Electronic Warfare is a
discipline characterised by rapid change, where openings and
vulnerabilities are exploited quickly and always to the detriment of the
slower party. As with Electronic Warfare at lower frequencies, the
electronic battle at optical frequencies is equally intense and can have
an equally decisive impact upon a conflict. As an instance the Russians'
deployment of the Fulcrum and Flanker fighters equipped with laser
ranging Infrared Search and Track sets (IRS&T) was a successful
attempt to circumvent overwhelming Western capability in microwave
Electronic Warfare techniques by shifting the aircraft's key
fire-control sensor into the optical frequency range.
This move was hardly unexpected as Russian efforts in
IRS&T equipment development have been noted for at least half a
decade, what is truly alarming is the lack of response to this event by
Western air forces. As emerging Stealth technologies erode the
capability of microwave radar based systems to provide missile guidance
and fire control parameters, optical systems in the visible and infrared
bands will assume an ever increasing importance. Optically guided
weapons will proliferate in land, sea and air battle scenarios at the
expense of conventional systems. To these issues must be added the
practical implications of an air and ground environment where laser
beams of eye-damaging wavelengths and power levels are used by almost
every fire control system, be it air-to-air, air-to-surface or
surface-to-surface. As with microwave EW, a clear understanding of the
technology of electro-optical systems is essential to the conduct of
offensive and defensive electro-optical warfare. The following will
hopefully clarify some of the issues.
Infrared Trackers
The infrared tracker is the technology at the core of infrared
missile warning systems, heatseeking missiles and IRS&T sets. Its
function in the simplest of terms is to scan a field of view (FOV) for
any sources of infrared radiation of a particular wavelength or band of
wavelengths and to measure with some accuracy the angular coordinates of
these sources. These sources will be exhaust plumes and airframes which
radiate heat and thus infrared radiation (for more detailed discussion
of the infrared signature of an aircraft, please see TE March 1982, TE May 1987).
There are numerous ways of building infrared trackers, the
most important of these are the reticle, rosette scanning and focal
plane array techniques, which are common to heatseeking weapon guidance,
infrared warning systems and IRS&T sets.
In spite of their differences in construction and performance,
all infrared trackers will have some features in common.
All will employ either an array of or a single detector
element. A detector is a small piece of a semiconductor material which
is photosensitive, ie its electrical properties will change when
illuminated by light of a given wavelength. Numerous materials have been
used for detectors over the last four decades, with PbS very popular in
early days, later yielding to InSb. The favoured materials today are
Mercury Cadmium Telluride (HgCdTe) which is unfortunately very tricky to
fabricate detectors from but provides superior sensitivity to longer
wavelengths and Platinum Silicide which is less sensitive but far easier
to process.
The performance of any detector material is usually judged by
its sensitivity over some range of wavelengths (colours). Sensitivity
is a measure of the amount of electrical response the detector will
provide to some given level of incident radiation, ie the more sensitive
the detector material the fainter the radiation it can detect.
Sensitivity will vary with wavelength as a result of quantum physical
effects in the detector and more mundane considerations such as detector
surface finish. The quantum physical aspect is given by the
photoelectric effect where photons of impinging light or infrared excite
electrons (carriers for the physics oriented reader) within the
material, if the photon is energetic enough (ie its wavelength short
enough) the electron will be dislodged from its place in the detector's
crystalline lattice and becomes detectable as a minute electrical
current. The minimum photon energy required to dislodge a carrier
determines the longest wavelength detectable by the detector material
and hence the lowest possible temperature target which may be detected.
This is important as the infrared signature of a rocket exhaust plume,
engine tailpipe, fuselage rear end and airframe all differ in peak
wavelength, the hotter the target the shorter the brightest wavelength
and the broader the range of wavelengths emitted (readers are advised to
look up blackbody radiation in any decent physics textbook). A 2 micron
band detector will thus see only rocket exhaust plumes, afterburner
plumes and tailpipe cavities, a 4 micron band detector all of the above
and hotter parts of the airframe from all aspects, while an 8 micron
band detector all of the above and the whole airframe or even its wake.
As it turns out the longer wavelength detectors must be cooled for
operation, as room temperature itself will excite enough electrons in
the detector to completely swamp any photoelectric current from a
sought target.
