13 Range and Angle Tracking
13 Range and Angle Tracking
13 Range and Angle Tracking
410 _____________________________________________________________________
411 _____________________________________________________________________ Note that the sign and magnitude of the error will drive the tracking error toward zero only over a limited range (in this case, about 20 samples) and the linear region over which the magnitude of the range error can be used to estimate the actual error is more limited (about 8 samples).
The range tracking loop uses this error to drive the smoothed estimate of the target range until the amplitudes of the early and late gate samples are equal. If, however, the error falls outside the bounds shown by the transfer function, then the error voltage falls to zero (in a noise free environment) and there is no driving function.
r =
2.5 2 S / N
(13.1)
Note that the tracking error will take on the dimensions of , so it can be a range error in metres or a delay error in seconds.
412 _____________________________________________________________________ Secondly, by sampling the echo signal close to its peak ensures that the measurement SNR is as high as possible. Typically a range tracker will include three gates, the early and late gates discussed earlier, and a sum gate which straddles the other two gates and samples the whole received pulse. This gate can be a physical realisation as shown in the figure below, or it can be the sum of the voltage signals from the early and late gates.
Sum Vsum Gate Early Gate VE Integ Late Gate VL + Integ
AGC
Transmit Trigger
Range ramp
Sample Trigger
Smoothed Range
To maintain a constant loop bandwidth, the error function VL-VE must be normalised otherwise targets with a large RCS, or closer to the radar will produce larger errors than one with a small RCS or further away even if the true range error is the same. Normalisation is achieved using an automatic gain control (ACG) driven by the sum channel range gated video signal to maintain a constant target amplitude in the receiver irrespective of the range or the target RCS The normalised range error drives a second order tracker, represented in the figure by a pair of cascaded integrators. In most modern trackers digital implementations of the tracking loop are used as they are easy to implement and can be optimised to suit specific conditions. Problems with pure digital implementations such as the one shown in the figure below is that the S&H and ADC must have sufficient dynamic range to digitise the full range of target amplitude variations. This can exceed 80dB. The primary object of such tracking filters is to minimise the output noise variance under specific conditions of input. (usually constant velocity).
Sample & Hold late Radar Video Range Trigger early Sample & Hold ADC L-E D R=K-------L+E Normalisation Function ADC
D R
Range Tracker
413 _____________________________________________________________________ Kalman filters and - trackers are widely used for track-while-scan and single target tracking applications. The - Filter The - tracker is a fixed gain formulation of the Kalman filter and is still widely used because it is easy to implement performs well in general. The - Tracker equations for range tracking are defined below Smoothing =R + R R R n pn n pn
(13.2)
(13.3)
(13.4) (13.5)
Rn = Measured Range R p ( n +1) = Predicted range after T seconds V = Predicted Range Rate after T seconds
p ( n +1)
= Predicted range at the Measurement Time R pn = Predicted Velocity at the Measurement Time V
pn
Ts = Sample Time , = Smoothing Constants It has been suggested (Benedict and Bordner) that to minimise the output noise variance at steady state, and the transient response to a manoeuvring target as modelled by a ramp function, then the gain coefficients are related as
2 . 2
(13.6)
Other criteria can also be used. For example, it has been suggested that the filter should have the fastest possible step response (critical damping), in which case the coefficients are related as
=2 .
(13.7)
414 _____________________________________________________________________ The actual values of the gains depend on the sample period, the predicted target dynamics and the required loop bandwidth. The main disadvantage of the fixed gain - tracker is that it estimates the position of an accelerating target with a constant lag. The magnitude of this lag can easily be estimated and the filter can use adaptive gains to improve the RMS tracking accuracy under these circumstances. The cost function that is minimised is [lag2 + r2]. For estimates that include accelerations, the -- tracker can be substituted. However, its performance if the target is moving with constant velocity is poorer than that for the simple - tracker because of the additional noise introduced by the acceleration estimate.
