CA1297562C - Radar intrusion detection system - Google Patents

Abstract

ABSTRACT OF THE DISCLOSURE

A bistatic Doppler radar intrusion detection system which utilizes a high duty-cycle phase-coded pulse compression signal to provide a range-gated detection zone is described. The combined transmit and receive antenna beam patterns define the shape of the detection zone in azimuth and elevation. The distance from the antennas to the detection zone and depth of the detection zone are determined by the pulse compression code sequence selections. The system consists of a transmitter having a code sequence generator for generating one of a selected number of codes and a code sequence. An antenna transmits the modulated signal through the detection zone. A programmable digital delay circuit is used to provide a delayed replica of the code sequence and thereby specify the detection zone range. A receiver is provided having an antenna for receiving signals scattered from the detection zone. The received signal is split into two paths, in-phase and quadrature-phase and reference signals from the delayed replica used to synchronously detect the received signal.
Processing circuitry computes the Doppler frequency spectra of the received signal and performs automatic detection of targets traversing the detection zone.

Description

~7~i6;2 The presen-t invention relates to intrusion detection sys-tems and, specifically -to a system using bistatic radar and Doppler radar processing techniques in conjunction with phase-coded pulse compre~sion methods to achieve a high resolution detection ~one "window".
The term "pulse compression" is used in the sense given in the I.E.E.E. Standard Dictionary, namely: "The coding and processing of a signal pulse of long time dura-tion to one of short time duration and high range resolution, while maintaining the benefits of high pulse energy."
Intrusion detection systems can be of a line sensor type or volumetric sensor type~ Line sensors provide perimeter security. In ~uch systems, targets passing between the antennas cause partial or complete blockage of the transmit signal, resulting in the declaration of an alarm. Such systems cannot, however, provide true Doppler detection since there is no net radial movement. Leaky cables have also been used to provide perimeter coverage but are difficult to deploy rapidly. The operation of a leaky cable sensor above ground is troubled by moving foliage, sunlight, temperature drift and moisture around and on the cables. The result can be a high false alarm rate and an unsatisfactory detection performance.
The present invention relates to a volumetric sensor.
There is a requirement for such a volumetric security sensor to detect intruder3 and vehicles. For example, weapons stockpiles, mobile C3I resources and garrisons must be alerted to an intrusion well before the intruder can wreak havoc on the resource. The kh/

1297~;~i2 requirement typically ie for a detection zone which i3 at leacJt a few ten~ of meters away from the reeource to allow unre~trictsd movement about the re~ource itcelf. It may be impractical to poeition -the intru~ion detection ~y~tem in or near the detection zone. Consequently, there i~ a requirement for a oecurity ~en~or that can provide a detection zone far enough away from the re~ource and the seneor it~elf to allow ~ecurity forcee to react.
A well-defined detection zone centered 20-200m away from the re~ource ic cufficient for a number of ~ecurity applications. The cecurity sensor should be able to detect and track intrusion~ to allow eecurity forcee to quickly locate and intercept the potential intruder.
Any intrueion detection radar cystem ehould have a well defined detection zone. Movement outcide the detection zone ohould not reault in an alarm. Personnel movement in the vicinity of the antennas, ~uch ao within an encampment, must be tolerated by the eystem. Since the detection zone iB uaually ~ome di~tance from the antennas, radar range-gating techniquee are necessary.
The eensor must provide detection of a variety of ground level intrucionc, euch ae vehicles and both a crawling and on-foot intruder. Airborne intrueion (e.g., hang glider, parachute) mu~t also be detected by the radar syctem. The detection procese for euch targete can be optimized by u~ing coherent or ~ynchronous detection techniquee, thuc enabling Doppler signal procesaing.
Thi~ allowc both magnitude and pha3e information of the returned eignal to be uced in procescing the data.

kh/

Foliage penetration capabilities are required for a number of security applications. The sensor must aloo be ab]e to diecrimina-te between the eignal returned from blowing foliage and a legitimate targe-t, Signal processing techniques, for example Fourier analysis and Kalmus filtering, are commonly uc3ed to help "unmask" an intruder's scattered signal from the clutter return signal. The ~3ensor must be able to maintain a high detection capability/low false alarm rate over all weather conditions.
There are a variety of techniques that can be used to determine the range of the received aignal. The traditional means to achieve this has been pulse-type radar; that is, a pulse or burst of RF energy i8 transmitted, with the received signal sampled once or a number of times. Each successive sample corresponds to a more di~tant range cell. The depth of the detection zone is approximatsly equal to the length of the pulse multiplied by the speed of light. Eor example, a 200nsec pulse can yield a range resolution of 60m or better. As a result, fine resolution in range requires a short pulse and therefore a higher peak power. In an effort to reduce the peak power requirement while still maintaining the same range resolution, radar designers utilize pulse compression signals. The more common pulse compression signals used are the frequency chirp and phase-coded waveforms. A chirp waveform is accomplished by a gradual (or otep-wise) increase or decrease in the rate of change of phase of the transmitted signal. Phaee-coded signalc3 are obtained by changing the phase, at instante determined by a code ~equence, in a smooth or abrupt faehion of an otherwise continuous wave kh/

~97S6~
~ignal. There are a variety of code sequences that may be sui-table fvr radar pulse compression. Pseudo-noise (PN) sequences, Barker codes and complementary series are examples of code sequences -that have favorable characteristics in terms of radar detection. The means to compress the returned pulse compression signal so as to achieve the same integrated response as a single pulse having the same total duration as the pulse compression signal (without sacrificing range resolution) may also take a variety of forms. Some of the more common techniques and technologies include digital correlation, active correlation, SAW
devices, CCD correlators, and acousto-optic devices.
A line sensor using pseudo-random codes is disclosed in U.S. Patent No. 4,605,922. This patent teaches a microwave motion sensor system using spaced transmitting and receiving antennas.
The transmitted signal is modulated by a pseudo-random code to cause a spreading of the transmitted signal over a wide frequency band. This renders any jamming techniques ineffective. The receiver has a similar pseudo-random code generator to that in the tran.smitter and locks on to the transmitted code. The random code sequence is not used for range gating as is done in the invention of this application.
U.S. Patent No. 4,458,240, i~sued July 3, 1984 (issued on a divisional application of U.S. Patent No. 4,187,501) shows a system using transmission line ,sensors in which the starting phase of the transmitted signal i8 switched by 0 or 180 from pulse to pulse under the control of a pseudo-random code generator. As in U.S. Patent No. 4,605,912, this spreads the spectral energy and kh/

greatly reduces the effect of any interfering signa]. The ranrlom code sequence is no-t used for range gating as it is in the invention of the present application.
A Continuous wave (CW) or multi-CW radar, though capable of providing a simple receiver design because of a high transmit signal duty-cycle at or near unity~ cannot satisfy many of the above requirements The lower bound on the sampling rate for a CW
radar, given by the Nyquist sampling theorem, can be as low as a few tens of llertz for the targe-ts of interest. Movement around the antennas can overwhelm the return signal from targets just a few tens of meters away since there is no range-gating capability with such a signal. Similarly, large targets which are past the desired detection area cannot be suppressed; consequently, railways and roadways near the desired detection range can result in an unacceptably high nuisance alarm rate. Because of the inability to provide range-gating with CW signals, the non-fluctuating portion of the received signal, also referred to as the profile or stationary clutter, is usually quite large, often placing limits on the receiver sensitivity. U.S. Patent No.
4,595,924 describes a Very High Frequency (VHF) CW ~oppler radar.
A more conventional pulse-type radar, while capable of providing one or more "range cells", has numerous drawbacks.
These include a much faster, more complex and therefor more costly data collection process, susceptibility to intentional or unintentional interference, ease of targeting by hostile forces, and an increased peak transmit power because of the transmit signal's low duty-cycle (typically well under 10 %). As well, sampling in excess of 10 MHz is required for a detection zone kh/

~756~

resolution of 30 m or less. U.S. patent No. 3,6()3,996 describes such a pulsed Dopp]er radar.
SUMMARY OF THE INVENTION
The present invention pertains to a personnel intrusion detection radar that achieves a range-gated detection zone with a high duty-cycle phase-coded pu!se compression signal. The range to the detection zone and its depth or width are programmable. By using a high duty-cycle pulse compression signal the more favorable attributes of CW and pulse-type radar systems are combined.
The result is a radar system with the range resolution of a pulse-type radar and the sampling/preprocessing sirnplicity of a CW radar. Though designed primarily as an intrusion detection system, the sensor may also be used for object detection (e.g., railcars).
This result is achieved in accordance with the invention by using a pulse compressed signal containing a pseudo-random code to establish both the range and range window for targets of interest.
Specifically the invention relates to an intrusion detection system comprising: means transmitting an r.f. signal formed from a continuous wave modified by phase changes at selected instants; means providing a code se~uence to control the selected instants; means receiving a portion of the transmitted signal which may have been modified by the presence of a target; and means mixing the received signal with a delayed replica of the transmitted signal to establish a detection zone external to the space between the antennas, the delay establishing the range of the detection zone; whereby the system provides an enhanced response relating to objects in the detection zone.

B rn/

~2~75fi~
An embodiment of the lnvention will now be desc-ribed in conjunction with the accompanying drawings in which:
Figure 1 is a schematic diagram ~howing the apparatus and detection zone of the system of the invention;
Figure ~ shows the layout of the apparatus of the invention;
Figure 3 is a diagram illustrating the configuration of the detection zone;
Figure 4 is a schematic diagram of the transmitter portion of the sy~tem;
Figure 5 is a schematic diagram of the receiver portion of the system; and Figure 6 i~ a schematic diagram of the signal proce~sing portion of the ~ystem.
BRIEF DESCRIPTION OF THE PREFERRED EMBODIMENTS
The intrusion detection system providing a three-dimensional range-gated detection cell is shown in Figure 1. The combined antenna beam patterns of the transmit and receive antennas shape the detection zone in azimuth and eleva-tion. The detection zone may exist over a complete hemisphere, as illustrated in Figure 1, or a portion thereof. The depth and range of the detection zone are programmable. An intruder approaching the antenna~ is illuminated with a phase-coded VHF
signal. A portion of the scattered signal i9 received a-t the receive antenna. ~nen the intruder enters the detection zone his return signal causes a deviation from the nominal or quiescent received signal spectrum. The radial movement of an intruder in kh/

~29756Z
the detection zone changes the phase and magnitude of the returned .~ignal, due to the Doppler effect. A ~ubstan-tial change in the signal spectrum result~ ln the declaration of an alarm. The signal received at the receive antenna from objects not in the detection zone is negligible following synchronous detection because of the autocorrelation properties of the code 3equence, thereby providing a range-gated detection capability.
Consequently, the system achieves a range resolution equal to that of a pulse-type radar, but because the sy~tem transmits and receives a high duty-cycle signal, requires only the sampling and preprocessing speed of a CW-type radar.
The system consists of a power source, two antennas, lead-in cable for the antennas, a termina~ and an electronics unit, as shown in Figure 2. The unit is powered using ac or dc power. A terminal allows the operator to set the detection zone range, the detection zone depth, the receiver gain, and other system selections, and to receive the results of the processed receive signal such as alarm information. The transmit signal generated in the electronics unit 18 transferred to the transmit antenna by coaxial lead-in cable. Similarly, the signal received at the receive antenna propagates down another lead-in cable to the receiver section of the electronics unit.
Referring again to Figure l, the electronics unit generates the transmit signal and sends this signal via lead-in cable 2 to one of the antennas 4. The other antenna 5 is used to receive the reflected ~ignal. The received signal is transferred to the receiver portion of the electronics unit by way of lead-in kh/

