Dual frequency anti-theft system.

Dual frequency anti-theft system. - Dubbelfrekvent stöldbekämpningssystem.

The present invention relates generally to electronic product control systems, and more particularly to a goods control system that transmits two distinct radio frequency signals, one of which is audio modulated and picked up by a recognizing transceiver by mixing them through a nonlinear impedance

Previous and known monitoring systems of this type, such as those disclosed in, for example, U.S. Patent 4,063,229,

Weis et al., Send! a single radio frequency received by an antenna in an identifying transceiver tag or label, wherein a nonlinear impedance, such as a semiconductor diode, generates a selected harmonic of the transmitted signal that is irradiated for re-detection until the frequency transmitted by the receiving circuit is adjusted.

However, these systems proved to be unsatisfactory in practice in that they lack the sensitivity to reliably detect the presence of an identifying transmitter in the control area and, in addition, gave false alarms in response to many other conditions.

Another important aspect is that the inherent nonlinear characteristics of the transmitter circuits and elements often caused harmonics to be transmitted along with the per transmission frequency, causing the receiver to respond without the presence of a nonlinear impedance element in the detecting transceiver. If the sensitivity of the receiver is to be reduced in order to prevent these directly transmitted harmonics, then in some circumstances the lower energy harmonics generated by the retransmitter element transmitting the transceiver element could be masked. Although this problem can be minimized by proper shielding and RF filtering at both the transmitter and receiver, the filters should have extremely sharp cropping characteristics, so even a small frequency shift from the transmitted signal that is out of the harmonic could easily cause the re-irradiated frequency to be out of the receiver filter pass. Frequency slippage can also be monitored by the Doppler effect caused by moving the identifying transceiver rapidly within the control area, exacerbating the transmitter slip effect.

On the other hand, such high frequency signals could easily propagate outside the intended control range, in which case a remote identifying transmitter invas receiver could cause the alarm to be triggered incorrectly. As a result, protected products often could not be located or handled near the control area. Even then, the frequency energy could propagate through unpredictable reflections or even along waveguide tubes or electrical wires to and from distant points in the area of ​​the shielded structure, causing the alarm system to be triggered incorrectly.

These systems are also susceptible to incorrect triggering by metal objects such as umbrellas, prams and shopping carts, as the interference between different metals^lthe seam or point of contact provides a non-linear impedance diode effect by generating and retransmitting the harmonic of the transmitted signal. On the other hand, the receiver could react to false radio frequency noise coming from elsewhere, such as engine ignition systems and electronic devices.

On the other hand, the system may not respond to the actual presence of the recognizing transceiver element in the control area if the received and retransmitted energy is insufficient. This could occur, for example, when the transceiver antenna is misaligned with respect to the polarization of the transmitted field, or if the antenna is electromagnetically isolated from the transmitter by a body or metal surface. Similarly, when the transmitter is close to the human body, the resonant output circuit may become out of sync, dissipating the harmonic energy ready to be retransmitted to the receiver. Furthermore, although a signal tracking circuit may be provided to adjust the frequency response of the receiver to compensate for frequency shifts in the transmitter, the efficiency of the transceiver suffers badly whenever the tuned output circuit is forced to oscillate at frequencies other than its normal resonant frequency.

Subsequent attempts to solve the problems of these earlier systems have led to many variations. One such is disclosed in U.S. Patent 3,631,484 to Augenblick, in which a single radio frequency is transmitted to a detecting transmitter only to an antenna for repetitive irradiation while comparing a harmonic to signals received by the receiver to detect Dppjkler frequency shifts caused by motion of the detecting transceiver. Although this system eliminates the problems associated with transmitter frequency slip and false alarms from nearby fixed sensors, a product slowly passed through the control area would not cause sufficient Doppler frequency offset to trigger the alarm.

