GB2432999A

GB2432999A - RF tag detection

Info

Publication number: GB2432999A
Application number: GB0525622A
Authority: GB
Inventors: Nicholas Patrick Roland Hill
Original assignee: Individual
Current assignee: Individual
Priority date: 2005-12-16
Filing date: 2005-12-16
Publication date: 2007-06-06
Anticipated expiration: 2025-12-16
Also published as: GB0525622D0; GB2432999B

Abstract

Proximity of an RF tag to a reader is detected in a manner that does not rely on close coupling, or on an exact frequency match between an interrogation RF field and a resonant frequency of the tag. This is achieved by utilising a read RF field which chirps in frequency as well as decaying in amplitude with time. The generator comprises a resonant circuit including an antenna coil L1, in series with capacitors C1, C2, C3 and a non-linear element FET1. A stimulus signal comprising square wave is applied to the antenna. Since FET1 switches on and off in response to the square wave, the resonant frequency of the antenna circuit is modified according to the mark/space ratio of the stimulus and the amplitude of the resulting ringing signal. The result is a chirp pulse increasing in frequency as the amplitude decreases. The frequency range of the chirp includes the resonant frequency of a tag to be detected. On approach of the tag, its resonant circuit absorbs energy from the interrogating field as its frequency equals the tag resonant frequency. This causes changes in the phase of the RF signal which are detected. The tag may be an implanted cat ID tag. Detection of the tag switches on the full ID reading arrangement of the reader, which may, on correct ID, operate a cat flap.

Description

I

RFTaa DetectiQfl

Field of the Invention

Embodiments of the invention relate to the field of proximity detection of an RFID (radio frequency identification) transponder, more particularly to method and systems for low power proximity detection of an RF]D Transponder.

Background of the Invention

A large amount of effort has gone into the development of radio frequency identification (R.F1D) systems. Initially these started off being simple analogue detectors of passive LC resonatorS fixed to the object of interest. A common application was tagging of items within a store to avoid theft. An early method is disclosed in US2774060 where an antenna operating at a given frequency is loaded by the presence of a transponder with a similar resonant frequency. This extra loading is registered indicating the presence of a transponder.

Further developments moved into the time domain where short RF bursts are sent out by an antenna to activate the resonant circuit in a transponder. The ring signal generated in the transponder was subsequently picked up by the reader and offered greater sensitivity than systems where the stimulus and measurement were carried out together.

An exemplary reference is US3818472.

As the field has matured many more developments have taken place, including the move to higher information content transponders. Here the transponder still contains at least one resonant circuit, which derives power from a RF field, powering a digital circuit in the transponder. This digital circuit subsequently modulates the transponder response transmitting data such as an identification number.

As the technology has moved from analogue to digital, the power required to read the transponder has increased. Sufficient power must be transferred to the transponder in order to run the digital circuit contained therein for a sufficient length of time to transmit the required information. This increased power requirement may be problematic where low power is required, such as a portable or remote reader run from a battery power source. The issue of power drain was recognised in US6476708 and GB2278631, where a low power reader mode was proposed. This low power mode makes use of the fact that the digital RFID transponder has a LC resonant circuit, whose presence may be detected without full power being applied to the transponder to run the digital circuit. Once the presence of a transponder within the read range has been determined the reader may switch to a full power mode that activates the digital circuit in the transponder and reads/transmits the required information. This approach can lead to very large power savings where the reading operations are required over a low proportion of the time, as the reader stays mostly in its low power mode.

The analogue detection method proposed in US6476708 and GB2278631 is the generation of ring signals in the reader antenna. In GB2278631 a 2ms pulse actuates the antenna natural resonance every lOOms. The waveform in the antenna is a damped resonance. If the transponder resonance frequency is close to the antenna resonant frequency then when it is brought close to the reader it will further damp the antenna resonance. Such an increase in damping may be registered through a measurement of the decay amplitude in the antenna; changes in the decay amplitude indicate a changed antenna damping and therefore the presence of a transponder.