The Kalman Filter
A simplified view of the Kalman filter when compared to one of the sub-optimal filters is that it uses optimum weighting coefficients (somewhat analogous to and ) that are dynamically computed each update cycle. In addition a more precise model of the target dynamics can be used. The benefits of the additional complexity include Improvement in tracking accuracy A running measure of the accuracy Method of handling measurements of variable accuracy, non uniform sample rate or missing samples Higher order systems are easier to handle
Figure 13.6: The target echo and the split gate position at the start of the simulation
415 _____________________________________________________________________ To perform the split gate tracking function, three gates are shown, the early and the late gates straddling the sum (central) gate (o,+,*). Due to receiver noise, the tracking gates will move in a random fashion until they are reached by the target echo, at which time they will lock onto it.
Figure 13.7: Target echo and split gate position during tracking
Note that the early and late gates are sitting astride the target echo peak which would produce a small tracking error voltage that drives the tracking loop to maintain its position straddling the target echo perfectly, even though the latter is moving.
(a)
(b)
Figure 13.8: Target range, measured range and track error with time for (a) = 0.3 (b) = 0.1
416 _____________________________________________________________________ two coincide. In general this settling period can be reduced by injecting course estimates of the target rate into the track filter, or by starting out with a filter with a wider bandwidth and progressively reducing it as the tracking error decreases The tracking noise, being the difference between the measured and the true target position (for constant vel target) is inversely proportional to the filter bandwidth. The filter performance is often characterised by the ratio in the variances of the output and input noise. This is known as the Variance Reduction Ratio (VRR). Measurements made over n samples combine in the filter to provide an output whose RMS noise is reduced by 1 / n compared to a single pulse measurement. In terms of the equivalent noise bandwidth n of the range filter,
n= 2 n fr = 1 , VRR
(13.8)
where VRR Variance reduction ratio, fr Pulse repetition frequency (Hz), n - Bandwidth of tracking filter (Hz). For the split gate tracker, the filtered thermal noise output will therefore be
r =
2.5 (S / N )( f r / n )
(13.9)
The following figure shows the filter transfer function (for the position estimate) for various values of and (using the Benedict Bordner relationship). The filter half power bandwidth is measured at |H(z)| = 0.707.
Figure 13.9: The - filter transfer function for a sample rate of 1kHz
It can be seen that the filter bandwidth increases as increases, but that the relationship is not completely linear. As stated earlier in these notes, a filter can be
417 _____________________________________________________________________ characterised completely by its impulse response. This allows the VRR to be determined as a ratio of the power in the impulse response to the power in the impulse. This is valid because the frequency spectrum of an impulse is flat whereas the response will be low-passed as shown in the figure.Alternative methods of calculating the VRR is to inject white noise of known variance into the filter and to measure the variance on the output, or by integrating the area under the |H(z)|2 curve in the frequency domain.In general, the impulse response method converges very quickly and so has been used to determine the VRR plotted in the following figure. Using the VRR formulation, the filtered tracking noise can be rewritten as
r =
2.5 2 S / N
VRR .
(13.10)
(1 ) ,
(13.11)
where L Lag (m), & r& - Range acceleration (m/s2), Ts Sample interval (sec), - Filter gains. For a sample rate of 1kHz (Ts=1ms) and an acceleration of 1m/s2, the following figure shows the lag as a function of for a Benedict Bordner - filter. Because of the fast sample rate, this lag is insignificant unless the target acceleration is very high
418 _____________________________________________________________________
Figure 13.11: The - filter dynamic lag for a sample time of 1ms and an acceleration of 1m/s2 as a function of
Figure 13.12: RMS noise and dynamic lag and the root mean squared sum determines the optimum gain
Using the equations developed to determine the RMS noise and the dynamic lag, use Matlab to calculate the root mean squared sum of the two parameters. In this case the optimum position estimation gain will be = 0.29 from the graph. Using the Benedict Bordner relationship, the velocity estimation gain, = 0.049.