Ideally the designer of an infrared tracker would seek a
detector matched to the wavelength of the sought target while requiring
minimum cooling. Cooling a detector is another can of worms and many
techniques have been used in practice. Missiles will employ gas or
thermoelectric cooling, whereas IRS&T equipment may also employ
closed cycle refrigeration to reduce turnaround time. Gas cooling
involves the rapid expansion of a compressed gas usually provided from a
bottle installed in the launcher or launch system, as instances the
Nitrogen cooled USN AIM-9D PbS detector or Argon cooled USN/USAF AIM-9L
Sidewinder InSb detector. Thermoelectric cooling is achieved via the use
of a Peltier heat pump, passing an electrical current through such a
device cools one end of it and heats the other, the USAF AIM-9E/J's PbS
detector was cooled in this fashion.
The choice of detector material cooled to an appropriate
temperature will determine the sensitivity and longest detectable
wavelength seen by the infrared tracker. Most detectors are sensitive to
visible light due to the nature of the photoelectric effect and this can
cause problems in the real world. The sun is a bright source of visible
and infrared light (6000K blackbody radiator) and any highly reflective
object such as a cloud will also appear as such to a tracker, this is
the reason why inferior heatseeking missiles chase the sun or clouds.
The solution to this problem is simple in concept although often tricky
in implementation; the detector will view the outside world through an
optical filter which will suppress all visible and infrared light of
those wavelengths which are not desired. Filters are usually
implemented by depositing very thin layers of glass with alternating
refractive indices and thicknesses of the order of the wavelengths of
light involved, upon the surface of the glass window used by the
detector. Clever choice of thicknesses will cause an effect termed
interference; undesirable wavelengths are reflected and the remainder
passed. The gold hue apparent on the nose windows of infrared missiles
is such an interference filter.
A cooled detector with its optical filter is the core of an
infrared tracker. What remains is some mechanism to focus the incoming
light on to the detector and point the detector's field of view within
the tracker's field of view. This mechanism is the tracker's optics,
which are often stabilised in pitch and yaw to provide a jitter free
output.
The oldest and perhaps dominant optical arrangement today is
the rotating reticle scheme, first put to use by the Germans on A-4
ballistic missiles. The rotating reticle scheme (see TE March 1982 for
illustration) uses a rapidly spinning reticle (or mirror) which has a
pattern of transparent spokes etched on half of its surface, the other
being semi-transparent, the detector views the outside world through the
whirling pattern of the reticle. The pattern chops all incident light
and the resulting output from the detector is a train of pulses
synchronised to the reticle. The pattern on the reticle is designed such
that the size of the pulse train is proportional to the angular
displacement of the target relative to the axis of rotation of the
reticle. The remaining angular coordinate is derived from the relative
timing (phase) of the pulse train relative to the reticle rotation.
The rotating reticle tracker is penalised by the need for
precision fabrication of the reticle which must be exact mechanically
while also often doubling up as an infrared filter, alternate
construction employs a concave mirror with the reticle pattern etched on
its surface. The strength of this technique lies in the simplicity of
the analogue electronics required to decode the target position. Its
practical limitations lie in its narrow field of view and high
susceptibility to amplitude modulation (AM) jamming, the simple
technique of flashing an infrared source at a rate comparable to that of
the reticle rotation can cause the electronics to lock on to the
jammer's pulse train and drive the tracker away from the target (as with
AM jamming of conical scan radar). A newer and mechanically/optically
simpler arrangement is that of the rosette scanning tracker. A rosette
scanning tracker will usually employ a single fixed detector and moving
optics which have a small instantaneous field of view. The optics then
scan a much larger field of view in a rosette pattern, which has the
appearance of a flower with multiple petals. The strength of this
technique lies in the simplicity of the scanning mechanism and a minimum
of hardware in the optical path; the penalty is in the additional
complexity of the electronics required cf. a rotating reticle tracker.
Jam resistance of these trackers has not been discussed at
length in the open literature, no doubt for fear of compromising weapons
such as the FIM-92 Stinger SAM.