The key advantage of the ring signals described above is low power. Since the waveform is generated with only a single pulse, there is no requirement for an oscillator to generate the signal in the antenna for the duration of the decaying waveform. This saves a significant amount of power. Furthermore the waveform in the antenna is well suited to exciting the transponder resonance when it is centred on the transponder resonance frequency. Furthermore, provided the antenna Q is not much less than the transponder Q the ring signal lasts for an appropriate amount of time to stimulate the transponder.

There are however disadvantages of the prior art method, which are listed below.

Frequency selectivity The method requires that the antenna resonance frequency is similar to that of the transponder. This cannot always be guaranteed, for example manufacturing tolerances in either the transponder or the reader antenna can mismatch the systems. Alternatively, environmental factors such as temperature or metallic objects nearby have the capability to detune either antenna. Still one further example is if the transponder resonance frequency is not known, such as a low frequency animal identification transponder where both 125kHz and 134kHz are commonly used frequencies.

Sensitivity The method is based on an energy amplitude measurement, which will work well provided the coupling between the reader and the transponder is strong enough to appreciably modify the decay of the antenna ring signal. However, where the system is required to work with low levels of coupling between the antenna and the transponder, increased sensitivity would be beneficial.

An example of such a low coupling application is a battery operated cat flap, where the operation of the door lock is controlled by the sub-dermal RFID transponder implanted into the pet. The levels of coupling may be significantly less than 1%, requiring a more sensitive low power transponder detection method in order to sense the approach of the animal before moving into a full-power mode to detect the transponder identification number.

In summary, the prior art methods for a low power detection mode of a digital RFID transponder through the increased damping of an antenna ring signal are well suited where there is good frequency matching between the antenna and the transponder and relatively high levels of coupling. There is however a need for a similarly low power detection method that may be used where the transponder frequency is not well known and for reduced coupling levels.

Summary of the Invention

We will describe a low power detection method for the presence of a RFID transponder.

The system may move into a full power identification mode when the presence of a transponder is detected.

Rather than a conventional LC reader antenna resonant circuit, these embodiments use a circuit that contains a non-linear element, such that the natural resonance frequency depends on amplitude. In response to a pulse excitation the reader antenna circuit generates a chirp signal that simultaneously decays with time and sweeps across a frequency range.

The antenna circuit is preferably significantly higher Q than the transponder such the decay of the chirp is relatively slow compared to the transponder response. In this manner there is still an appreciable duration of the antenna waveform when the transponder frequency is close to the chirp instantaneous frequency. When this occurs the transponder absorbs energy from the antenna, which is subsequently registered in the reader. This chirp method therefore retains the low power feature associated with the prior art method of ring signals, but no longer requires matching between the resonant frequency of the antenna and the transponder.

It is further an object of these embodiments that the sensitivity of the detection is increased relative to the prior art ring signal method. These embodiments employ a decaying chirp waveform that links the amplitude of the chirp to the frequency of the chirp. This waveform allows a more sensitive measurement of energy loss through the phase of the decaying waveform. Small changes in the energy absorbed from the antenna cause corresponding changes to the frequency of the decay, in addition to the amplitude. Comparing two levels of damping, once the two chirps are operating at different frequencies their phase difference increases with time. Consequently, after a delay of many cycles the small difference in damping between the two waveforms may be picked up by a straightforward sampling of the chirp waveform. The phase difference between the two waveforms, once it amounts to a significant fraction of a cycle, translates to a large fractional change in the sampled voltage. This is in contrast to an amplitude measurement, which would display a much smaller effect. Registering changes in damping through the phase of the chirp decay offers improved sensitivity

over the prior art.

It is further an object of these embodiments that the frequency sweep of the chirp decay may be used to determine the transponder resonant frequency. As the frequency sweeps over, the transponder resonance, the effect of the transponder on the antenna is maxirnised. For the case where the chirp starts at low frequency and sweeps to a higher frequency then the effect of increasing the transponder resonant frequency is to delay the corresponding change to the chirp. Rather than sampling the chirp at one point only, sampling at a number of points along the length of the decay therefore provides information relating to the transponder resonance frequency. This measurement may either be carried out every time the chirp is generated, or alternatively a change in the system may be registered from a single point measurement and subsequently the transponder frequency determined from a multi-point measurement of the chirp. Once the transponder resonant frequency has been determined the antenna may be tuned accordingly for the subsequent full power identification mode. This multi-point measurement may also beneficially discriminate between a transponder with a clear resonance and interfering objects that absorb energy over a wide frequency range, for example metallic objects.