419 _____________________________________________________________________ For a 50Hz sample period, the filter bandwidth can be obtained from Figure 13.9 by interpolation and scaling to be about 3.5Hz
% determine the optimum tracking gains for an % ablag.m acc = 10; % target acceleration fs = 50; % sample frequency snrdb = 10; % SNR tau = 3; % Range gate size(m) snr = 10.^(snrdb/10); ts = 1/fs; % Calculate the RMS range tracking noise out of the split gate % from the pulse width and the signal to noise ratio sigma = tau/(2.5*sqrt(2*snr)); k=0 sigmaout=zeros(1,90); alp=zeros(1,90); lag=zeros(1,90); for alpha=0.1:0.01:1, k=k+1; beta = alpha.^2/(2-alpha); % position transfer function for the alpha beta filter b=([alpha, beta-alpha]); a=([1,beta+alpha-2,1-alpha]); % measure the power in the impulse response to determine the VRR sig=[zeros(1,500),ones(1),zeros(1,500)]; out=filter(b,a,sig); alp(k)=alpha; vrr=sum(out.^2); % calculate the noise after the track filter sigmaout(k) = sigma*sqrt(vrr); % calculate the dynamic lag from the formula lag(k) = acc*ts.^2*(1-alpha)/beta; end comb = sqrt(lag.^2+sigmaout.^2); plot(alp,lag,alp,sigmaout,alp,comb); grid xlabel('alpha') ylabel('dynamic lag and RMS noise (m)') title('Optimising the Filter Gain')
420 _____________________________________________________________________
In the typical configuration shown in the figure above, the standard pulsed time of flight measurement technique is followed by a second micro-controller that processes the measured data to estimate the position and speed of the target. Estimates the velocity can be made using a tracking filter of the kind discussed above, or by measuring the change in range as determined by the varying time of flight between successive pulses. From the figure below it can be seen that once the filter has settled, the estimates are smoother than those produced by the difference method. The amount of smoothing is determined by the value of the gains. In practical applications, the filter is seeded with a reasonable estimate of the rate based on the position difference over a few samples. This speeds convergence considerably.
Figure 13.14: Tracker performance comparison between an filter estimate of the speed and a simple difference shows the superiority of filtering
421 _____________________________________________________________________
% laser radar range tracking simulation % uniform distribution of range error +/-10cm % sample rate 20Hz xtar = 400; % start range of target (m) vtar = -130.0; % speed km/h v = vtar/3.6; % speed m/s fs = 20.0; % sample rate (Hz) ts = 1/fs; % sample time (sec) % determine the actual position of the car with time t=(0:ts:5); % allowed time to get a fix 5s x=xtar+v*t; % uniformly distributed range errors between -0.1 and +0.1m xnoise = 0.2*(rand(size(t))-0.5); xmeas=x+xnoise; % measure rate by subtracting pairs of pulses rate=diff(xmeas)./ts; len=length(rate); t1=t(1:len); %plot(t1,rate); % measure the rate by filtering using an anpha beta filter alpha = 0.5; beta = alpha.^2/(2-alpha); b1=([alpha,beta-alpha]); % position transfer function a1=([1,beta+alpha-2,1-alpha]); b2=beta*([1,-1,0])/ts; a2=a1; xdfil=filter(b2,a2,xmeas); % rate transfer function plot(t1,rate,t,xdfil,t,xdfil,'+') grid title('Target Speed Estimate'); xlabel('Time (sec)') ylabel('Speed (m/s)')
Riegl FG21-P
Figure 13.15: Riegl speed scope
Operating principle pulsed time of flight Beamwidth 2.5mrad Measuring time 0.4 to 1s Speed 0-250km/h Accuracy +/-3km/h to 100km/h +/-3% of reading above 100km/h Range 30-1000m Accuracy +/-10cm typical Target marker, circular reticle matched to beam diameter
Figure 13.16: Effectiveness of narrow beam for measuring speed on congested roads
422 _____________________________________________________________________
13.4.Angle Measurement
13.4.1. Amplitude Thresholding
At its most simple, a received echo amplitude threshold can be used to determine that a target is within the beam. This gives a very rough measure of the target direction.