The newest family of optical trackers are imaging or Focal
Plane Array (FPA) trackers, which will become the dominant tracker in
the next decade or so. A FPA tracker will employ a planar array or field
of detector elements which stare at the whole field of view
simultaneously. By reading the outputs of the detectors with electronics
embedded in the same chip/carrier, an image of the viewed scene can be
recreated for viewing by an operator (eg FLIR applications) or analysis
by an image processing computer to detect and track targets. The
construction of the necessary optics is simple with no moving parts - a
lense/ window/filter and a support to position the chip carrier within
the required focal plane of the optics, this arrangement is thus cheap
to produce, align/ calibrate and has a high tolerance for acceleration
thus being very suitable for missile guidance. Four technologies are
being used or evaluated for use in imaging/array trackers; Mercury
Cadmium Telluride(HgCdTe), Platinum Silicide, Indium Antimonide(InSb)
and Iridium Silicide.
HgCdTe is a well established material with excellent
sensitivity extending down to the 8-12 micron band (cool targets, ie
FLIR applications, satellite tracking, detecting stealth vehicles ) but
it is difficult to fabricate arrays from because of a very large
variation in sensitivity from detector to detector. Such an array will
introduce clutter (noise) into the image it views and this will
understandably make it more difficult for the image processing algorithm
to sift targets from the background. As a result additional signal
processing is usually required to compensate for array nonuniformity. At
the time of writing Rockwell International were offering 128 x 128
element arrays in HgCdTe. The alternative to HgCdTe in 5.5 micron band
applications (warm targets such as airframes from all aspects, vehicles,
exhaust plumes) is InSb which is easier to fabricate and offers good
uniformity. 128 x 128 and 256 x 256 element arrays are offered by Amber
Engineering of Goleta, California, who are developing arrays for
civilian and military applications.
Another alternative is the use of Platinum Silicide which is
unfortunately about fifty times less sensitive than HgCdTe and is
spectrally limited to the 2.5-4 micron band(ie hot targets such as
aircraft/airframes, vehicles, missile exhaust plumes); its strength lies
in high uniformity of array sensitivity and ease of fabrication and thus
low cost, commercial Silicon fabrication techniques are used as for eg
MOS memory chips. This infers another major advantage, the ability to
embed within the same slab of Silicon all the necessary electronics to
scan and read out the image from the array, thus radically cutting the
cost and complexity of the optical system as a whole.
Platinum Silicide arrays of 512 x 512 elements are now offered
by Hughes, the highly sought (by the military) 1024 x 1024 array size is
expected to become available in the next two years as the technology
matures. Iridium Silicide offers like HgCdTe operation in the 8-12
micron band but is very immature as a production technology today.
The choice of detector array having been made for the
application, the designer of the tracker will have to select a tracking
algorithm and suitable image processing computer or logic for the task.
In established applications such as missile guidance (eg AGM-65A/B/D
Maverick) tracking can be done with a hardwired electronic contrast lock
which places 'gates' about an operator specified area of contrast in
the viewed scene or image. The position of the 'gates' in time relative
to the video signal (ie scanned line by line picture) from the array
readout electronics then determines the angular position of the target.
A newer technique is to save the image in a dedicated area of computer
memory and then have a powerful processor sift through the image and
accomplish the above in software; this strategy offers flexibility in
algorithm performance and the ability to autonomously search for and
recognise targets.
As with rosette scanning trackers little has been written in
the open literature about suitable techniques for jamming imaging
trackers. At a guess possible techniques could involve seduction of the
tracking gate with flares (ie gate stealing) or causing the target to
scintillate with phased IR jammers at its extremities (eg wingtips),
this latter technique would cause jitter in a centroid tracking
algorithm. Its effectiveness would be very limited, only during the
terminal phase of a fighter or missile attack with the purpose of
forcing a near miss just outside proximity fuse range or spoiling the
gun aim. In general the passive nature of the IR tracker makes it all
the more difficult to jam as the jammer has no means of directly
assessing its effect on the threat as with microwave jammers. While all
designs will exhibit generic and specific vulnerabilities to jamming,
good intelligence and other sensors may be required to identify the IR
tracker equipped threat in order to select an appropriate jamming
technique. An alternative to jamming is the use of a suitable laser to
blind the tracker by saturating its detector(s), this is a brute force
method not unlike barrage jamming which can be very effective against
trackers with a slow auto gain control response. By the time the tracker
recovers the target will have drifted out of its field of view and the
threat has been defeated. In the context of defeating infrared
trackers, Stealth or more formally Low Observables technology is an
potent alternative. A LO design will conceal its engine hot end from
trackers by reducing the viewing aperture (ie range of viewing angles).