Similarly, attempts were made to study systems in which the nonlinear impedance element in the detector acts as a signal mixer to generate sum and difference frequencies in response to two different frequency transmitted signals, as described in the prior art in U.S. Patent 3,895,368 to Gordon et al. However, such dual-frequency mixer systems had many practical shortcomings, including the problem of how to exclude higher-frequency transmissions from the intended control area. To overcome this problem, said Gordon et al patent describes the use of a dual field system using a high frequency electromagnetic field and a high power, low frequency electrostatic field formed between non-uniform conductors located on opposite sides of the control space. The nonlinear impedance element subordinated to these two fields acts as a mixer to generate sum and difference frequencies which are repeatedly irradiated to the receiver for detection. However, the power required to generate the required electrostatic field in the control area is considerable and such low frequency electrostatic fields may be out of reach of the sensor, e.g. Likewise, nearby pipes and other metal structures can easily turn a small electrostatic field to distant targets, triggering the number of identification cards far out of control, as well as false alarms due to metal joints, metal carts, and the like. through metal structures.

The present invention provides a product monitoring system characterized by the features set forth in the characterizing portion of claim 1. In this system, a non-linear impedance element, such as a semiconductor diode, is connected to a metal antenna on a removable label or card attached to a garment or other merchandise. The antenna is preferably in the form of a folding dipole, wherein the diode is connected between opposite sides of the closed loop portion at one end to form a connected output circuit, the resonant frequency being twice the selected center frequency. The longer antenna portion extending beyond the diode is very close to the voltage length at the selected center frequency, which may be, for example, 915 megahertz. The resonant frequency of the output circuit is determined by the capacitance of the diode and the inductance of the closed loop portion adjacent to the antenna, whereby the resonant frequency is twice the selected center frequency (e.g. 1830 megahertz).

Two different radio frequency signals are both transmitted from dipole transmitting antennas located on opposite sides of the monitoring area. The second signal is generated from a very stable crystal oscillator as an unattenuated wave at a certain frequency (e.g., 905 megahertz) that deviates from the selected center frequency by about 1%. The second transmitted signal is audio modulated, preferably with an audio or audio signal of 1 to 20 kilohertz, to provide a radio frequency offset of plus and minus 5 kilohertz to the carrier, which is also obtained from a very stable crystal oscillator at a frequency (e.g.

925 megahertz), which deviates equally from the selected center frequency on the opposite side so that the average center frequency of the two signals is the same as the selected center frequency. Both transmitter signals are transmitted across the monitoring area from dipole antenna segments oriented at right angles to each other on the same sides and wherein the corresponding dipole segment for transmitting the same frequency from opposite sides is also directed at right angles to each other. This causes cross-polarization of the monitoring area 11a when transmitting the two radio frequencies from opposite sides, thus ensuring that the radiation of both frequencies between the monitoring area 11a transmitters is sufficient in all directions to account for any orientation of the card, while outside is minimized due to their different polarizations. On the other hand, the sound modulation of the second radio frequency causes no stationary waveforms to form, which may cause dead spots or blind spots in the monitoring area and result in cards outside the desired range performing the wrong trigger of the system.

It is also important that the dual frequency operation reduces the effect of the frequency shift of the transmitter and increases the bandwidth of the system in terms of the ability of the detecting transceiver to retransmit the incoming radio frequency signals. In particular, the frequency at which the antenna of the transceiver is tuned can be located anywhere between the two transmitted frequencies, and the power of the transceiver is still not significantly reduced, thus eliminating the need for precise antenna sizing and minimizing human problems, wherein the normal tuning point of the recognizing transceiver shifts downward in frequency due to the dielectric loading effect of the human body in contact with or very close to the card. For example, if the antenna of the sensing transceiver is tuned down and away from the selected center frequency, this will only increase the performance of the sensing transceiver at a lower transmission frequency and will not severely degrade mixer operation as proper mixing occurs at radio frequency power ratios of 10: 1 or even higher. Similarly, the effects of transmitter frequency slip are minimized in the sense that the shift from one transmitter is not multiplied in the same way as retransmitted harmonics from single frequency systems, and all slips from one can be compensated for by performing the opposite shift from the other transmitter.

The signal strength and frequency stability of the retransmitted identification transceiver, as well as the unlikely triggering of a false alarm from transceivers outside the monitoring area, allow for maximum receiver sensitivity and minimum bandwidth. The signals received from the rotating polarized receiving antennas on each side are passed through a very narrow bandpass filter which rejects the transmitter frequencies and then the signals are amplified so that the modulation tone is obtained using quite conventional demodulation methods. Preferably, audio is used (e.g. 2 kilohertz) to the radio frequency carrier taajuusmoduloimiseksi so that the receiver antennas filtered and amplified signal can be derived from a passive double balance mixer which receives the lower side injection signal (e.g. 1808.600 MHz), developed a stable local oscillator suitable intermediate frequency (e.g. 21.4 MHz) to form a mixer output. This intermediate frequency from the mixer is amplified and directed to another precision filter with a narrow passband (e.g., 30 kilohertz), which determines the pre-detection bandwidth. The detection of the modulation tone is then performed using a narrowband (e.g. 30 kilohertz) crystal d mincr imination, the output of which is locked to the ground until its input side is strong enough to generate for the automatic gain control a detector voltage exceeding a preselected reference level which is adjusted to set the sensitivity of the system. When the lock is open, the sound is routed to a phase-locked loop sound interpreter circuit with a voltage-controlled oscillator idle frequency equal to that of the sound and capable of achieving any uniform sound in a narrow frequency range (e.g., plus or minus 10%). When the loop receives an audible signal, the phase difference detector detects the phase locked state and generates a DC output voltage to operate the drive amplifier with a capacitive feedback that lasts to the output signal to trigger the alarm for some minimum time (e.g.

seconds) regardless of the short duration of the expressed sound. In this way, the alarm operates regardless of how short the transceiver is in the monitoring area as soon as the detected signal is strong enough and has the correct modulated frequency content. This eliminates the false alarms that would be caused by return signals from identifying transmitter receivers outside the monitoring range8 and signals from external sources, which external sources may accidentally generate signals corresponding to the retransmitted frequency but lacking the required audio modulation.

In the following, the invention will be explained in more detail with reference to the accompanying drawings, in which:

Figure 1 is a block diagram of the basic elements of a circuit and shows in partial perspective the antenna placement of a product monitoring system according to the invention.

Fig. 2 is a more detailed diagram showing the cross-polarized orientation of the transmit antenna segments and showing a perspective view of the operating antenna and non-linear impedance elements of the identifying transceiver.

Fig. 3 is a more detailed block and circuit diagram showing a preferred embodiment of the narrowband audio modulated RF transmitter of Fig. 1.

Fig. 4 is a detailed block and circuit diagram showing a preferred embodiment of the unattenuated RF wave transmitter of Fig. 1.

Fig. 5 is a block and circuit diagram showing a preferred embodiment of the linear amplifiers shown in Fig. 1.

Fig. 6 is a detailed block and circuit diagram showing a preferred embodiment of the narrowband audio modulated receiver of Fig. 1, wherein the transmitted signal is frequency modulated.

Figure 1 shows a product inspection system according to the invention, in which suitable sets of transmitters and receivers are mounted at respective points on free-standing plates 10 and 12, or possibly on door frames on both sides of the observation area, the observation area typically the person passes between the antennas. Although the antenna arrays are shown in Figure 1 mounted slightly obliquely for clarity, the antennas are normally directly opposite on both sides so that the respective antenna elements are located in parallel vertical planes. As best shown in Figure 2, the transmitter antenna arrays 14 and 16 each consist of pairs of rectangularly spaced metal strips 18, 19, 20 and 21 attached to a vertical washer on either side of a shielded passageway or other area. Each band extends outward from the center pole region and the individual pairs are aligned to form a conventional center feed dipole transmission antenna that is approximately a quarter wavelength with respect to the frequency to be transmitted and can be suitably oriented to extend horizontally and vertically as shown. The individual strips 18 to 21 may be cut from a conventional copper-coated adhesive-based tape commonly used in printed circuit boards and attached to a non-conductive dielectric, low loss suitable support and further to a plate or door frame, or a series of four strips may be etched by simply etching. A conductive metal plate or low-mesh network (not shown) may be placed behind and parallel to the antenna bands 18-21 to reflect and condense the transmitted signal energy and radiation pattern inwardly across the shielded space for more efficient operation and to prevent signal radiation from the opposite side to the area behind plates 10 and 12. In a preferred embodiment of the system, the copper-coated tape strips are attached to the surface of a G-10 fiberglass sheet, which board is glued inside a light anodically treated aluminum body that covers the entire back surface of the sheet 10 or 12 and structurally supports antenna mounts and associated circuit elements.

Attached to both sides are rotationally polarized receiver antennas 22 and 24, such as cross-folding dipole structures such as a reversing gate antenna or a helical antenna. The length of each receiver dipole segment should be a quarter of the wavelength of the frequency transmitted signal, which signal, as explained below, is equal to the sum of the two transmitted frequencies.

Two different radio frequency signals and f? developed for transmission from the corresponding dipole band segments 18,

19, 20 and 21, which form the transmitter antenna arrays 14 and 16. The f 1 signal is a narrowband modulated radio frequency generated by a very stable oscillator 26 connected on one side to the vertical dipole segments 18 of the transmitter antenna array 14 and also via a linear amplifier 28 on the other side of the monitoring area. opposite horizontal band segments 21. The second transmitter signal f £ ^ developed in the same manner with a fixed radio frequency by means of a very stable oscillator 30, a second side of the transmitter antenna array 14 of the horizontal band segments 19, and on the other side of the linear amplifier 32 of the transmitter antenna array 16 on the opposite vertical band segments 20. Both oscillators 26 and 30 preferably use temperature compensated crystal oscillators with a cascade frequency factor and narrow bandpass filters to generate an unattenuated wave f2 and a radio frequency carrier for the audio modulated signal f1, as explained in more detail in connection with Figures 3 and 4.

In general, the distance between the metal band antenna segments 18 21 and the reflective surface of the conductive plate or network behind it, which depends on the thickness of the low loss dielectric background, is selected to form a low voltage constant wave ratio (VSWR) at the antenna input impedance to provide an effective radiation pattern of about 60 ° as the beam emits from the transmitter series 14 and 16 on each side.

Both radio frequencies and thus originate from opposite sets of transmitters 14 and 16 with opposite polarizations, whereby the frequencies intersect and collide on both sides with a recognizing transceiver 34 located in the monitoring area between the two plates 10 and 12. The identifying transmitter receiver 34 is shown schematically in Figure 1 as a rotation-polarized helical antenna loop, with the diode 36 connected across a short closed portion of the loop. However, as shown in more detail in Figure 2, a preferred embodiment of the transceiver 34 consists of an elongate flat metal antenna 38 having a central slot on one side that forms a band dipole structure. The total length of the antenna is ideally a quarter of the wavelength of the center frequency between the two transmitted radio frequencies. The nonlinear impedance element 36 is in the form of a semiconductor diode and connected between opposite sides of the loop near the other end about halfway out of the slot in the stick so that the capacitance of the diode 36 and the f and sum or ts. the resonant frequency is twice the selected center frequency of the transmitter signals.

The exact setting of the diode 36 to provide the antenna resonant 38 with the desired resonant frequency in the output circuit is not critical and is most often determined empirically based on the capacitance of the selected diode and the conduction characteristics of the antenna loop. In practice, the short and straight metal segment on the diode side of the slit acts as a dipole transmission antenna for the voltage in the resonant frequency of the output circuit.

The maximum power and selectivity of the detecting inverter anode are achieved when the frequency difference between the two transmitter signals and the direction Allen is 2% of its center frequency. In the current version of the system, the frequency of the unattenuated aa 1to signal generated by the source 30 is selected to be 905 megahertz, while the frequency of the fan modulated carrier of the second signal transmitted from the source 26 is 925 megahertz. Thus, their center frequency is 915 megahertz, and the frequency of the resonant output circuit is 1830 megahertz. These frequencies are selected to be within the range of available spectrum transmission bands available for these purposes in the United States. On the other hand, in order to comply with international transmission standards, it is clear that the system can, for example, be designed with a resonant output circuit frequency of about 4,900 megahertz and transmission frequencies of about 2,420 and 2,480 megahertz.

