY OF RADAR TECHNICAL NOTES It wasn’t until 1942 that the US Navy coined the official acronym for modern day RADAR, which stands for radio detecting and ranging. However, the scientific concept of radar developed millions of years ago in nature as the ultrasonic sensor of a bat. Bats release a short “cry” through their nasal passages and receive its echo through antennae-like ears. Interestingly, the oldest radar used as a warning 8 Rick Mannello system also evolved in nature as defense against bats. Tiger moths, a staple diet for bats, have evolved with ears equipped to detect and scramble a bat’s radar system as a method of evading attacks. FIRST RADARS Human involvement with radars first began around the beginning of the 20th century. In 1904, German inventor, Christian Hulsmeyer, acquired a patent for the Telemobiloskop or, the Remote Object Viewing Device. This device was developed as a tool to detect and avoid ship collisions. Unfortunately, Hulsmeyer’s creation never really caught on and the Telemobiloskop quickly fell into marketing oblivion. Despite Hulsmeyer’s failure, the theory behind primitive radar technology continued to advance and change form and function. Radio transmissions, remote radio measuring techniques and object detection devices throughout the 1920s and ‘30s all utilized the concept of radar. Experiments in the ‘20s and ‘30s at the US Naval Research Labs by Dr. Alfred Hoyt Taylor and Leo C. Young demonstrated rudimentary CW detection of ship and aircraft targets. Limitations of technology held back development until Robert Page’s breakthrough demonstration of pulsed radar in 1936. A year earlier, in Great Britain, another breakthrough, the so-called Daventry Experiments of Sir Robert Watson-Watt, demonstrated detection of aircraft at over eight miles. This led to the immediate development of the first operational radar system know as Chain Home, a series of radar towers along the south coast of England which were deployed in 1937. RADAR USE By 1939 Britain, France, Germany, Russia, Switzerland and the US were all counted among countries with developed and functional radars. World War II marked the first organized use of radars during battle. Throughout the war, several countries relied upon their radars to prevent attack as well as to search and destroy targets. Significant efforts continued to enhance the rudimentary science. The refinement of the magnetron enabled higher frequency (3GHz) performance leading to higher resolution and more compact radar systems; thus opening a range of applications for ground, land and sea-based systems. In spite of the rapid development of radars, they were still not fully understood, nor did their use have well developed concepts of operation. A significant missed opportunity in history for radars occurred on December 7, 1941 or Pearl Harbor Day. On that day, radar operators actually detected the presence of invading squadrons approaching the Hawaiian Islands. However, when they reported their findings their superiors quickly dismissed the observations. Had senior military staff trusted in less tenured 2100 Crystal Drive personnel and their radar technology, that historical day may have had an entirely different ending. Suite 650 Arlington, VA 22202 SUMMARY By the end of World War II most of the radar technologies used today were in place, though they were T+ 1.866.458.ICXT dependent upon the limited technical means of the time. Chirp radar, synthetic aperture radar and monopulse techniques were all in use. Future technical developments in magnetrons, electronics and signal process would bring radar forward into the second half of the century.