Range Antenna pattern G1 Target 2 G2
Antenna
Target 1
Note that because the antenna gain drops off sharply, the angular position uncertainty is usually constrained to within the 10dB (one way) beamwidth of the antenna. Both pulsed and CW techniques can be used for angle measurement over a broad range of frequencies from microwave to the infrared band. In general complex modulation schemes are generally used for the following reasons: Discriminate against ambient solar radiation, Eliminate interference from fluorescent lights, Reduce the probability of interference from other sensors.
One of the disadvantages of this technique is that the cross-range resolution degrades with range because sensors operate with a constant angular resolution, as shown in the figure below.
Figure 13.18: The cross range resolution degrades with range as shown. At position (a) the targets are resolved but at position (b) the targets, with the same spacing, are not resolved
The main applications are limited to CW or modulated IR proximity detectors for industrial & robotic applications.
Proximity Detector Example
IR proximity detectors are commonly used for short range robotic applications but not normally used to measure range, just the presence of a target within the beam. They operate in the near IR (at a wavelength just longer than visible light, typically 880nm) and are visible to CCDs so can be observed using a video camera (which can be useful)
423 _____________________________________________________________________ A typical receiver made by Sharp Electronics (GP1U52X) is shown here with a near infrared LED transmitter. This receiver package includes a photodiode, amplifiers, filters and a limiter. The receiver responds to a burst modulated 40kHz signal with an on-off period of 600+600s. The digital output goes low is a target is detected.
Figure 13.19: Infrared proximity detector
Q1 Q2
Q3 Q2 Q1
Time
t =
3dB
k 2( S / N )
deg
(13.12)
where: 3dB Antenna Beamwidth (deg) k Constant dependant on the tracking type S/N Signal to noise ratio The following are the most common null steering techniques: Lobe Switching Conical Scan Monopulse
424 _____________________________________________________________________
425 _____________________________________________________________________
13.7.Conical Scan
A single beam displaced in angle by less than the antenna beamwidth is nutated on its axis. (Nutated means spinning without rotating the polarisation). Beam displacement is often achieved by incorporating a rotating sub-reflector in a Cassegrain antenna as shown in the figure below.
Figure 13.23: Cassegrain antenna with a canted spinning sub-reflector to implement conical scan
The scan rate is generally limited to between 5 and 25 pulses per revolution for long range operation, but at short range where the PRF is higher, many more pulses are
426 _____________________________________________________________________ received. Amplitude modulation of target returns will be a function of the position of the boresight with respect to the target and phase detectors (multipliers) using quadrature phase references for the two orthogonal axes demodulate the received AM signal out of the range tracking gate to generate the angle tracking error as described in the following section. There are a number of different methods that can be used to perform this demodulation function as shown in the following figure.
Range Gated Radar Video
Signal Conditioner
Reflective Encoders
count index
Index Counter
Radar Video
Lowpass Filter Lowpass Filter Azim Error Elev Error
Shaft Encoder
count index
Counter count
Lookup Table
sine cos
M-DAC
M-DAC
(b)
Range Bins
Radar Video
FFT
Target Selection
Shaft Encoder
count index
Counter count
(c)
Figure 13.24: Some conscan angle error demodulation techniques (a) analog, (b) digital in hardware and (c) digital in software
An implementation of the digital solution in software is discussed in detail in the following section.
427 _____________________________________________________________________
0.16 0.14 0.12 0.1 0.08 0.06 0.04 0.02 0 0 0.2 0.4
0.8
0.2
0.6
0.8
1.0
The greater the slope of the error signal, the more accurate will be the angle tracking. This occurs at q/B just greater than 0.4 The gain on boresight is about 2dB down on the peak The range tracking accuracy is determined by S/N, and so requires the maximum on boresight gain q/B = 0 which is not feasible. A compromise is used with q/B = 0.28, which corresponds to an antenna gain about 1dB below the peak.
428 _____________________________________________________________________ To perform this measurement, a conscan antenna mounted on a pan-tilt unit, as illustrated in the following figure, is swept past a point source of radiation (in the far field of the antenna). The received signal level at the output of the antenna is logged as a function of angle (or time if the angular rate is constant).