In practice, when the antenna loop 38 of the annotator of the detecting transmitter receives both transmitted signals ^ 1 ^ 2 ' ^ofä^mThe signals are mixed due to the nonlinear impedance effect of the half-lead diode 36, starting at the resonant frequency of the output circuit oscillation equal to and the sum of the frequencies. Better mixing and the overall efficiency of the sensing transceiver are achieved by using a fast switching level diode, a low RF threshold and a low forward bias. In particular, lower cost germanium diodes are preferred due to their relatively low threshold of about 0.3 volts compared to more expensive silicone diodes with thresholds of 0.6 volts.

A frequency difference of about 2% between transmitted signals provides important benefits by maximizing the efficiency of the detecting transceiver and enabling the system to avoid false alarms because the detecting transceiver return signal differs from the signal that could be generated by various metal objects such as umbrellas arrangements. In particular, the bandwidth of the detecting transceiver 34 with respect to incoming radio frequencies widens without compromising its efficiency, as the receiver antenna 38 can be attempted to settle anywhere between the two transmitter frequencies, which also minimizes human-induced tuning problems. This is due to the fact that tuning or de-tuning antenna 38 closer to one transmitter frequency than the other improves the signal strength at this frequency without compromising mixer conversion capability, as proper radio frequency mixing can occur at power ratios of up to 10: 1 or greater.

Furthermore, due to the cross-polarization of the two frequencies transmitted from both antennas 14 and 16, their propagation at one transmitter point to points far outside the monitoring area is rarely similar for both signals. A rare reflection pattern that can result in a second transmitted signal condensing at a recognizing transceiver somewhere far away almost never causes the second and oppositely ipolarized transmission to be reflected in the same form and reach the same area with sufficient power. Therefore, if only one signal is received, the nonlinear impedance of the diode 36 can only generate a frequency doubling and not the necessary scrambling effect, so the next return signal is at a frequency very different from the return signal1 of the desired transceiver. For example, a transceiver that identifies with current system parameters would generate twice the frequencies of 1810 or 1850 megahertz14, both of which differ by a full 20 megahertz from the normal return frequency of 1830 megahertz.

These shifted frequencies would be considerably attenuated from the tuned output circuit and can be easily separated by a conventional filtering technique from the correct and mixed frequency response at 1830 megahertz.

In this connection, it can be seen that the signals received by the receiver antennas 22 and 24 on both sides are fed via a conventional scrambling interface 40 to the narrowband audio modulated receiver 42. The mixing of the two transmitted signals from the return signal of the identifying transceiver allows the response of the receiver 42 to be limited to very narrowband operation, eliminating false alarms due to external noise and transmission signals from other sources. In fact, the required bandwidth of the receiver depends for the most part only on the frequency severity of the transmitters 26 and 30, thus providing a very narrow detection window corresponding to the possible frequency slip of the transmitter. When using very free transmitter oscillators, as will be explained below, the bandwidth of received signals (i.e., pre-detected bandwidth) available for modulation tone detection can be extremely narrow, and the receiver bandwidth (after detection) can be further narrowed by accurately detecting the modulation tone. In addition, the reliability and sensitivity of the system is further improved by causing the transmitter 42 to provide an output signal to start the transmitter 44 only when the detected modulation audio signal strength exceeds the selected minimum amplitude level within a predetermined time to ensure that the identifying transceiver is actually present in the detection area.

The preferred embodiment in operation in Figure 3 generates a very free, narrowband frequency modulated signal for the transmitter signal to maximize system sensitivity and selectivity. A stable sound generator 46 of conventional construction, which may be of the simple RC type, generates a frequency sound in the hearing range of 1 to 20 kilohertz. This sound is 2 kilohertz in the present system and is fed as a modulation signal to a voltage controlled crystal oscillator 48 to frequency modulate its output. In a preferred embodiment, the crystal oscillator 48 is of conventional construction and its precise temperature compensation is capable of maintaining a frequency stability of 0.7 cycles per million at a temperature of 5 to 45 ° C at a frequency of about 51.4 megahertz. The amplitude of the modulation signal coming from the sound generator 46 and directed to the voltage control circuit is adjusted to provide a maximum possible frequency deviation of only plus or minus 0.25 to 0.30 kilohertz, followed by only very narrowband modulation of the oscillator carrier. The modulated output of the oscillator 48 is then directed to a conventional frequency factor 50 which triples the oxidizer frequency, which is then applied to a narrowband, bipolar bandpass filter 52. This filtered coefficient signal is then applied to a second conventional frequency factor 54 which again triples the available frequency. The filtered output of the bandpass filter 56 is then passed to another frequency factor 58, this time only doubling the input frequency to produce the desired modulated output signal (fj), the frequency of this signal being 925 megahertz and the narrowband modulation offset being plus or minus 5 kilohertz. To RF-1 amplifier 60 and power amplifier 62. This amplifier transmitter signal fj is routed through a narrowband three-pole bandpass filter 64 to a power splitter 66 which reproduces the transmitter signal to the vertical antenna bands 18 of the transmitter array 14 of the plate 10 and further to the linear amplifier 28 in the second plate 12 via a lightweight cable connector.

Figure 4 shows the second transmitter frequency f? similarly, using a conventional temperature-compensated crystal oscillator 68 capable of maintaining a frequency of 0.5 ppm at 5 to 45 ° C with an output frequency of about 50.3 megahertz. This output frequency is tripled by a frequency factor 70 and filtered by a bipolar bandpass filter 72. The narrowband output of filter 72 is then passed to a second frequency factor 74, which again triples the frequency to direct it through the second bipolar bandpass filter 76, and then the filtered output frequency is doubled in the last frequency factor 78 to generate the desired f2 signal at 905 MHz.

The f2 ~ signal is applied to the input of the adjustable RF boost amplifier 80 and the second amplifier stage 82 to achieve the desired transmission power. The amplified output is then filtered through a narrowband, three-pole bandpass filter 84 to remove amplified distortions or excesses and direct the output to a power divider 86 for further transmission to the antenna bands 19 and the transmitter 14 on the plate 10. Due to the high efficiency and sensitivity achieved, the transmitted power of these signals is on the order of magnitude lower than that required in previous systems, thus eliminating potential health concerns from potential tissue transmissions due to microwave transmissions.

As shown in Figure 5, the respective outputs f1 and f1 from the power divider 66 or 86 can be connected to the linear amplifiers 28 and 32 on the opposite antenna plate 12 by simple wire conductors or a light cable, eliminating the need to install expensive and difficult to adapt and heavy RF components. ape li li practices that were needed in previous systems to avoid power losses. Each of the linear amplifiers 28 and 32 consists of a simply adjustable radio frequency amplifier stage 88, the output of which is passed through a narrowband three-pole bandpass filter 90, thereby eliminating any signal distortion or noise generated or developed in the gain line. The gain of amplifier step 88 is adjusted to keep the strength of the transmitter signal substantially at the same level as when fed to the transmitter antenna segments on the opposite side.

As shown in Figure 6, in a preferred embodiment using narrowband frequency modulation, the signals picked up by the receiver antennas 22 and 24 are routed through a mixer 40 to a very narrowband, forward bandpass filter 92. In the particular system shown, the valid return signal of the recognizing transceiver 34 is frequency modulated by a fixed single audio tone, preferably at a frequency of 2 kilohertz, resulting in a maximum deviation of only 5 kilohertz on either side of the 1830 megahertz carrier frequency. The bandpass filter is designed to discard lower frequency transmitter signals with a minimum value of 60 to prevent internal interference due to nonlinearities in the circuit. The filtered output of the bandpass filter 92 are applied by double-balance mixer 94 for mixing the underside injektiotaajuuteen ^ f, with a value of 1808.600 MHz, for example, is stable to paikällisoskillaa11or

To generate an intermediate output frequency (IF) of 21.4 megahertz at its output when a valid return signal from the identifying transceiver is present. The lower side is formed by the injection in the same way a very stable, temperature compensated crystal oscillator 96, which is a value of about 50.24 megahertz. The first oscillator 98 as frequency multiplied by four and two sequentially fed to triple the frequency multiplier 100 through 102 and the narrow-band-pass Iinapaiseen 104 injektiosignaaIin lower side feeding into the mixer 94.

The intermediate output frequency of the balanced mixer 94 is applied to a low noise amplifier 106 to stabilize the overall value of the receiver noise to read 12db for input to a four-part monolithic crystal bandpass filter 108, preferably model 1619-1622 manufactured by Piezo Technology,

Inc. is a registered trademark of COMLINE, where the amplitude response versus frequency is 30 kilohertz to -3 db points.

The crystal bandpass filter 108 effectively determines the width of the pre-detection band and, together with the 12 db noise reading and the modulation value 5, gives a total receiver sensitivity of -113 dbm at a ratio of 20 db S + N / N at the output of the crystal discriminator 110, as will be explained in more detail below. The output signal from the crystal bandpass filter 108 passes through successive RF amplifier stages 112 and 114, both of which are arranged in a strip or chip capable of automatic gain control to provide the desired input level to the crystal discriminator 110. proportional to the amplitude of the output. These corresponding AGC values ​​from the separate phases 112 and 114 are summed and act as an automatic total gain detector 116 whose output is a direct current proportional to the combined output amplitude of each phase, which in turn indicates the signal strength of the original detecting transceiver signal from the bandpass filter 108. This combined output of the AGC detector is fed to a low-pass filter 118 having a predetermined time constant to form a gradually increasing charge at a rate proportional to the intensity of the detected detecting transceiver return signal. The output charge of the low pass filter 118 is provided to the reference circuit 120 for comparison to a predetermined threshold value set at the sensitivity setting of the potentiometer 122.

In a preferred embodiment of the system, the crystal discriminator 110 is a monolithic crystal filter such as

From Piezo Technology, Inc. as model 2378F, which is connected to an RCA integrated circuit model CA 3 08 9 E as shown in the accompanying data sheet to form an extremely narrowband stable discriminator with a bandwidth of only about 30 kilohertz. With a valid identification transceiver return signal, the output of the discriminator 110 generates a modulating audio sound that is 2 kilohertz in the system used. However, the output of discriminator 110 is held from ground potential by lock circuit 124 until the trigger value from reference circuit 120 indicates that the charge generated by low pass filter 118 exceeds the selected sensitivity setting from potentiometer 122. This allows the system to be set to a sensitivity level that ignores transient or weak return signals from distant identifying transceivers or other sources.

As soon as the interlock circuit 124 is open, the 2 kilohertz audio is routed through a low-pass filter 126 for decoding by conventional phase-locked loop technology 11a using a phase difference detector 128 and a phase detector 130 capable of receiving all smooth tones in a range of 10? Ό In the conventional manner, the output of the phase detector 130 is passed to a loop filter 134 to generate a signal to adjust the frequency and phase of the voltage controls and the oscillator 132 to provide a phase lock.

The phase difference detector 128 then transfers its output to a conventional drive amplifier 136, which is maintained by a feedback capacitor 138 to trigger a suitable alarm 44 to trigger an auditory or visual response for a selected period of time, regardless of how short the initial response is. In this way, a strong response from a detecting transceiver present in the monitoring area between the antenna plates 10 and 12 triggers a full alarm no matter how fast the protected product passes through the area, but the system is able to ignore even continuous low power reaction signals outside the immediate protection area.

Although the system has been described with reference to a preferred embodiment using specifically described circuit elements and methods and having operating parameters associated with an existing preferred embodiment using audio frequency modulation, the invention can be practiced using modifications and modifications of circuit elements and methods without departing from the claims. For example, the system may be implemented using amplitude modulation of the second transmitted radio frequency instead of frequency modulation, or by using modulation tones outside the audio range without still losing the basic operating benefits of this unique operating system.