This is achieved by exhaust placement (cf Northrop B-2A upper surface
exhausts), shape and the use of louvres and shielding structure. Exhaust
plumes are then cooled to a temperature at which the brightest infrared
colours are absorbed by atmospheric carbon dioxide and water vapour
over fairly short distances. The cumulative effect of these measures is
that an infrared tracker can detect a LO design only from a very narrow
range of angles, usually from astern/above, at relatively short range
under clear weather conditions only. Tracking is only possible when
these conditions are maintained, ie manoeuvre by the LO target or use
of cloud will defeat the tracker. As it is most apparent, there is a
broad range of tracker types available and each will have its
appropriate offensive and defensive application. Offensive applications
involve both weapon guidance and fire control systems.
Infrared Search and
Track Systems
The IRS&T set has been in use, in its various guises,
since the 1960s when USAF air-defence fighters such as the F-101B Voodoo
and F-106A Delta Dart carried nose mounted nitrogen cooled lead selenide
cross array IRS&T sets to detect high flying Russian bombers such
as the Bear and Bison. The strength of the IRS&T is in its ability
to detect and track a target passively and thus undetectably.
This is considered to be a major asset in air defence
operations as it vastly reduces the warning time to attack on a target.
In a situation where a large bomber target may carry powerful onboard
jamming equipment the use of a Air Intercept (AI) radar by a interceptor
will not only provide the bomber with timely warning of impending
attack, it will also provide the bomber with an opportunity to defend
itself by jamming. This strategy was adopted by the US Navy which
equipped its F-4B air defence fighters with chin mounted AAA-4
IRS&T sensors, used in part to cue early AIM-9B Sidewinder seekers,
these were deleted from later versions as the aircrafts' role shifted to
air superiority/strike. The new F-14A became the Navy's principal air
defence fighter and it also carried an IRS&T set, although these
have been largely replaced by AXX-1 Television Camera System TV
telescopes.
The Russians have certainly never been shy about copying
Western designs and they too deployed IRS&T sets from the 1960s
onward, fitting these to Foxbat and Flogger aircraft of the
IA-PVO-Strany. While this equipment has not been publicised in the Free
World it is likely to be similar to late sixties US designs, providing a
2 micron band detection capability in clear weather conditions.
The subject of greater concern is the current generation of
Soviet IRS&T equipment which is fitted with a boresighted laser
rangefinder and in effect provides the same fire control functions as an
air intercept radar, under appropriate atmospheric conditions. The
fundamental failing of the passive IRS&T is its inability to measure
range to the target, although some estimate could be made given the
intensity of the target's infrared signature. The ability of a Fulcrum,
Flanker or Foxhound to carry out a full GCI(intercept) and missile or
gun attack with electro-optical systems alone renders the hundreds of
pounds of microwave jammers aboard Western strike aircraft quite
ineffective.
The Soviet thrust into the area of Infrared Search &
Track/Laser fire control has been paralleled in the Free World by the
deployment of FLIR/laser systems such as Lantirn on Western tactical
aircraft such as the F-15E and F-16C/D.
Thermal Imaging/Laser
Systems
Forward Looking InfraRed (FLIR) is a generic term referring to
the broad range of thermal imaging equipment. The first true thermal
imagers appeared in the seventies when the USAF deployed a range of
systems based upon the Texas Instruments family of thermal imaging
modules (TE March 1984). The technology has since proliferated and is
now used on almost every class of platform. Of particular interest here
is its application to fire control systems used by tactical aircraft,
such as the AVQ-26 Pave Tack carried by the RAAF's F-111C, battlefield
helicopters, such as the TADS/PNVS turrets carried by MDC AH-64A Apache,
and SAM/AAA systems, such as the Philips Trackfire .
In all of these applications the stabilised thermal imaging
telescope is boresighted with a laser rangefinder and thus can provide
all three coordinates of a target - this allows a fire control computer
to calculate a fire control solution for an unguided weapon or gun or to
set up optimal release conditions for a guided weapon.
The laser however also doubles up as a target designator by
transmitting an appropriate pulse code. A rapidly flashing spot of
appropriately coded laser illumination on a target is used to guide
semi-active laser homing weapons, which home in on the laser spot. Both
conventional laser guided bombs such as the GBU-10/Mk-84 and
GBU-12/Mk-82 used by the RAAF and laser guided missiles such as the
AGM-65E Maverick or AGM-114A Hellfire use this technique, relying on a
simple four quadrant detector (ie four detectors arranged in a Maltese
cross pattern) arrangement. The accuracy of these weapons is very high,
with average miss distances of feet.