Conical Scan Receiver
94GHz Source
Pan-Tilt Unit
PTU Control
PC stores the angle error voltages from the receiver and the PTU angle
The received signal are as shown in the figure (a) below: The AC component of the signal level starts out low, when the offset is large It then increases to a peak when the beam squint angle equals the target offset on the one side On axis, the beam offset is symmetrical so there is no modulation, and so the AC component reduces to zero At the cross-over point the phase of the modulation is reversed but the shape of the modulation is the mirror image due to the symmetry of the process In (b), both the modulation signal and the reference signal are shown. Note the 180 phase shift at crossover which is an indication that the sign of the error is reversed. In (c), the synchronously demodulated signal is shown. Note that there is a DC component and an AC component at twice the modulation frequency. Figure (d) shows the filtered demodulated signal showing only the DC which is the conscan angular transfer function for the Cassegrain antenna shown.
429 _____________________________________________________________________
(a)
(b)
(c)
(d)
Figure 13.27: Conscan demodulation process showing the following: (a) modulated signal as antenna sweeps past source (b) modulated signal and reference signal showing phase reversal (c) the demodulation product of the signal and the reference (d) filtered demodulated error signal mapped to angle
13.7.3. Application
Conscan systems are typically used in tracking radars as shown in the figure below. They are simple to implement, using a single beam and a single receiver and transmitter, and beacon tracking can be implemented without the transmitter and without the range gating circuitry.
430 _____________________________________________________________________ As shown in the schematic diagram of the tracker, below, the AM signal out of the range gate is demodulated by the azimuth and elevation reference signals to produce the two angle error signals. These angle errors drive the angle servos which in turn control the position of the antenna, and drive it to minimise the error (a null tracker).
Transmitter Receiver with AGC
Duplexer
cos t sin t
Detector
Range Gating
Scan Motor
431 _____________________________________________________________________
On receive the four horns produce 3D beam patterns as shown in the figure below.
When viewed from the front, the beam patterns are as shown in the cross section shown below.
432 _____________________________________________________________________
Figure 13.32: Combining beams using a microwave hybrid circuit (monopulse comparator) to produce sum and difference channel signals
A B
+90
C D
Phase Shifter -90
Hybrid 1
A-B A+B C+D
Hybrid 2
C-D
Hybrid 3
Hybrid 4
(A+B+C+D) Sum
Figure 13.33: Hybrid configuration to produce sum and difference channel signals at the transmit frequency
It can be seen that if a target is on boresight, then the amplitudes of the signals received in the four channels (A,B,C,D) will be equal, and so the difference signals will be zero. However, as the target moves off boresight, the amplitude of the signals received will differ, and the difference signal will take on the sign and magnitude proportional to the error.
433 _____________________________________________________________________ The difference channel signals are normalised with respect to the sum channel to produce an error signal that is independent of the echo amplitude as shown in the figure below.
Figure 13.34: Normalised sum and difference channel gains as a function of angle off boresight
This ratio can be obtained using an AGC circuit that operates on the two difference channels and is driven by the detected sum-channel output of the tracking gate, or by division in a digital tracker. Phase detectors demodulate the azimuth and elevation error signals using the sum channel IF signal as a reference to produce the two error voltages. These must also be range-gated so that their magnitudes represent the error signals from the correct target. The sum and difference channel voltage signals can be modelled quite accurately using the following
Esum = cos 2 (1.14 ) Edif = 0.707 sin( 2.28 )
(13.13)
where Esum Normalised sum channel output voltage, Edif Difference channel output voltage (normalised to sum channel), - Angle from the beam axis normalised wrt to the half power sum channel beamwidth. New error measurements are produced with every pulse, hence MONOPULSE
434 _____________________________________________________________________
Transmitter AGC Range Tracker
Mixer
Duplexer Sum Channel IF Amp
Amplitude Det.
Video Amp
Mixer
IF Amp
Monopulse Hybrid
Monopulse Antenna Elev Difference Channel
Mixer
IF Amp
Local Osc.