In summary, the embodiments provide an improved low power detection mode for the presence of an RFID transponder. The features include: 1) Low power

2) Increased sensitivity over prior art methods.

3) No requirement for antenna-transponder accurate frequency matching.

4) Information relating to the transponder resonant frequency may be determined.

Further Aspects of the Embodiments Below is a list of points detailing further aspects of the embodiments.

1) A resonant circuit with a non-linear element in it such that on pulsing it produces a chirp decay that links the amplitude to the decay. A measurement of phase is used to determine small differences due to the presence of an object.

2) Point 1 where the circuit uses at least one FET and variable duty cycle of two capacitive branches in order to generate the chirp.

3) Point 1 where the circuit uses a varactor as the nonlinear element that generates the chirp.

4) Point 1 where the circuit uses electrically variable inductor as the nonlinear element that generates the chirp.

5) Point 1 where the object to detect is a RFID transponder 6) Point 1 where the object to detect is a passive resonant circuit 7) Point I where the object to detect is a metallic object 8) Point 1 where the object is coupled inductively 9) Point I where the object is coupled capacitively 1O)Averaging sequential samples may be used to track environmental changes or battery droop 11) Averaging sequential samples may be used to reduce noise interference.

12) A cat flap with a built in RFID reader, where the low power mode is used to detect the presence of a cat prior to full power reading of the sub-dermal RFID chip.

13)A pet feeder with a built in RF]D reader, where the low power mode is used to detect the presence of a cat prior to full power reading of the sub-dermal RFID chip.

Brief Description of the Drawings

Figure 1 is a schematic of the first embodiment of the reader.

Figures 2A and 2B are decay waveforms of the first embodiment where no transponder is present in the proximity of the reader. Figure 2C is the instantaneous frequency of the decay.

Figures 3A and 3D are decay waveforms of the first embodiment where a 125kHz transponder is present in the proximity of the reader with coupling constant 0.3%.

Figure 3C is the instantaneous frequency of the decay.

Figure 4A is a graph showing both envelope functions from the decays in figures 2A and 3A. Figure 4B is the difference between these two envelope functions plotted on the same vertical scale.

Figure 5A is the difference waveform between the two decay waveforms in figures 2A and 3A. Figure 5B is the difference waveform where the transponder resonance frequency has been increased from 125kHz to 134kHz.

Figures 6A, 6B, 6C show the difference waveforms for the 125kHz transponder with coupling constants 0.1%, 0.3%, 1.0%, respectively.

Figure 7 is a schematic of a second embodiment of the reader.

Figure 8 is a schematic of a third embodiment of the reader,

Detailed Description of the Embodiments

The following description of a low power proximity detector for a digital RFID transponder, where the transponder resonant frequency may also be determined, is merely exemplary in nature and is in no way intended to limit the invention or its applications or uses. Those skilled in the art would readily recognise that it may also be applied to the detection of any object that modifies the reader response, including a resonant analogue RFID transponder and metallic or magnetic non-resonant objects.

Coupling to the reader antenna of the object to be detected may be capacitive or inductive. The frequency range around 125kHz used in the embodiments may be varied over a very wide range, encompassing the whole range of antennae from sub-audio frequencies to microwaves.

Figure 1 shows the first embodiment of the invention. A high Q antenna comprises 32 turns of 660-strand 46AWG Litz wire, with overall diameter approximately 20cm.