56;~

cable 3, The properties of the tran.smit waveform are such that a range-gated detection zone ie establi~hed about the antennas. By u~ing omni-directional antenna.~, omni-directional coverage is obtained. Directional antennas will reduce the azimuth and/or elevation coverage The detection ~one is ellipsoidally-shaped with the antennas repre~enting the foci of the ellipsoid. Omni-directional antennas are as3umed. Referring now to Figure 3, the operator makes two selections:
(i) antenna separation : the antennas are separated by a distance of 2xo, with a straight line between the antennas defining the x-axis.
(ii) nominal range : a nominal range of b is selected, 3pecifying the range along the y and z-axes.
The detection zone range along -the x-axis, a, is determined by the two selections:
a = (xo2+b2)l/2 The detection zone exist3 for any (x,y,z) such that (x-xo)2 y2 + z2 _______ + __--------a2 b2 Typical detection zone value~ for the inventive system are: an antenna separation of 30m, a nominal range of 80m, a detection zone depth of 20m. The approximate volume for these detection zone parameter3 is 1.6x106 cubic meters.
The operational and performance benefit~ of an intruder detection radar in the low VHF band and, in particular, near 60 ~z, are well under3tood. The radar cro3s-section (RCS) for kh/

~2~7S6~

humans is a maximum in the low Vl{F band. This occurs because the effective length of a human while standing on ground is approximately a quarter-wavelength. Small animals aad birds, by contrast, have a minimal RCS at this frequency range because they are much smaller than the 5m wavelength of a 60 M}l~ signal; these smaller taIgets are, in fact, referred to as Rayleigh scatters.
Similar results are also applicable for propagation through the forest; leaves, pine needles and branches are Rayleigh scatterers as well. As a result, the propagation 1088 through the forest and its clu-tter return are substantially less than that experienced by sensors operating at higher frequencies.
Figure 4 illustrates the transmitter portion of an intrusion detection system of this invention. A con-tinuous wave source 15 supplies a VHF signal to a modulator 17 through a power splitter 16. The VHF signal is preferably at 60 MHz. Modulator 17 is also supplied with a code sequence from code generator 13 which has been filtered by lowpass filter 14 to remove the higher frequency components of the code. This results in the output from modulator 17, which is supplied to transmitting antenna 4 through a wideband amplifier 18, being a continuous wave wi-th smooth phase changes at eaGh change of the code from generator 13. The rate at which the code is generated is controlled by a clock 12 which, in turn, is controlled by a control unit 11. By changing the clock rate the code sequence rate i8 changed and, as will be shown below, the depth of the range window altered.
A digital delay unit 19 is coupled to the output of the code sequence generator 13 and supplies a delayed version of the kh/

~29~56~

code eequence -to a lowpae~ filter 20 and subcaeqllently to modulator 21 substantially identical to modulator 17. This provide~ a reference signal for use in the receiver for demodulating any received signals from a target. So that both in-phase and quadrature-phase received signal~ may be detected, quadrature power splitter 22 supplie~ appropriate signala on lines 23 and 24 to the receiver unit. The amount of delay in -the digital delay circuit 19 determines the nominal range at which the detection zone will be located and the clock rate determines the depth of the detection ~one.
The pre~ence of a target in the detection ~one causea a reflection of some of the transmitted signal. The reflected ~ignal, together with some signal received directly from the transmitting antenna, i3 picked up by receiving antenna 5 (see Figure 5). The received signal i~ bandpa~s filtered and amplified in amplifier 34 and divided u~ing power splitter 35 into two channels. These signal~ are applied to modulators 36 and 37, where they are modulated by the in-phase 23 and quadrature-pha~e 24 reference signals, respectively. Referring to the in-phase channel for example, the received signal is mixed with the delayed replica of the transmitted signal in modulator 36. Only those signal component~ which are correla-ted with the delayed code sequence are detected and, hence, an enhanced signal representing any reflection at the particular range defined by this delay is produced. These in-pha~e and quadrature-phase detected ~ignal~
are then processed in the normal manner through the remaining signal proces~ing channels ~hown in Figure~ 5 and 6.