  17. 17. WHAT TO CONSIDER WHEN DESIGNING TECHNICAL NOTES A SECURITY SYSTEM WITH RADARS “Detection, delay and response are all required functions of an effective physical security system. These functions must be preformed in order and within a length of time that is less than the time required for the adversary to complete his task. A well designed system provides protection in depth minimizes the consequence of component failure and exhibits balanced protection.”(Sandia National Labs- Design 9 Howard Borst and Evaluation of Physical protection systems) PHYSICAL SECURITY SENSORS Time. Time is the enemy of any physical protection system. Most physical security sensors provide information about a point in space at a specific time. More specifically, most systems tell you where an adversary was, giving you the challenge of finding an adversary’s current location before they are able to carry out their objective. Having enough time to complete yourtask before the intruder is able to complete their’s is the challenge. If properly considered, positioned and integrated, ground surveillance radar can provide the responder critical time to interdict and respond to a threat. The proper implementation of any security system, including radar, requires understanding of the threat, possible avenues of approach, sensor system performance and environmental conditions. Understanding the threat(s) to your facility can have a large impact on the selection, placement and operation of your system. For example if you are protecting a perimeter are the threats vehicle-borne, foot-borne or both? In the case of radar, a vehicle-borne threat is easier to detect than a foot-borne infiltration. To detect vehicles radar must be carefully placed with direct line of sight to any vehicle capable avenues of approach. If the threat is foot-borne, then the line of sight issues become more complex. Blind spots must be filled and the probability of detection must be well understood. Response forces must also have a good working knowledge of the specific battle space the system is defending. SENSOR PERFORMANCE In describing sensor performance we use three terms: Probability of Detection (Pd), Nuisance alarm rate and vulnerability to defeat. Perfect probability of detection is measured as 1.0 for the ideal sensor. This means that a sensor will detect an intruder 100% of the time in all conditions. While no sensor is perfect, the closer the sensor performs to ideal the better overall system performance. Depending on the criticality of the asset being protected multiple sensors can be layered to provide the level of detection desired. Nuisance alarms are alarms that are not caused by an intrusion. The ideal nuisance alarm rate would be zero. The most likely cause of nuisance alarms are environmental factors such as wildlife, vegetation and weather conditions, Selection of sensor systems and radars should take these factors into account. For example, some radar systems are easily fooled by moving vegetation while others are not. A cluttered background of shrubs and trees can cause challenges for any radar being able to “see” intruders. Careful consideration of the space the radar will survey will improve the overall security system performance. VULNERABILITY OF DEFEAT The next factor for consideration is an understanding of the sensor system’s vulnerability of defeat. Most physical security sensors can be defeated depending on the skill of and time available to the adversary. Radar systems are particularly difficult to defeat given the proximity of the sensor to the edge of the detection zone. For example a single fence sensor can be defeated any number of ways most of which involve fairly simple techniques. A radar system may be defeated but only through complex technical means unavailable to most adversaries SUMMARY Radars operate at their best with long line of sight and clear environmental conditions. Superior security radars excel when the environmental conditions are less than perfect and the terrain is challenging. If careful consideration is given to the design of a physical security system; taking into account the threat, environment, and overall system performance requirement, it is evident that ground surveillance radars provide superior performance and value over other perimeter sensors.
  18. 18. The table below presents the various frequency bands used for radar with their most common applications. Frequency Band Nominal Range Specific Radar Typical application Designation Frequency Assign- ment (1) HF 3 ~ 30 MHz no specific bands Over the horizon radar VHF 30 ~ 300 MHz 216 ~225 MHz Very long range Ground penetration radar UHF 300 ~ 1000 MHz 420 ~ 450 MHz Very long range 902 ~ 928 MHz Foliage penetration L 1.0 ~ 2.0 GHz 1.215 ~ 1.390 GHz Long range surveillance S 2.0 ~ 4.0 GHz 2.305 ~ 2.385 GHz Long range surveillance 2.417 ~ 2.483 GHz Air traffic control 2.700 ~ 3.650 GHz C 4.0 ~ 8.0 GHz 5.250 ~ 5.85 GHz Air surveillance Air traffic control Airborne altimeter X 8.0 ~ 12.4 GHz 8.500 ~ 10.55 GHz Long range ground surveil- lance Airborne weather radar Weather observation Marine radar Police radar Ku 12.4 ~ 18 GHz 13.4 ~ 14.0 GHz Guidance 15.7 ~ 17.7 GHz Medium range ground surveillance K 18 ~ 27 GHz 24.05 ~ 24.25 GHz Police radar Ka 27 ~ 40 GHz 33.4 ~ 36.0 GHz Short range ground surveil- lance Targeting Imaging V 40 ~ 75 GHz 59 ~ 64 GHz Automotive anti-collision W 75 ~ 110 GHz 76 ~ 81 GHz Imaging 92 ~ 100 GHz Automotive anti-collision Airborne wire detection Fig 1: Radar frequency bands and applications (1) NTIA, Office of Spectrum Management, October 2003 2100 Crystal Drive Suite 650 Arlington, VA 22202 T+ 1.866.458.ICXT
  19. 19. THE POWER OF SLEW-TO-CUE TECHNICAL SURVEILLANCE CAPABILITY NOTES Traditional video surveillance consists of a large number of cameras feeding back to a simple “command center” comprised of a security guard viewing a large bank of monitors. This presents workload issues for the guard. Studies show that an individual can effectively view the bank of monitors for an average of 20 minutes. If a typical shift for a security guard is eight hours, how effective is the security for the remaining seven hours and forty minutes? 10 Nirav Pandya CAMERA SURVEILLANCE It is left to the security guard to find a potential threat on one of the many monitors and than start a series of operations to determine if the threat warrants action or not. In most cases, the first action to occur is the security guard must orient themselves to determine where the potential threat is. This is often very difficult when dealing with a bank of monitors displaying imagery from different cameras at different locations and a myriad of viewing angles. After the location is resolved, action can be taken to address the potential threat. The guard will most likely call a mobile team which can reach the potential threat and intervene. Simultaneously, the guard in the command center must continuously relay the location of the threat to the mobile team. As the intruder continues to move, the entire problem is made more difficult as the command center operator will have to try to control the cameras via joystick and relay coordinates at the same time. It becomes increasingly difficult as the intruder transfers from one camera’s field of view to another. Additional and valuable time will be lost in reacquiring the target on a different camera, each with typically a limited field of view. SLEW TO CUE SURVEILLANCE A more effective and strategic solution combines ground-based radars, pan and tilt imagers (visible CCD and thermal) and software technology to automate many of the critical challenges that the security guard faces. As a result, the guard will be able to give the proper attention to relaying the pertinent information to all necessary individuals. This solution will detect moving objects and automatically slew the camera to the place the object in its field of view. In addition, the camera will track the object as it moves across the area. If the tracked object moves out of the field of view of one camera, it will be picked up automatically by another camera and tracking will continue seamlessly. All of this information will be displayed in a command and control center that will consist of a map of the entire location. Once an object is detected by the radar, the tracks will appear on the geospatial map so the guard knows the exact location of the threat as well as its relative location to landmarks in his perimeter. As the threat moves around the entire location, the map is constantly updated to indicate real time location for the intruder. Meanwhile the cameras are automatically following the intruder, providing a real time video display in the command and control center which offers the operator a clear assessment of the target. SUMMARY Depending on the terrain and shape of the perimeter that you are interested in protecting, various radars can be networked together in an integrated solution. Use of ground based radars reduces the numbers of cameras that must be deployed. The 360 degree surveillance capability of ground radars allows multiple 2100 Crystal Drive targets to be tracked, and cameras to be used to selectively view areas of interest. This greatly simplifies a Suite 650 perimeter design, as well as the installation and maintenance of camera only systems. Arlington, VA 22202 Radars have the capability of detecting a moving person at distances up to 5,000 meters and when matched with thermal imagers, people can be recognized up to 3,000 meters. With specific integrated options you T+ 1.866.458.ICXT can customize your solution by choosing the radars, CCD and thermal imagers which best suit the area you wish to protect. Slew to cue can also be performed utilizing other sensors. For example, if an access control location is breached, the system can be designed to slew a camera immediately to view the door that was breached. With slew to cue functionality, you can optimize your security solution to ensure that you do not miss any potential threats. The command and control center is simplified providing automatic detection and assessment and allowing the operator to focus on making an appropriate response. These are effective tools in perimeter surveillance allowing for the most efficient and effective use of valuable security assets.
  20. 20. MULTILAYERED PERIMETER SURVEILLANCE TECHNICAL FOR AIRPORTS NOTES Every seaport, airport, critical infrastructure facility or base perimeter is different, with unique physical layouts and surrounding terrain. Land, water, hills, trees, buildings and roads all impact the way in which these perimeters are protected. In airports, for instance, runway layouts, staging areas and terminal 11 Dan Manitakos buildings differ in design and layout – all of which affect sensor sight lines. Add to these widely varying physical layouts the need for adequate warning and response times for an unwanted intrusion and it is not sufficient to simply know a breach of the perimeter has occurred. Security personnel must be able to assess the threat, track movement both before and after the intrusion and react quickly to deter the security violation. PERIMETER SECURITY Perimeter security is about deterrence, detection, assessment and action. Traditional methods focus primarily on fence systems and fence alarms. The flaw in these systems are three fold: first and foremost, when a fence alarm sounds the security breach is either in progress or has already occurred; second, current technology is prone to false alarms; and third, these systems provide no tracking, assessment or situational awareness capabilities. Security officers need more information in order to react appropriately. They need to know the type of threat with which they are dealing, from where the threat is coming, and where it is headed. While fences remain an important physical deterrent, they do not provide a complete solution. MULTILAYERED SURVEILLANCE A multilayered surveillance approach is critical to protecting high value assets and infrastructure. This approach includes a combination of sensors such as imaging cameras, thermal imagers and radars. These sensors can be networked through a common command and control software backbone, providing situational awareness and implementing video analytics to improve efficiency. A standard scenario might include medium to long range radars, which provide a command view of a large facility and/or perimeter, such as an airport. These can be augmented with shorter range radars, as needed, to eliminate blind spots. These radars would be interoperable with selected cameras equipped with “slew-to-cue” functionality. With this functionality, the radar can detect an intruder, track movement and pinpoint geo positioning, while the cameras provide visual identification and information concerning the impending threat to assist in measuring the response. Although cameras (either daylight or thermal) are a fundamental part of a security solution, camera-only surveillance can also provide flawed coverage and security. A solution of strictly cameras requires security operators to simultaneously monitor a large quantity of video. Further, cameras provide a relatively narrow field of view as compared to wide-area radars. This means that a relatively large number of cameras must be deployed, along with the associated infrastructure support, to provide acceptable 360° coverage. Successful solutions for wide area surveillance should provide 360° situational awareness using 2100 Crystal Drive an intelligent combination of radar and cameras. Suite 650 Arlington, VA 22202 SUMMARY T+ 1.866.458.ICXT The deployment of a multilayered, wide-area surveillance solution enables comprehensive perimeter coverage. Wide area radars combined with slew-to-cue cameras and thermal imagers, provide an important advantage in total system cost and responsiveness. The most cost effective solution uses a family of radars with various range capabilities which are integrated with a family of cameras with diverse range performance, all fully interoperable. The use of radar in the perimeter solution reduces the camera requirement while providing a longer response time which can enable a better deployment of resources. Both factors contribute to lower total system cost and overall maintenance. Thus, an integrated radar/ camera solution provides advantages in both up front acquisition and life cycle costs.
  21. 21. SOME COMMON IR APPLICATIONS TECHNICAL NOTES IR camera technology or thermal imaging has an abundance of applications in the modern world. Some of the most notable include building inspection, electrical/mechanical inspection, surveillance and searching for buried objects. 12 Dustin Radney When applied to a surveillance system, thermal technology allows the user to see in the absence of light. In an application where there is a fixed perimeter to maintain, the IR cameras are typically em- ployed to supplement or replace the need for high mast lighting that is required for most CCD cameras to be effective in darkness. These lights are expensive to install and maintain and are still ineffective in rain or fog. However, the IR cameras give the user a complete solution for night time surveillance even in adverse weather conditions. Additionally, IR technology gives the freedom of seeing at night without being detected. Since thermal cameras are a passive technology, they do not emit a beam of energy that can be traced back to the user. Building infrastructure assessment is another function of IR technology. The use of thermal imaging allows an inspector to see support structures of a wall that are otherwise hidden behind the visible exterior because the surface directly attached to the brace becomes a different temperature. The in- spector can determine if the building infrastructure has been compromised and/or meets code regula- tions. Fig. A and B illustrate both internal and external walls as seen through an IR camera where the support braces are clearly visible. Figure A Figure B Similarly, inspectors use IR cameras to check for water damage, termite damage and HVAC inspec- tion, insulation and water infiltration quality Thermal imaging is also commonly used to locate objects buried under the ground such as land mines, 2100 Crystal Drive pipelines, storage tanks, etc. Additionally, thermal cameras are used to locate leaks in buried pipes. Suite 650 If an underground pipe containing chemicals, fluids, gas, oil, etc. leaks it changes the temperature of the soil in vicinity of the leak. Therefore, an overhead thermal image of the area allows the user to ac- Arlington, VA 22202 curately determine the location of a leak. T+ 1.866.458.ICXT
  22. 22. IR SPECTRAL BANDS AND TECHNICAL PERFORMANCE NOTES The word “infrared” refers to a broad portion of the electromagnetic spectrum: everything between visible light and microwaves. Much of the infrared range is not useful for ground- or sea-based im- aging because it is blocked by the atmosphere. The remaining portions of spectrum are often called “atmospheric transmission windows,” and define the infrared bands that are usable on Earth: Near 13 Chris Douglass Infrared (NIR), Short-Wave Infrared (SWIR), Medium-Wave Infrared (MWIR), and Long-Wave In- frared (LWIR). Atmospheric transmission of infrared bands (Courtesy Raytheon) Since NIR and SWIR are so near the visible bands, their behavior is similar to more famil- iar visible light. Energy in these bands must be reflected from the scene in order to pro- duce good imagery, which means there must be some external illumination. Both NIR and SWIR systems can take advantage of sunlight, moonlight, starlight, and an atmospheric phenomenon called “nightglow,” but typically requires some type of artificial illumination at night. Arrays of infrared Light Emitting Diodes (LEDs) often provide a very cost ef- fective solution for short-range illumination, but achieving good performance at distances of over tens of meters requires more directed illumination. Typical medium to long-range systems employ a focused beam from a laser or specialized spotlight, though special consideration of eye-safety issues is required. While NIR and SWIR imaging systems often employ sensors that are more exotic than those found in consumer-grade camcorders and digital cameras, glass is transparent to wavelengths as long 3μm, so normal lens systems can be used and windows can be seen through. Because NIR has a wavelength lon- ger than visible light, and SWIR a wavelength that is longer still, energy in these bands is scattered less by particles suspended in the atmosphere. This means that SWIR, and to a lesser extent NIR, systems are tolerant of low levels of obscurants like fog and smoke. The MWIR and LWIR bands are often called “thermal” bands because a typical scene emits radiation in these ranges. An imaging system that operates in these ranges can be completely passive, requiring no external illumination because it is able to sense the energy that is radiated directly from objects in the scene. Two major factors determine how bright an object appears to a thermal imager: the object’s temperature and its emissivity. As an object gets hotter, it radiates more energy and appear brighter to a thermal imaging system. Emissivity is a physical property of materials that describes how efficiently it radiates. Because cloth has a lower emissivity than skin, to a thermal imager cloth will appear darker than skin even when both are exactly the same temperature.