Monopulse
4.0
3.0
Conscan
2.0
435 _____________________________________________________________________
Angle Tracker
Motor Current
Figure 13.37: Angle tracking loop rotates the antenna to minimise tracking error
This physical displacement is achieved by supplying the angle servos with the error signals from the conscan or monopulse modules. These servos power the motors which rotate the antenna in the correct direction. The RMS tracking error due to thermal noise is now
t =
3dB
k 2( S / N ) f r / n
deg,
(13.14)
where: 3dB Antenna Beamwidth (deg), k Constant dependant on the tracking type, S/N Signal to noise ratio, fr Radar pulse repetition frequency (Hz), n Angle servo bandwidth (Hz).
436 _____________________________________________________________________
A simulation model was built using the ultrasonic transducer normalised gain pattern supplied by the manufacturer as shown in the polar plot below.
The difference between the signal voltages received from the left and the right transducer is calculated to produce the receiver transfer function as shown in the following figure. It can be seen that the peak magnitude of the error signal is dependent on the squint angle. It can also be seen that there is a region of about +/-20 over which the relationship between the voltage and the beacon angle is almost linear.
437 _____________________________________________________________________
Figure 13.40: Angle error transfer function with squint angle as a parameter
A category I ILS provides guidance information down to a decision height (DH) of 200ft and with good equipment ILS can even be used for Category II approaches, 100ft on the radar altimeter. The ILS provides both lateral and vertical guidance necessary to fly a precision approach if glide slope information is provided.
438 _____________________________________________________________________
Localiser Transmitter
The transmitter provides lateral guidance. It operates at VHF in the range 108.1MHz to 111.95MHz. The transmitter and antenna are situated on the centreline at the opposite end of the runway from the approach threshold. The antenna radiates two vertical fan shaped beams that overlap at the extended centre line of the runway. To differentiate between the two overlapping beams that are radiating at the same frequency, the right hand side of the beam (as seen by an approaching aircraft) is modulated at 150Hz and the left hand beam at 90Hz. The total width of the beam pair can be varied between 3 and 6. It is adjusted to provide a track signal 700ft wide at the runway threshold increasing to 1nm wide at a range of 10nm.
Localiser Receiver
The localiser receiver activates the needle of one of the cockpit instruments as illustrated above. For an aircraft to the right of the beam, in the 150Hz region only, the needle will be deflected to the left, and visa versa in the 90Hz region. In the overlap region, both signals apply a deflection force to the needle causing a deflection in the direction of the strongest signal so that when the aircraft is precisely aligned, there will be zero net deflection, and the needle will point straight down..
Glide Slope Equipment
The glide slope equipment consists of a transmitter and antenna operating in the UHF at a frequency between 329.30MHz and 335.00MHz. It is situated between 750 and 1250 ft down the runway from the threshold, offset 400 to 600ft to the one side of the centreline and it is monitored to a tolerance of +/-0.5. It consists of two overlapping horizontal fan beams modulated at 90 and 150Hz respectively and the thickness of the overlap is 1.4, or 0.7 above and below the optimum glide slope. The glide slope may be adjusted between 2 and 4.5 above the horizontal plane depending on any obstructions along the approach path. Because of the antenna construction, no false signals can be obtained at angles below the selected glide slope, but are generated at multiples of the glide slope angle. The first is at about 6. It can be identified because the instrument response is the reciprocal of the correct response. The glide slope signal activates the glide slope needle in a manner analogous to that of the localiser. If there is sufficient signal, the needle will show full deflection until the aircraft reaches the point of signal overlap, at this time the needle will show partial deflection in the direction of the strongest signal. When both signals are equal, the needle shows horizontally indicating that the aircraft is precisely on the glide path. With 1.4 of overlap, the glide slope area is approximately 1500ft thick at 10nm reducing to less than 1ft at touchdown. A single instrument provides indication for both vertical and lateral guidance.