kh/

A typical ayctem operate~ at a -tran~mitted power of 20 mW with a code length of between 100,000 and 1 million bit~ and at a clock rate of 10 M}lz. The longer the code length i~ the ~maller i~ the probability of any ambiguity in the returned ~ignalc. There i~ a limit to code length, however, becau~e the spacing between -the ~pectral component~ mu3t be greater than the Doppler bandwidth to avoid other ambiguitiec in the received cignal. Typically, for a maximum Doppler bandwidth of lOHz the spectral components are spaced at leact 20 Hz apart. Ac previou~ly mentioned, a variety of code ~equences cuch a~
pseudo-noi~e, Barlcer code~ and complementary ~erie~ can be u~ed.
Specifically, the in-pha~e detected signal goe~ through a low pa~s filter 38, an amplifier 39, and a ~ample-and-hold circuit 42. Both oignals are then multiplexed in multiplexer 44, converted to digital ~ignals in converter 45, and procecsed by proce~eor 46 for eub~equent Doppler frequency re~pon~e. Signale exceeding a given thre~hold produce an alarm warning eignal.
Although a particular embodiment hao been described, it will be clear that variations are poeeible while remaining within the ecope of the inventive concept. A3 the size of the detection zone increaseo, it becomee more important to locats the target in range, azimuth and elevation. Thie additional direction finding requirement can take a variety of form~. A high gain receive antenna can be ~canned over the desired detection zone, with the azimuth/elevation reeolution being determined by the antenna beamwidth. Alternatively, target azimuth and/or elevation can be kh/

12~756~

de-termined by using two or more receive antennas to compare the relative pha~e of the return signale.
As a related embodiment, either the tran~mit or receive element could be replaced by a leaky coaxial cable. ~or example, the cable could encircle the antenna at a fixed radius. The effect of the pulse compression ~ignal is to achieve a plurality of detection ~one~ along the cable. Consequently, the perimeter along the cable i9 effectively divided into a number of sectors, thereby providing an indication of the intrueion location.
Although the apparatus of the preferred embodiment described above change~ phase by 180~; the equipment will function with different amounts of pha~e change, 45 or 90~ for example.
The effect of the different angle of phase change is to modify the spectrum of the tran~mitted signal slightly by reducing it~ higher frequency content. Further, it is not necessary that there be only switching between two phase angles. Instead, the transmitted waveform could be switched by three or more different phase shifters, provided only that a delayed replica of the transmitted signal is used in the receiver.

kh/

Claims (5)

1. An intrusion detection system comprising:
means transmitting an r.f. signal formed from a continuous wave modified by phase changes at selected instants;
means providing a code sequence to control the selected instants;
means receiving a portion of the transmitted signal which may have been modified by the presence of a target; and means mixing the received signal with a delayed replica of the transmitted signal to establish a detection zone external to the space between the antennas, the delay establishing the range of the detection zone;
whereby the system provides an enhanced response relating to objects in the detection zone.

2. A system as in claim 1 wherein a plurality of detection zones are provided at different ranges from the antennas.

3. An intrusion detection system comprising:
a transmitting and a receiving antenna, their combined beam patterns establishing a substantially continuous detection zone spaced in azimuth and elevation from the antennas;
a c.w. source;
a code sequence generator;
phase modulating means responsive to the code sequence generator and the output from the c.w. source to provide a signal to the transmitting antenna consisting of the c.w. signal with phase changes at instants determined by the code sequence;

demodulator means reversing the phase changes on that portion of the received signal that corresponds to a selected range to provide a monitoring signal;
whereby the monitoring signal exhibits an enhanced response to the reflections from targets at the selected range.

4. An intrusion detection system comprising:
a leaky coaxial cable extending along a perimeter to be protected and an associated antenna, transmission between them defining a detection zone along the cable;
a c.w. source;
a code sequence generator;
phase modulating means responsive to the code sequence generator and the output from the c.w. source to provide a transmission signal to one of the antenna and cable consisting of the c.w. signal with phase changes at instants determined by the code sequence;
means connected to the other of the antenna and cable to provide a received signal;
demodulator means reversing the phase changes on the portion of the received signal that corresponds to a selected range to provide a monitoring signal;
whereby the monitoring signal exhibits an enhanced response to the reflections from targets in the detection zone.

5. An intrusion detection system as in claim 4 wherein the demodulator means is supplied with a series of delayed replicas of the transmission signal whereby the enhanced response to each delayed replica represents a sector of the cable within the detection zone.