Around the target operating frequency of 125kHz the antenna has inductance 300.tH (Li) and effective series resistance 0.7 (Ri), giving a Q of 340. The antenna is placed in series with the capacitor network Cl, C2, C3 and an n-type FET. The capacitive network presents two different net capacitances in series with the antenna, depending on whether the FET is on or off. The duty cycle over which the FET is on depends on the oscillation amplitude through the associated variation of the FET source potential. The natural resonance frequency of the circuit is therefore determined by the amplitude. A high level of oscillation gives a near 50% duty cycle, whereas an amplitude less than Vth gives a 0% duty cycle. These two extremes of duty cycle correspond to two extremes of frequency, given by the following equations: fSO% 1 IL.(ci + (c2-' + C3' )-)+ JL.(ci + C3)) 24L.(C1+(C2 +C31)j The above equations give a frequency range over which the circuit may resonate in response to a stimulus.

When a negative voltage is placed on the FET gate then a larger amplitude is required to keep the same duty cycle of FET conduction. Therefore a negative gate voltage may be used to increase the amplitude of oscillation to the required level. In this manner the circuit shown in figure 1 may be used in an RFID reader to generate an interrogation field over a wide frequency band. One benefit of this system is that excitation of an antenna may be achieved at a pre-determined frequency, without fine-tuning of an inductor and/or a capacitor. Furthermore, the system is tolerant to some degree of detuning e.g. with metallic or magnetic material placed in the vicinity of the antenna, provided the operating frequency band still encompasses the stimulus frequency.

This embodiment uses an additional feature of the circuit shown in figure 1, where the free decay of the natural circuit resonance generates a frequency sweep i.e. a chirp signal. A 41.ts square voltage pulse of amplitude 5V is applied to Vstimulus and the resulting traces of the chirp decay shown in Figure 2. Figure 2A shows the decay of the FET drain voltage and figure 2B shows the antenna current. The drain voltage is asymmetric when the FET duty cycle is non-zero, which is a natural consequence of the two different capacitances coupled into the resonance. The antenna current on the other hand is more symmetric.

Figure 2C shows the instantaneous frequency of the chirp as a function of time. The chirp starts off at 117.9kHz, rising to a maximum of 144kHz. The high frequency limit compares well to a predicted f0% of 145kHz, however the start frequency is above the other limit of f$Q% = 106kHz. This is because 50% duty cycle is only approached where the amplitude is much greater than the threshold voltage Vth. If a lower starting frequency is required for the same capacitive network then one or more of the following methods may be employed: 1) Increase the initial amplitude of the oscillation, for example by a double pulse stimulus.

2) Change the FET for one with a lower Vth.

3) Put a positive gate voltage on the FET of amplitude less than Vth. This keeps the operation of the circuit the same i.e. the FET conducts only on the negative cycle of the FET source voltage. However, the amplitude required for FET turn on with such a positive gate voltage is reduced.

Figure 3 shows similar chirp graphs as figure 2, however a resonant transponder has been introduced. The transponder comprises a lml-1 inductor in parallel with a l.6nF capacitor, giving a resonant frequency of 125kHz. The effective series resistance of the inductor is 50 =, which gives the transponder a Q of approximately 15. The coupling between the reader antenna and the transponder is set low at 0.3%. There is no obvious difference between the two sets of traces with and without the transponder. This observation is clearer in figure 4A, where the decay envelopes of the FET drain voltage are overlaid and are virtually indistinguishable. Figure 4B shows the difference in the decay envelopes on the same vertical scale, showing their very close match. In fact the maximum difference is approximately 2%, a very low level that would be hard to pickup with a measurement of energy decay.

Rather than looking at the decay envelope, figure 5A shows the difference in the actual waveforms between the transponder present and transponder absent. This measurement has the great advantage that it is sensitive to differences in phase and the result has much improved sensitivity. As the chirp decays and sweeps through its frequency range, the transponder absorbs a small proportion of energy, changing the frequency of the decay. A phase difference relative to the reference waveform builds up over time and this gives the large difference voltage in figure 5A. The voltage difference is maximal at approximately I.9ms (0.9rns from the start of the chirp), reducing after this time as the chirp decays further. Note that once the amplitude of the chirp decays below Vth and the duty cycle of the FET is zero, the phase difference between the traces stays constant and the difference waveform decays with amplitude of the chirp; in this example the zero duty cycle threshold is approximately 2.1 5ms.