  23. 23. At these long wavelengths, infrared radiation be- haves differently from visible light. Glass is opaque in the LWIR band, and blocks most energy in the MWIR band. Consequently, LWIR and MWIR sys- tems cannot use inexpensive glass lenses, but are forced to use more exotic materials like silicon or germanium. Glass windows are also not transparent in these bands, so they appear brighter or darker ac- cording to their temperature. Since radiation in the MWIR and LWIR bands is not transmitted by water, rain can coat a scene and wash out much of its ther- mal contrast resulting in a duller image. Atmospheric obscurants cause much less scatter- ing in the MWIR and LWIR bands than even the SWIR band, so cameras sensitive to these longer wavelengths are highly tolerant of smoke, dust and fog. Even small effects like atmospheric turbulence can add up over very long distances to impact range performance, allowing LWIR an edge over MWIR. Hotter objects emit more of their energy at shorter wavelengths. The peak emissions of an object at room temperature falls in the LWIR band, so for ob- jects at normal earthly temperatures, a MWIR system must be more sensitive than a LWIR system to achieve identical imaging performance. The emissive peak of hot engines and exhaust gasses occurs in the MWIR band so these cameras are especially sensitive to vehicles and aircraft, but since hotter objects emit more total radiation, they are still easily detected by LWIR imagers. 2100 Crystal Drive Suite 650 Arlington, VA 22202 T+ 1.866.458.ICXT
  24. 24. IR TECHNOLOGY PARAMETERS TECHNICAL AND TRADEOFFS NOTES There are many technical parameters used to describe the performance of infrared detectors and cam- era systems. They are ways to help quantify image quality and predict range performance. Among others, NETD measures image quality. Responsivity and thermal time constant are parameters of the detector and help understand the quality of the detector. Finally, MTF and MRTD measure bar targets 14 Jeff Van Anda of varying size and distance apart. The performance at these differing spatial frequencies can simulate range performance. NOISE EQUIVALENT TEMPERATURE DIFFERENCE (NETD) NETD is the most commonly discussed IR technical parameter. It is a measure of noise in an IR im- age. It therefore directly relates to the overall quality of an image. In the audio world, before an audio signal is converted to sound by speakers, there could be noise on top of the desired audio signal. Once it is converted to sound, this noise can be measured in terms of sound, or decibels. The same is true for noise on top of an infrared signal. Once the infrared signal is converted to an image representing temperatures, the noise can be measured in terms of temperature, or degrees. This measurement is NETD, or the equivalent temperature difference of the noise. Typically NETD is expressed in units of Kelvin (K). Cooled infrared camera systems typically have low noise levels, in the range of 10 – 30mK. Uncooled infrared cameras systems are typically a little noisier, in the range of 30 – 120mK. Noise in an image can be spatial or temporal. Spatial noise is noise across the image at any given point in time. It is perceived as an unchanging fixed pattern on top of the image. Temporal noise is noise at any point in the image over time. It is perceived as the static that moves in an image. NETD is typically the measure of both these noise types. RESPONSIVITY Responsivity is the measure of how much an infrared detector’s output changes given a certain tem- perature change. It is typically represented as mV/K. Each element in a detector outputs a voltage that represents the amount of signal, or temperature, the element is “seeing.” Higher responsivity cre- ates better image quality. As the amount of signal goes up, the signal to noise ratio increases which, by definition, decreases NEDT. This comes at a price however. Detectors can only output a specific range of voltages. In addition, analog to digital (A/D) converters only have a specific range of voltages they can input. Increasing the responsivity decreases dynamic range as smaller temperature differences will consume the available output of the detector or input of the A/D converter. In arrays with many detector elements, the reponsivity differs from element to element; this is one of the reasons long cali- bration procedures are necessary. In order for an infrared picture to be usable, the reponsivity across elements has to be normalized in post processing. THERMAL TIME CONSENT Thermal time constant refers to the amount of time it takes a given pixel to change from one tem- perature to another. Since cooled detector materials do not hold the temperature they are sensing, they have no thermal time constant. So, this parameter is not relevant to cooled detectors. Resistive bolometer elements actually change resistance based on the pixel itself becoming a proportion of the temperature it is sensing. The thermal time constant is the amount of time it takes a resistive pixel element to change from one temperature to another. The result of a long thermal a long time constant is smearing of objects, particularly of very hot or very cold temperatures, when moving in an infrared image. The longer the thermal time constant the longer it takes to change from one temperature to the next, and therefore more smear. MODULATION TRANSFER FUNCTION (MTF) MTF is the ability of an optical system to transfer contrast at a given resolution. It is measured by looking at four bar targets. Each target has slightly smaller bars and less distance between the bars. These targets are said to have varying spatial frequencies. The targets are placed at a specific tem- perature that is different from their surroundings. The peak to peak output of the target against its surrounding is measured. As the spatial frequency becomes no longer resolvable, the two tempera- tures will blend together reducing the peak to peak output. The results from this can then be related to range performance as smaller targets can simulate objects at farther distances.