There is however a more insidious aspect to these systems.
Nearly all laser rangefinders and designators employ Nd:YAG 1.066 micron
near infrared lasers which have both the power level and short
wavelength required to damage the human retina. Partial and complete
blindness may result from direct or indirect sighting of the beam from
such a laser. The presence of laser rangefinders on almost all armoured
vehicles, tanks and battlefield close air support aircraft and
helicopters renders battlefields optically hazardous to the unprotected
eye. The Russians have taken a particular liking to the use of laser
rangefinders as antipersonnel weapons, sweeping the battlefield ahead of
their advancing armour to damage eyes and less robust optical fire
control sensors. The USAF had investigated this application under the
Compass Hammer program, proposing an optical tracker capable of
detecting gun muzzle flashes and directing a powerful blue green laser
at the source to blind the gunner or fire control system. A Westinghouse
designed Advanced Optical CounterMeasures pod was successfully flight
tested in the mid eighties with planned application to the B-52, however
no information appears to be available on deployment plans. The
countermeasure is the wearing of protective goggles, usually employing
multiple interference filters for specific laser wavelengths, all the
time by all personnel. Invisible and narrow laser beams offer no warning
to the victim.
High Energy Laser Weapons
The High Energy Laser (HEL) is a weapon which is within a
decade of deployment, if development is sustained. While often ridiculed
as science fiction technology, the current generation of near and
mid-infrared chemical lasers has demonstrated usable power levels of
hundreds of kilowatts and the ability to deliver power through beam
folding optics to a beam director and target. The early eighties saw the
demise of the USAF Airborne Laser Lab (ALL) program (see TE Dec 1981)
during which a carbon dioxide 10.6 micron gas dynamic laser weapon on
an NKC-135A airframe successfully engaged and shot down transonic
drones.
Objectives met, the funding was stopped. Work is however
currently being done on the US Navy's land based Miracl deuterium
fluoride 3.8 micron chemical laser which coupled to a Sealite beam
director has been involved in successful trials against supersonic
Vandal missile targets. The USN sees the HEL Weapon System (HELWS) as a
shipboard point defence weapon with the ability to kill aircraft and
missiles in a high density air/sea battle (interested readers are
directed to USNI Proceedings Oct 1988). Coupled to Aegis battle
management, optical fire control will be used.
High power tunable Free Electron Lasers are also under
development for SDI applications (Lawrence Livermore labs and TRW), but
these devices are so huge that they are unlikely to be practical in air
defence applications.
Heatseeking and Imaging
Weapon Guidance
Heatseeking and imaging guidance have been used in missiles
for over two decades. Heatseeking guidance has progressed from simple
rotating reticle tail aspect only capable missiles to jam resistant
flare rejecting all aspect weapons. The most recent variants of the
established Sidewinder, the Mike and Romeo are described as
substantially better than their predecessor, the Lima model which had a
success rate of 86% in the Falklands conflict. Similarly heatseeking
SAMs have progressed considerably, with the current FIM-92C Stinger
offering a dual colour rosette scanning seeker with a field programmable
guidance processor. The Russians have also expended considerable effort
in this area with the new SA-13 Gopher fitted with a cooled dual colour
seeker. All aspect heatseekers are thus a real threat to all classes of
aircraft at all altitudes and trivial countermeasures such as flares are
largely ineffective with newer weapons. Capable jammers, warning
systems and sufficient aerodynamic performance to sustain manoeuvre are
thus necessary for survival.
Imaging missile guidance saw its first major application
during the SE Asian conflict when the USAF demolished several communist
targets with HOBOS TV guided bombs. Soon followed by the TV guided
AGM-65A/B Maverick tank killing missile, imaging guidance found a wide
range of applications all requiring pinpoint accuracy. Imaging weapons
will employ either conventional television or thermal imaging to provide
an operator with a view of the target. An automatic tracker is then
used (as described above) to ensure that the weapon tracks the selected
target until impact. Simpler weapons such as Maverick are
Lock-On-Before-Launch where the launch platform flies to line-of-sight
to the target, the operator then locks the missile on and launches it,
more complex weapons such as AGM-84E SLAM and GBU-15 are
Lock-On-After-Launch where a datalink carries the TV image to a remote
launch aircraft. The latter family of weapons is about to expand with
the deployment of optical fibre guided weapons which unwind a hair thin
optical cable as they fly, this cable carries the TV image from a nose
camera back to the launch platform which sends guidance messages back to
the weapon via the same cable. Unjammable and electromagnetically
silent, these weapons (missiles, glidebombs) will be used against
aircraft, helicopters and surface targets.