Figure 5B shows a similar waveform difference measurement where the transponder frequency has been increased from 125kHz to 134kHz (resonance capacitance decreased from I.6nF to 1.4nF). A change to the difference voltage is clear, in particular the onset of a sharp rise in the difference voltage is delayed with respect to the 125kHz transponder. This is as expected since the chirp frequency is rising with time and therefore matches the new transponder frequency later in the decay. The amplitude of the effect is also lower, which is a result of less chirp duration after the transponder has started to have an appreciable effect on the decay. The chirp moves into the zero duty cycle region relatively soon after the transponder is excited and the phase difference relative to the reference stops increasing. The comparison of figures 5A and SB shows how the shape of the difference waveform may be used to discriminate between transponders with resonant frequencies 125kHz and 134kHz. Further analysis, particularly of the start point of the difference waveform may allow the transponder frequency to be determined with finer resolution.

Figure 6 shows difference waveforms corresponding to a 125kHz transponder for a range of coupling constants to the reader antenna. The coupling constant increases from 0.1% in figure 6A to 0.3% in figure 6B (same data as figure 5A) and 1% in figure 6C.

As the coupling constant increases the effect of the transponder on the chirp decay also increases resulting in an increase in the difference waveform. For the case of 1% coupling the effect of the transponder is sufficiently large to cause multiple cycles of phase difference. The effect of these multiple cycles is the oscillating envelope seen in figure 6C. The phase difference may be calculated and unwrapped to determine the onset time and maximum phase change associated with the transponder; this process may help the determination of transponder resonant frequency in relatively high coupling situations.

Figure 7 shows a second embodiment, which includes two FETs, one p-type and one n-type. This circuit operates in a similar manner to the single FET version, shown in figure 1, but with a more symmetric output waveform. As such it may be advantageous where reduced distortion is required, for example to fit within regulatory permitted frequency limits. For low amplitudes both FETs are non-conducting, whereas for high amplitudes both are conducting. The design equations for the frequency limits at 0% duty cycle and 100% duty cycle (here duty cycle is the fraction of the cycle that either FET is conducting) are as follows: = r(/i. (Cl + C3) + JL (Cl + C5))

I

0% 2r(\IL. (ci + (cr' + C3))+ i. (ci + (c4 + C5)-)) The values shown in figure 7 give a maximum frequency range of f100%lO4kHz and f0%=I 47kHz.

Figure 8 shows a third embodiment based on a single FET whose duty cycle controls the resonance frequency, but with an additional capacitor (C4) and FET (FET2). When the FET is turned off the antenna current flows through C4, which consequently generates a voltage. In this example C4 is much larger than the other resonance capacitors and therefore does not strongly affect the frequency range of the chirp.

Furthermore, because of the large capacitance, the voltage generated has an amplitude less than O.7V and does not therefore conduct through the body diode of FET2. The voltage at the junction of Cl and C4 may be subsequently taken to an analogue to digital converter (ADC) for sampling. The ADC will generally have a sample and hold function included such that it digitises the value of the waveform on a timescale significantly less than the time period of that waveform.

The main advantage of this circuit is that the ADC input may be bypassed by turning FET2 on, in which case the antenna current is returned to ground rather than flowing through C4. In this manner the ADC input, which is set-up for the low power proximity detection of the transponder, is protected when the reader switches into a full power identification mode. The large voltages generated in the full power mode, such as at the drain of FETI, are not loaded by the proximity detection circuit elements and therefore do not cause any damage.

An alternative manner in which the circuit in figure 8 may be used is to keep FET2 on for most of the chirp decay, only turning it off when the decay waveform is sampled. If, by the time it is sampled, the chirp is decayed significantly from its start level then a smaller value for the capacitance C4 may be used, still keeping the amplitude below O. 7V (and therefore avoiding the body diode path through FET2). In this manner the voltage sampled by the ADC may be increased (up to O.7V), improving the signal to noise of the measurement.

One further alternative is keep FET2 off for the duration of the chirp decay, as originally proposed, and introduce some gain before the ADC samples the voltage. This may also include a voltage offset such that the dynamic range of the signals is appropriate for the input range of the ADC. An increase in signal to noise by this method is at the expense of power drain from the additional gain circuitry. In order to minimise this additional power drain, the amplification stage may be powered at a reduced duty cycle corresponding to the repeat rate of the chirp.