  25. 25. MINIMUM RESOLVABLE TEMPERATURE DIFFERENCE (MRTD) MRTD is measured by also looking at a four bar targets of varying spatial frequencies. For this test, the temperature of the target is adjusted in small steps (both from hot to cold and cold to hot) toward the surrounding temperature until the operator can no longer resolve the target. The difference between the two temperatures at this point is the minimum resolvable temperature difference. This measure can relate to range performance also however since a human element is involved it can be a more subjective test. Some people still prefer to also understand MRTD of system, however, for additional performance criteria 2100 Crystal Drive Suite 650 Arlington, VA 22202 T+ 1.866.458.ICXT
  26. 26. RANGE PERFORMANCE MODELING TECHNICAL NOTES The Night Vision Thermal Imaging Systems Performance Model, or NVTherm, is an extremely useful tool that is commonly used to calculate range performance, which is the distance upon which a given thermal imager can produce an image of a given target. This image produced is then classified as detection, recognition or identification depending on the quantity and quality of data collected about 15 Justin Thompson the target. This tool was developed by the U.S. Army Night Vision and Electronic Sensors Directorate and can be used to predict range performance for a variety of mid-wave infrared (MWIR) and long- wave infrared (LWIR) sensor types.i It considers the influence of many parameters including, but not limited to, atmospheric attenuation, system noise, detector pitch and array size, optical blur, lens attenuation, focal length, aperture size, even the type of monitor being used. The model then uses these parameters to calculate estimates for expected range performance at detection, recognition and identification levels. As noted above range performance estimates are typically described in terms of target detec- tion, recognition or identification. These terms are clearly defined but often misunderstood. Target detection suggests that a target may be seen in the image as a small, but detectable, blur. It simply means that the target is visible with minimal number of pixels, as few as one pixel element, and that there is a reasonable probability that what is visible is something of interest.ii It is very useful for knowing when an area needs further investigation because some unknown intruder is present. Classical recognition, or class discriminationiii, is a commonly misunderstood range performance category. Recognition of a target is being able to discern it with sufficient clarity that its specific class can be differentiated.iv This requires a small cluster of pixels on the target, and suggests that the operator could distinguish between a person and a car, or between a truck and a tank, etc. Often an operator will want something more than the stan- dard definition for recognition when evaluating a potential threat in the image making it important to choose a camera system that offers a range performance comfortably greater than what your expected requirements really are. Identification, or object discriminationv, of a target implies that there is an ability to discriminate be- tween objects, or to discriminate the correct vehicle, not just the vehicle Neglecting atmospheric effects, the recognition range would be 25% of the detection range, and identification range would be 12.5% of the detection 2100 Crystal Drive range. In the real world we cannot neglect atmo- Suite 650 spheric effects so the recognition and identifica- Arlington, VA 22202 tion ranges are generally cut even more than these percentages – the extent to which is determined by the atmospheric attenuations assumptions that are T+ 1.866.458.ICXT used (e.g. clear weather is obviously going to yield greater range than a hazy day). i U.S. Army Night Vision & Electronic Sensors Directorate (NVESD), Night Vision Thermal Imaging Systems Performance Model User’s Manual & Reference Guide, Rev 7, 2002 ii NVESD, section 2.8.4 iii Holst, Gerald C. Common Sense Approach to Thermal Imaging. Section 18.3.3 Discrimination. SPIE Press / JCD Publishing, 2000. iv NVESD, section 2.8.6 v NVESD, section 2.8.5 vi NVESD, section 2.8.5