The Electro-Optical
Defensive Suite
Defending an aircraft against optically guided weapons and
fire control systems will involve detection, tactical flying and jamming
where applicable. Detecting an inbound missile or tracking by a hostile
fire control system is a primary concern here for obvious reasons, and
may be accomplished by several means. Optically guided missiles are
passive and their only detectable emission is the heat of their exhaust
plumes. Two strategies are therefore used for detection and tracking.
The first is that of a missile warning radar which scans the airspace
about the aircraft looking for returns with the appropriate Doppler
for an inbound missile. The second strategy is the use of an
infrared detection set which is a tracker optimised for detecting and
tracking missile exhaust plumes. An example of the former strategy is
the ALQ-156, its strength lies in its ability to estimate the range of
the threat, its weakness is that it is broadcasting the location of the
aircraft it is protecting. An example of the latter is the AAR-34 as
fitted to the F/FB-111 or AAR-44 or 47 as fitted to C-130 aircraft. The
latter strategy offers truly passive operation and should not compromise
the position of the carrying aircraft.
Detecting tracking by a hostile optical fire control system
can be more difficult as its platform need not be easily detected.
Thermal imagers and IRS&T sets in passive (angular) tracking mode
cannot be detected unless one resorts to something drastic, such as
scanning airspace with a laser and looking for glint off the infrared
filter covering the threat's optics. Doing this may detect the bad guy
but will certainly compromise your location. However if the bad guy is
serious he will probably want to get range information for a fire
control solution and will thus turn on his laser rangefinder thereby
blowing his cover and intentions to a defending aircraft carrying a
Laser Warning Receiver (LWR). Devices such as the Perkin Elmer AVR-2
Laser Warning Receiver are analogous to radar warning receivers in that
they use a set of four suitable optical detectors and listen for the
pulse trains associated with particular fire control systems or guided
weapons, feeding warning information into existing aircraft/helo radar
warning equipment.
Once the threat has been detected defensive measures can be
taken. The objective of these is to cause the tracker in the threat
system to break lock. If the threat is a rotating reticle tracker (AAM,
SAM, IRS&T) dropping flares may seduce the tracker, just as rotating
the tailpipe away from the threat may break lock. The most effective
strategy is however jamming with a device such as the Northrop AAQ-4,
AAQ-8, MIRTS, Loral ALQ-123, MATADOR or Sanders ALQ-144. These devices
typically use an alkali metal vapour (eg cesium) lamp which is
electrically or mechanically modulated to flash with a selectable code
known to interfere with the tracker in the threat system.
A newer alternative may be the tunable free electron laser
which could be tuned to the most sensitive colour of the threat tracker
and jam it (as above). If the threat is an imaging tracker this may not
be adequate and terrain masking (ie placing a hill between oneself and
the bad guy) may be the only choice. A possible countermeasure as
suggested above would be the use of a laser to damage or dazzle the
threat tracker, the deployment of any such system has not been
publicised in the open literature.
The modern battlefield
environment is optically hostile and survival will eventually
necessitate the carriage of laser warning systems, infrared missile
detection equipment, infrared jammers and the wearing of laser
protective goggles by aircrew and footsoldier alike. The threat of
blinding by hostile and friendly laser equipment is very real on the
battlefield of the nineties and should be considered carefully . Unlike
situations such as nuclear, biological and chemical warfare which are
arguably unlikely to be encountered, optical weapons are here to stay.
FURTHER READING:
Hudson R.D. 'Infrared System Engineering', Wiley Interscience,
N.Y. 1969.
Fitts R.E. 'The Strategy of Electromagnetic Conflict', Ch.8,
Peninsula Publishing, 1980.
Readers interested in free electron lasers should read the
following: Freund H.P., Parker R.K. 'Free Electron Lasers', Scientific
American,April 1989.