The following techniques may be applied to engineer a robust system from the fundamental concept of these embodiments.

1) Monitoring the reference signals over time allows the system to system to track longer term changes not associated with the proximity of a transponder. Examples of such changes that are beneficially ignored include temperature changes, changes to the absorption from surrounding metal, and a gradual droop in the battery voltage that will cause variations in the chirp decay.

2) Averaging signals over many samples may give improvements in the signal to noise and therefore the robustness of the system. Such averaging may be carried out at the same sample time relative to the chirp start, and averaged over several separate chirps.

3) Multiple points within one chirp may be sampled and the difference to a reference determined at each sample. The decision as to whether a system change has taken place may be made more accurately by combining these separate measurements.

Such a sampling of multiple points may be combined with the requirement to determine the transponder resonant frequency.

In summary, an improved low power method for the proximity detection of a RFJD transponder has been described using a chirp decay. The embodiments described use a FET as the non-linear element that links the amplitude of the decay to the instantaneous frequency, although it is appreciated that there are many alternative methods. The use of the chirp decay has the following advantages: 1) The phase of the decay waveform has increased sensitivity to absorption from a RFID transponder in proximity with the reader.

2) The frequency range of the chirp allows operation with a range of transponder resonant frequencies and does not require matching of the reader and transponder antennae.

3) The onset time and shape of the phase difference may be used to determine information relating to the transponder resonant frequency. This may be used to determine the transponder frequency or alternatively to discriminate between two or more known operating frequencies that the transponder may have.

No doubt many other effective alternatives will occur to the skilled person. It will be understood that the invention is not limited to the described embodiments and encompasses modifications apparent to those skilled in the art lying within the spirit and scope of the claims appended hereto.

Claims

CLAIMS: 1. An rf tag proximity detection system, the system comprising:

an rf signal generator to generate a chirped decaying rf pulse; an rf detection system to detect said decaying pulse; and a system to identify a change in phase of said detected pulse to determine proximity of said rf tag to said detection system.

2. A tag detection system as claimed in claim I wherein said phase change identification system comprises means to compare said detected pulse with a reference signal.

3. A tag detection system as claimed in claim 1 or 2 wherein said phase change identification system comprises means to determine a cumulative phase change of said detected pulse.

4. A tag detection system as claimed in claim 1, 2 or 3, wherein said rf signal generator comprises a resonator incorporating a non-linear element.

5. A tag detection system as claimed in any preceding claim wherein said rf signal generator comprises a controllable electric resonator comprising an inductor coupled to a first capacitor to form a resonant circuit, the resonator further comprising a controllable element, a second capacitor controllable coupled across said first capacitor by said controllable element, and a control device to control said controllable element such that a total effective capacitance of said first and second capacitor varies over a duty cycle of an oscillatory signal on said resonator.

6. A tag detection system as claimed in any preceding claim wherein said rf signal generator and said rf detection system comprise a common rf resonator.

7. A method of detecting proximity of an object to an rf signal generator, the method comprising: generating a chirped decaying rf signal pulse using said rf signal generator; and detecting dephasing of said chirped rf signal pulse to detect said object.

8. Apparatus for detecting proximity of an object to an rf signal generator, the method comprising: means for generating a chirped decaying rf signal pulse using said rf signal generator; and means for detecting dephasing of said chirped rf signal pulse to detect said object.

9. A method of operating an rf transponder tag reader, the method comprising operating said reader in a low rf output power mode to detect proximity of an rf transponder tag, in particular using the system, method or apparatus of any preceding claim; and operating said reader in a high rf output power mode to read data from said tag responsive to said detection of proximity of said tag.

10. Apparatus for controlling the operation of an rf transponder tag reader, the apparatus comprising means for operating said reader in a low rf output power mode to detect proximity of an rf transponder tag, in particular using the system, method or apparatus of any preceding claim; and means for operating said reader in a high rf output power mode to read data from said tag responsive to said detection of proximity of said tag.