  27. 27. UNCOOLED IR DETECTORS TECHNICAL NOTES Uncooled thermal imaging dates back to the mid-1960’s with the pyroelectric vidicon. This product was used in the production of fire-fighting cameras. Between that time and the mid-1990’s, most ther- mal imaging has been of the cooled variety, meaning that the sensor elements are cooled to -200 C, or -321 F. During this time research into uncooled thermal imaging was pursued in order to reduce cost, 16 Jon Van Anda weight, and increase reliability. In the mid-1990’s cooled arrays were very common, and uncooled arrays were coming out of the research labs and moving into production cameras. The main types of arrays today are known as microbolometer using vanadium oxide, microbolometer using amorphous silicon, and BST (barium strontium titanate). These three types have made the most significant mar- ket change of any uncooled technologies. VANADIUM OXIDE MICROBOLOMETERS The detector works by measuring the temperature of each of its pixels. The material is designed to change its electrical resistivity as the temperature changes. The temperature of the scene is focused on the material by the lens system. The detector has a quite small mass, and is thermally isolated from its supports so that its temperature changes rapidly with the small amount of focused energy. The material varies in its reaction to temperature across the array, so precise calibration needs to be done by the processing electronics in order to gen- erate a clear picture. So that the calibration data can be measured, it is common to have a shutter that can close off the optics. This calibration is done while the camera is operating every couple of minutes (between 5 and 60, depending on the design). The calibration is objectionable because the image is frozen for approximately 1-2 seconds, while it is being accomplished. Microbolometers are generally temperature stabilized by means of a thermo-electric (TE) device. The TE device heats or cools the detector in response to a particular voltage. Within the past 3 years, re- search has been successful in better understanding the detectors response to temperature. As a result, it is becoming more common that these detectors no longer need temperature stabilization, and are being delivered to customers without TE devices. These detectors are used in a variety of roles. There are several manufacturers, and they are spread across the United States and Europe. There are also several suppliers who sell the detector as a com- ponent, ready to be coupled with processing electronics, and then designed into a full camera system. They are generally a little more expensive then BST, however the image quality is often better depend- ing on how the system is designed. AMORPHOUS SILICON MICROBOLOMETERS The largest application for Amorphous Silicon (ASi) has been fire-fighting. Historically, they are lower resolution and lower frame rate (20 Hz) until recently when upgrades have been developed. ASi has been positioned as a lower cost alternative to Vanadium Oxide because it can be made with foundry machinery common for the manufacture of other electrical components, without many of the special- ized processes of other detector technologies. To use the detector, the system designer must focus the infrared energy on the detector, which heats the detector elements, just like vanadium oxide detectors. When the elements change temperature, they also change resistivity. The processing electronics collect data about the resistance change to generate the picture of the scene. A shutter is often employed every couple of minutes, which freezes the image for 1-2 seconds, to collect calibration data.
  28. 28. ASi detectors also employ a TE device to stabilize their temperature. Similar to vanadium oxide, design- ers have developed algorithms to use the detector without a TE device, allowing a further reduction in cost. Typically, ASi detectors have a slight disadvantage to Vanadium Oxide in image quality, while a slight advantage in cost. BST (BARIUM STRONTIUM TITANATE) The detector outputs infrared data by looking at the change in temperature over time. Therefore, if the detector were held fixed, looking at a stable scene, it would eventually (over a couple of seconds) show nothing at all, just one shade of gray. To correct this problem, designers added a shutter wheel in front of the detector that spins continuously at a known rate. The wheel alternates between semi-transpar- ent and transparent, which the readout electronics compare to determine the scene. A by-product of this moving disc is that there are fuzzy edges around everything in the scene. The fuzzy edge affects the image quality and the range performance of the system. BST is an uncooled technology meaning that it does not require cryogenic cooling. It does however re- quire temperature stabilization. Typically, the detector is warmed slightly and then a regulator keeps the temperature constant. 2100 Crystal Drive Suite 650 Arlington, VA 22202 T+ 1.866.458.ICXT