439 _____________________________________________________________________
13.12. Triangulation
Though triangulation is not really a range or angle estimation technique, it is probably the most common method in use today for measuring the position of a target in space. The basic technique is thousands of years old and includes the following options Bearing only Range only Hybrid range and bearing
R 62.65 mm 168.11 30.96 R 45.28 mm 94.09 Triangulation using Angles Triangulation using Range (TOF) R 71.59 mm
Applications include Navigation: GPS, Omega, Loran-C Optical surveying Laser triangulation Radio emission detection Pinpointing earthquakes
13.12.1. Loran-C
LoRaN is an acronym for Long Range Navigation. It is a highly accurate (though not as accurate as GPS) highly available, all weather system for navigation in the coastal waters around the US and many other countries. An absolute accuracy of better than 0.25nm is specified within the region of coverage. And as with GPS, it is also used as a precise time reference, A chain of three or more land based transmitting stations each separated by a couple of hundred km is used instead of a constellation of satellites. Within the chain, one station is designated as the master (M), and the other transmitters as secondary stations conventionally designated V,W,X,Y,Z. The master station and the secondaries transmit radio pulses simultaneously at precise intervals and a Loran-C receiver on-board a ship or aircraft measures the slight difference in the time of arrival of the pulses. The difference in the time of arrival for a given master-secondary pair observed at a point in the coverage area is a measure of the difference in distance from the vessel to the two stations
440 _____________________________________________________________________ The locus of points having the same TD from the pair, is a curved line of position (LOP).These curved lines are hyperbolas, or more correctly spheroidal hyperbolas on the curved surface of the earth and the intersection of two or more LOPs from different master-secondary pairs determines the position of the user as shown in the figure below.
Why would anyone want Loran-C? it is inexpensive to operate, at $17 million per year and user equipment is lowcost and of proven capabilities. GPS requires in excess of $500 million per year. Loran's signal format has an integrity check built in, the GPS does not. Loran is easy to service just drive to the transmitting station, GPS replacement requires a new satellite launch schedule. On-air time for Loran-C is about 99.99%, GPS about 99.6% These features have convinced over 25,000 users to sign petitions to keep Loran-C online to its original date, 2015. Foreign nations have purchased new solid-state Loran-C transmitters to form new Loran-C chains in Europe, China, Japan and Russia. This land-based navigation system is viewed as a desirable, stable complement to the future GNSS of the world civilian community. And Loran is totally unclassified and is operated by host country authorities not the U.S. military.
441 _____________________________________________________________________
Why does the military, who designed the GPS system, not rely on it totally? The ease of which GPS can be jammed either on purpose or by unexpected interferences is certainly a major reason. The deliberate jamming is well documented by the U.S. military, the International Association of Lighthouses and the Civil Aviation Authority (UK). A one watt jammer about 5x5x10cm with a 10cm antenna can block out a 60km diameter circle. Picture that near your local airport (such a unit costs about $100.USD). If you want a jammer for GPS and Glonass, (the Russian equivalent to the GPS), such units were offered for sale by the Aviaconversia Company, Russia, which displayed them at a recent Moscow Air Show. Their jamming range was said to be 200km. What was the reaction by the FAA? "Nothing new" because there are "hundreds of these devices on the market".
13.13. References
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] D.Barton, Radar Systems Analysis, Artech 1976. M.Skolnik, Introduction to Radar Systems 2nd ed, McGraw Hill, 1980 G.Morris, Airborne Pulsed Doppler Radar, Artech, 1988. FG21-P Principle of Operation, http://www.riegl.co.at/fg21p/21p_prin.html, 04/04/2000 SpeedTrap, http://www.audicoupe.demon.co.uk/speedtrap_types.html, 22/02/2000 M.Skolnik, Radar Handbook, McGraw Hill, 1970 N.Currie (ed), Radar reflectivity Measurement: Techniques & Applications, Artech House, 1989. M.Skolnik, Introduction to Radar Systems, McGraw Hill, 1980 Navigation Systems- The Instrument Landing System, http://www.allstar.fiu.edu/aero/ils.html, 29/02/2000. Marine Radio Beacons, http://www.navcen.uscg.mil/policy/loran/h-book/book-1.txt, 28/02/2000. Loran C, http://www.landings.com/_landings/reviews-opinions/loran-c.html, 18/04/2001
442 _____________________________________________________________________
Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.
Alternative Proxies: