SECURITY THREATS

CROSS-REFERENCE TO RELATED APPLICATIONS

The present application claims priority from Australian Provisional Patent Application No 2009903907, the content of which is incorporated herein by reference.

TECHNICAL FIELD

Embodiments relate generally to the detection of security threats within an object. Particular embodiments relate to systems and methods that can detect these threats by sensing the presence of one or more of the components of a device or substance within an object that can be indicative of a threat. The components of a security threat that may be detected include non-linear junctions, batteries, timers, controllers, wires, initiators, hand held phones, explosives, radioactive materials, biological materials, chemicals, closed circuit loops, connectors and metal pieces, for example.

BACKGROUND

The proceeding discussion of the background art is intended to facilitate an understanding of the context in which embodiments may be employed. Nothing in this application should be interpreted as an acknowledgement or admission that any of the material referred to forms part of the prior art or was part of the common general knowledge as at the priority date of the application.

Due to changes in technology, terrorists have a number of options by which they can attack a target. Improvised devices, such as improvised explosives devices (IED), improvised biological devices (IBD), improvised nuclear devices (IND), improvised chemical devices (ICD), improvised radioactive devices (IRD), and improvised incendiary devices (HD), can be used to destroy targets or render large areas unusable. These objects can be cheap to build and can be made from a wide variety of components. Detecting these devices in some instances can be problematic and difficult. To detect threats in objects, other automatic detection techniques have mainly focussed on detecting explosives, such as pulsed fast neutrons, gamma rays, nuclear magnetic resonance (NMR), 3D tomography X-ray, and trace detectors. All of these techniques have been shown to have some shortcomings when detecting explosives reliably and at a low false-alarm rate.

Additionally, the smuggling of hand held phones, video cameras and digital cameras into prisons has been become an issue for many prisons around the world. These electronic items are typically smuggled past security guards in packages or handbags brought into prisons. To date, there hasn't been an effective solution to prevent these devices from entering prisons.

Similarly, there is a requirement to be able to screen a range of IED threats transported in bulk air cargo to secure international supply lines.

It is desired to address or ameliorate one or more shortcomings or disadvantages associated with prior threat detection techniques, or to at least provide a useful alternative thereto. SUMMARY

Some embodiments relate to an inspection system for detecting a threat within an object, said inspection system having a detection volume therein, the system comprising:

a conveyor belt for moving said object in a first direction through said detection volume;

at least two inspection subsystems positioned serially along said first direction, wherein at least one of said two inspection subsystems transmits electromagnetic pulses to said object and wherein both of said at least two inspection subsystems detect electromagnetic fields generated by said object; and

a computer to receive data indicative of the detected electromagnetic fields, generate signals based upon said data and compare said signals to predetermined thresholds. Some embodiments relate to an inspection system for detecting a threat within an object, said inspection system having a detection volume therein, the system comprising:

a conveyor belt for moving said object in a first direction through said detection volume;

a first inspection subsystem positioned along said first direction, wherein the first inspection subsystem is adapted to detect metals;

a second inspection subsystem positioned along said first direction and adjacent to said first inspection subsystem, wherein the second inspection subsystem is adapted to detect non-linear junctions and operates within a transmission frequency range of 1OkHz to 16MHz; and

a third inspection subsystem positioned along said first direction and adjacent to said first and second inspection subsystems, wherein the third inspection subsystem is adapted to detect electronic circuits;

wherein each of said first, second, and third subsystems generate data indicative of electromagnetic fields and wherein a computer receives data indicative of said electromagnetic fields, generates signals based upon said data and compares said signals to a known threshold. Some embodiments relate to a detection method for detecting a security threat within an object disposed inside a detection volume, comprising: receiving signals from sensors disposed around the detection volume indicative of a quality of the object, processing the signals and determining whether the signals are indicative of a possible security threat being present within the object.

Some embodiments relate to a system and/or apparatus for detecting a security threat within an object disposed inside a detection volume, comprising: means for receiving signals from sensors disposed around the detection volume indicative of a quality of the object, for processing the signals and for determining whether the signals are indicative of a possible security threat being present within the object.

Some embodiments relate to a detection system comprising: a threat detector defining a detection volume and having at least one detection technique for detecting security threats in an object passing through the detection volume. The number of detection techniques may be: at least two; at least three; at least four; at least five; or at least six.

The threat detector may detect at least one of the components of a threat. If there are two or more detection techniques in the threat detector, at least two detection techniques may be used to indicate the presence of a threat before an alarm is generated. If there are two or more detection techniques in the threat detector, at least two detection techniques may be used to indicate the presence of a threat and the threats identified by the two or more techniques may need to be spatially correlated before an alarm is generated.

If there are two or more detection techniques in the threat detector, the detector may comprise at least one detection technique that can detect countermeasures and at least one other that can detect some other component of the threat.

The detector may be designed to comprise two or more detection techniques which target different or the same components of a threat.

The detector may be adapted to comprise electromagnetic detection techniques where the wavelength of the electromagnetic radiation is greater than the longest dimension of the object under inspection. The frequency of the electromagnetic radiation may be within the range 1 Hz-IOOMHz.

The threat may include one or more of: an IED, IND, IBD, IRD, ICD, HD, hand-held phone or a powered circuit,

The threat detector may comprise an electronic circuit detector. The threat detector may comprise a Non Linear Junction Detector (NLJD) for detecting electronic circuits. Alternatively or additionally, the threat detector may comprise an Active Electronics Detector (AED) for detecting electronic circuits. Alternatively or additionally, the threat detector may comprise a Quadrupole Resonance (QR) detector for detecting electronic circuits. The threat detector may comprise a Metal Detector (MD) to detect metallic objects. The threat detector may comprise a Non Linear Junction Detector (NLJD) for detecting ferrous materials. The threat detector may comprise one or more explosives detectors. At least one of the explosives detectors may use Quadrupole Resonance, X-rays, Neutrons, Gamma Rays, Nuclear Magnetic Resonance, trace detection or Terahertz imaging technologies to detect explosives.

The threat detector may comprise a Quadrupole Resonance (QR) detector to detect magnetoacoustic ringing or piezoelectric ringing. The threat detector may comprise a Metal Detector (MD) to detect metallic objects and locate their position in three dimensions.

The threat detector may comprise a Metal Detector (MD) to detect metallic objects and may have an operating frequency in the IHz to 10OkHz range, or in the IHz to 1OkHz range. The threat detector may comprise a NLJD to detect electronic circuits and may have an operating frequency in the 10 kHz to 100MHz range, or in the 100 kHz to 16MHz range.

The threat detector may comprise an AED to detect electronic circuits and may have an operating frequency in the 1 kHz to 100MHz range, or in the 100 kHz to 16MHz range.

The threat detector may comprise at least two different detectors and if more than one detector produces a positive detection, then an alarm is raised, otherwise the object is passed as clear.

Some embodiments relate to a threat detector for detecting threats within an object in a detection volume comprising: a first probe to irradiate the detection volume with electromagnetic radiation; a second probe to receive electromagnetic radiation from the detection volume; transmitting means and receiving means to drive said first and second probes; processing means to operate the transmitting means in conjunction with the receiving means and process signals received by the receiving means to identify a threat within the detection volume; and alarm means to alert an operator to the presence of an identified threat in the detection volume. The threat detector may employ one or more detection techniques. The number of detection techniques may be at least two, at least three, at least four, at least five or at least six. The threat detector can detect at least one of the components of a threat. If there are two or more detection techniques in the threat detector, at least two detection techniques may need to indicate the presence of a threat before an alarm is generated. If there are two or more detection techniques in the threat detector, at least two detection techniques may need to indicate the presence of a threat and the threats identified by the two or more techniques may need to be spatially correlated before an alarm is generated. If there are two or more detection techniques in the threat detector, the detector may comprise at least one detection technique that can detect countermeasures and at least one other can detect some other component of the threat.

The detector may comprise two or more detection techniques which target different or the same components of a threat. The detector may comprise electromagnetic detection techniques where the wavelength of the electromagnetic radiation is greater than the longest dimension of the object under inspection. The frequency of the electromagnetic radiation may lie within the range 1 Hz-IOOMHz. The threat may comprise an IED, IND, IBD, IRD, ICD, HD, hand held phone or a powered circuit, for example.

The threat detector may comprise an electronic circuit detector. The threat detector may comprise a Non Linear Junction Detector (NLJD) for detecting electronic circuits. The threat detector may comprise an Active Electronics Detector (AED) for detecting electronic circuits.

The threat detector may comprise a Quadrupole Resonance (QR) detector for detecting electronic circuits. The threat detector may comprise a Metal Detector (MD) to detect metallic objects. The threat detector may comprise a Non Linear Junction Detector (NLJD) for detecting ferrous materials.

The threat detector may comprise one or more explosives detectors. One of the explosives detectors may use Quadrupole Resonance, X-rays, Neutrons, Gamma Rays, Nuclear Magnetic Resonance, trace detection or Terahertz imaging technologies to detect explosives. The threat detector may comprise a Quadrupole Resonance (QR) detector to detect magnetoacoustic excitation or piezoelectric excitation. The threat detector may comprise a Metal Detector (MD) to detect metallic objects and locate their position in one, two or three dimensions. The threat detector may comprise a Metal Detector (MD) to detect metallic objects and may have an operating frequency in the 1 to 10OkHz range. The threat detector may have an operating frequency in the IHz to 1OkHz range. The threat detector may comprise a NLJD to detect electronic circuits and may have an operating frequency in the 1 OkHz to 100MHz range. The threat detector may comprise a NLJD to detect electronic circuits and may have an operating frequency in the 100kHz to 16MHz range. The threat detector may comprise an AED to detect electronic circuits and may have an operating frequency in the IkHz to 100MHz range. The threat detector may have an operating frequency in the 10OkHz to 16MHz range.

The threat detector may comprise at least two different detectors and be arranged so that if more than one detector produces a positive detection, then an alarm is raised, but otherwise the object is passed as clear (ie. no threat detected).

Some embodiments relate to a probe for detecting a threat within an object passing through a detection volume in a prescribed direction, the probe comprising: at least one coil to produce a respective at least one magnetic field, each of which is substantially oriented in one direction within the detection volume.

One of the coils may produce a substantially horizontal magnetic field within the detection volume, parallel to the prescribed direction to detect NLJ's. One of the coils may produce a substantially vertical magnetic field within the detection volume to detect NLJ's. One of the coils may produce a substantially vertical magnetic field within the detection volume to metal objects. One of the coils may produce a substantially horizontal magnetic field within the detection volume, parallel to the prescribed direction to detect metal objects. One of the coils may produce a substantially horizontal magnetic field within the detection volume, perpendicular to the prescribed direction to detect metal objects. One of the coils may receive substantially vertical magnetic fields within the detection volume to detect active electronics. One of the coils may receive substantially horizontal magnetic fields within the detection volume, parallel to the prescribed direction to detect active electronics. One of the coils may receive substantially horizontal magnetic fields within the detection volume, perpendicular to the prescribed direction to detect active electronics.

One of the coils may produce substantially horizontal magnetic fields within the detection volume, parallel to the prescribed direction to detect QR signals;

Some embodiments relate to a probe for detecting a threat within an object passing through a detection volume in a prescribed direction, the probe comprising: a plurality of coils which produce magnetic fields which are substantially oriented in two or three different directions within the detection volume. The different directions may be perpendicular to each other.

One of the coils may produce a substantially horizontal magnetic field within the detection volume, parallel to the prescribed direction to detect NLJ's. One of the coils may produce a substantially vertical magnetic field within the detection volume to detect NLJ's.

One of the coils may produce a substantially vertical magnetic field within the detection volume to detect metal objects. One of the coils may produce a substantially horizontal magnetic field within the detection volume, parallel to the prescribed direction to detect metal objects. One of the coils may produce a substantially horizontal magnetic field within the detection volume, perpendicular to the prescribed direction to detect metal objects. One of the coils may receive substantially vertical magnetic fields within the detection volume to detect active electronics. One of the coils may receive substantially horizontal magnetic fields within the detection volume, parallel to the prescribed direction to detect active electronics. One of the coils may receive substantially horizontal magnetic fields within the detection volume, perpendicular to the prescribed direction to detect active electronics. One of the coils may produce substantially horizontal magnetic fields within the detection volume, parallel to the prescribed direction to detect QR signals.

Some embodiments relate to a probe for NLJD apparatus for detecting a threat within an object passing through a detection volume in a prescribed direction, the probe comprising: a first coil to produce a substantially vertical magnetic field within the detection volume; and a second coil to produce a substantially horizontal magnetic field within the detection volume, parallel to the prescribed direction. The first coil may comprise two single turn saddle coils connected in parallel. The second coil may comprise a single turn coil.

Some embodiments relate to a probe for a MD apparatus for detecting a threat within an object passing through a detection volume in a prescribed direction, the probe comprising: a field sensor to receive at least one magnetic field from the threat, the field being either induced through an applied magnetic field or inherent to the threat within the detection volume.

A second field sensor may be provided to receive at least one magnetic field from the threat which is either induced through an applied magnetic field or inherent to the threat within the detection volume. The second field sensor may be arranged to receive and sense a field substantially perpendicular to the first field sensor.

A third field sensor may be provided to receive at least one magnetic field from the threat which is either induced through an applied magnetic field or inherent to the threat within the detection volume. The third field sensor may be arranged to receive and sense a field substantially perpendicular to the first and second field sensor.

Each of the field sensors may comprise air core coils. The first field sensor may be aligned with the prescribed direction.

Some embodiments relate to a probe for an AED apparatus for detecting a threat within an object passing through a detection volume in a prescribed direction, the probe comprising: a first sensor to receive magnetic and/or electric fields from the threat within the detection volume. The probe may also comprise a second sensor to receive magnetic and/or electric fields within the detection volume, substantially perpendicular to the first sensor; and optionally a third sensor to receive magnetic and/or electric fields within the detection volume, substantially perpendicular to the first and second sensors.

Each field sensor may comprise an air core coil. The first field sensor may be aligned with the prescribed direction. The first and third coils may comprise single turn saddle coils and the second coil may comprise a single turn coil. Some embodiments relate to a probe for a QR apparatus for detecting a threat within an object passing through a detection volume in a prescribed direction, the probe comprising: a coil to produce a substantially horizontal magnetic field within the detection volume, parallel to the prescribed direction; wherein the coil is a single sheet turn coil.

Some embodiments relate to a threat detector for detecting a threat within an object passing through a detection volume in a prescribed direction that includes one or more probes comprising one or more of the following: (i) a first coil to produce a substantially vertical magnetic field within the detection volume; a second coil to produce a substantially horizontal magnetic field within the detection volume, parallel to the prescribed direction; wherein, the first coil comprises two single turn saddle coils connected in parallel and the second coil is a single turn coil;

(ii) a first coil to produce a substantially vertical magnetic field within the detection volume, a second coil to produce a substantially horizontal magnetic field within the detection volume, parallel to the prescribed direction; and a third coil to produce a substantially horizontal magnetic field within the detection volume, perpendicular to the prescribed direction; wherein, all coils are multi-turn rectangular coils;

(iii) a first coil arranged to produce a substantially vertical magnetic field, a second coil arranged to produce a substantially horizontal magnetic field, parallel with the prescribed direction, and a third coil arranged to produce a substantially horizontal magnetic field perpendicular to the prescribed direction for detecting metal enclosed objects; wherein the first and third coils are single turn saddle coils and the second coil is a single turn coil; and

(iv) a coil to produce a substantially horizontal magnetic field within the detection volume, parallel to the prescribed direction; wherein, the coil is a single turn coil.

Some embodiments relate to systems and methods utilising one or more of the described probes. Some embodiments relate to computer-readable storage storing executable program code to cause a computer to execute any of the described methods.

Brief Description of the Drawings FIG. 1 is a block diagram of an arrangement of some embodiments. FIG. 2 is a flowchart of a method of some embodiments.

FIG. 3 shows a side view of the non linear junction subsystem inside a shield.

FIG. 4 shows axial magnetic field producing coils for a NLJD detection subsystem inside a shield.

FIG. 5 shows a vertical magnetic field producing transmit coil for the NLJD detection subsystem inside the shield.

FIG. 6 shows a vertical magnetic field producing receive coil for the NLJD detection subsystem inside the shield. FIG. 7 is a block diagram of the NLJD detection subsystem.

FIG. 8a shows the typical magnitude plot for a threat detection in the NLJD detection subsystem. FIG. 8b shows the typical complex and real plots for a threat detection in the NLJD detection subsystem. FIG. 9 shows a transmit/receive system for an AED subsystem.

FIG. 10 shows a single sheet receive coil for the AE detection subsystem.

FIG. 11 shows a vertical magnetic field producing receive coil for the AE detection subsystem.

FIG. 12 shows an axial magnetic field producing receive coil for the AE detection subsystem.

FIG. 13 shows a horizontal magnetic field producing receive coil for the AE detection subsystem. FIG. 14 shows all three AED coils inside a shield. FIG. 15 is a block diagram of the AED subsystem.

FIG. 16 shows a single turn sheet coil for the QR detection subsystem inside a shield.

FIG. 17 is a block diagram of the QR subsystem.

FIG. 18 shows an arrangement for an electronic circuit detection subsystem alongside a metal detector subsystem.

FIG. 19 shows an axial magnetic field producing receive coil for the metal detector.

FIG. 20 shows a horizontal magnetic field producing receive coil for the metal detector. FIG. 21 shows a vertical magnetic field producing receive coil for the metal detector. FIG. 22 is a block diagram for the metal detector subsystem.

FIG. 23 shows an arrangement of a NLJD detector used as an electronic circuit subsystem and a metal detector subsystem. FIG. 24 shows an arrangement of an electronic circuit detection subsystem alongside an explosives detector subsystem.

FIG. 25 shows an arrangement of an electronic circuit detection subsystem alongside a metal detector subsystem and an explosives detector subsystem.

FIG. 26 shows an arrangement of two electronic circuit detection subsystems alongside a metal detector subsystem and an explosives detector subsystem. FIG. 27 is a block diagram of the NLJD detector being used as an NLJD subsystem and an AED subsystem.

FIG. 28 is a block diagram of the NLJD detector being used as an NLJD subsystem and an MD subsystem.

FIG. 29 is a block diagram of the NLJD detector being used as an NLJD subsystem and an explosives detector.

FIG. 30 is a block diagram of the NLJD detector being used as an NLJD subsystem, a MD subsystem and an explosives detector.

FIG. 31 shows a metal object approximately half way between the transmitting and receiving coils of a metal detector. FIG. 32 shows the signal strengths received for the object in Figure 31. FIG. 33 is a block diagram of some embodiments.

FIG. 34 is a flowchart of a process of some embodiments.

FIG. 35 is a block diagram of some embodiments.

DETAILED DESCRIPTION Embodiments will now be described by way of example, with reference to the accompanying drawings. Some embodiments are directed to inspection and/or detection methods and a detector that correlates two or more detection methods to detect threats in objects in a detection volume. In particular, embodiments address the problem that some threats are typically made up of more than one significant component. The security threat may include, for example, one or more of the following components: non-linear junctions, batteries, timers, controllers, wires, initiators, hand held phones, explosives, radioactive materials, biological materials, chemicals, closed circuit loops, connectors or metal objects.

Throughout the specification an "object" is defined to mean cargo, luggage, packages, mail, container, circuit boards, manufactured item, clothing, handbag, sachet, briefcase, shoes, minerals, food, manufactured pharmaceuticals or any transportable item that needs to be screened for threats. A "threat" is defined to mean some substance, device, component or quality that can pose a security threat or problem and which is associated with or contained within the object and needs to be detected before the object is cleared from the detector. Some examples of a threat could be, but are not limited to: an Improvised Explosive Device (IED); Improvised Biological Device (IBD); Improvised Nuclear Device (IND); Improvised Chemical Device (ICD); Improvised Radioactive Device (IRD); Improvised Incendiary Device (HD); Weapons of Mass Destruction (WMD); unwanted contaminants, such as a powered mobile phone which has fallen into the object; contraband, such as a mobile phone which a perpetrator is trying to smuggle into a prison or a courthouse or any object that can produce an electromagnetic field when subject to or not subject to an electromagnetic field. A "subsystem" is defined to mean a device which can detect one or more threats through one or more detection methods. The detection/inspection systems may comprise one or more subsystems. In the case where there are two or more subsystems, these may be connected or unconnected but placed in the general vicinity of each other (eg. adjacent) and operated in sequence or in parallel.

Some embodiments improve the detection of threats within objects by correlating the results of a number of threat detection methods. The number of detection methods may be at least two, at least three, at least four, at least five or six or more. If at least two detection methods produce a positive detection of a threat then an alarm is raised. Even more preferably, if two or more detection methods produce the positive detection of a threat and at least two detection methods identify that the threat occurs at the same location within the object then an alarm is raised. This method of correlation of the results of two different detection methods reduces the number of false alarms caused by spurious signals, which can occur in one detection method but not in other detection methods. For example, an extremely sensitive detector may passively listen for noise emanating from the scanned object. Its internally circuitry may randomly produce noisy signals or it may receive noisy signals from external sources. However, unless one or more other detection methods in the threat detector produce a positive detection then no detection of a threat will occur. In the case where the signals need to be correlated in space then if one detection method indicates that there has been the detection of a threat at one end of the object and another detection method indicates a detection has occurred at the other end of the object, then the object is passed as being clear because they are not spatially correlated. The detection methods target detecting different components of a threat. One detection method can be aimed at detecting countermeasures. By combining at least one detection method which can detect countermeasures taken by a terrorist and at least one detection method which can detect at least one component of a threat, there is a reduction in the risk that the object is passed as being clear of threats. If there were only one detection method and that method were susceptible to a countermeasure, then the performance of the overall detector can be easily compromised. By adding a second detection method which focuses on some other part of the threat then the detection rate is improved. In the case of an IED, an example of a countermeasure would be enclosing the explosive in a metal to make it harder to detect by Quadrupole Resonance. Another example would be wrapping the circuitry used to detonate an IED in a metal surface to prevent it from being detected. These particular countermeasures can be detected by using a metal detector.

Figure 1 shows N subsystems 21, 22, 23 side by side connected by one conveyor belt 25. It may be possible that there is only one subsystem, in which case that subsystem can scan the object by two, three, four, five, six or more different detection methods.

The N subsystems 21, 22, 23 house at least one coil located inside a metal shield to screen out external interferences. In some cases the subsystems are passively listening for noise and thereby do not generate a transmit signal and in other cases there is at least one coil generating a transmit field and at least one coil which acts as a receiver.

Each transmit coil generates an electromagnetic field substantially in one direction. In the case where there is both transmission and reception, the object to be scanned is moved by the conveyor belt into the first subsystem's detection volume. Either during the motion of the object into the detection volume and through to the other side, or after the object has arrived at the centre of the detection volume, the coil transmits electromagnetic pulses at a frequency between 1 Hz-IOOMHz or a wavelength which is greater than the longest dimension of the object. These pulses interact with a component of the threat inside the object and cause the component of the threat to produce an electromagnetic field of its own. The receiver coil (which may be the same coil as the transmitter) receives this electromagnetic field which induces a voltage on the receiver coil, which is then filtered and down converted (if required) and sampled by an Analogue to Digital Converter (ADC) inside a computer. The digital signal is then Fast Fourier Transformed and a peak in the Fourier spectrum is compared to a known threshold. If the peak lies above a threshold, then this first detection method in this first subsystem has recorded a positive detection of the component it was designed to detect.

This process is then repeated if the first subsystem is able to detect components of a threat by one or more different detection methods.

In the case where there is only reception of noise emanating from the object, the same method is used, except that there is no transmission of a signal. The processing of this signal can be as explained above or the height of the time series data is compared to a threshold. In either case if the peak frequency signal lies above a threshold or the peak height of the time series data lies above a threshold then the subsystem has recorded a detection of the component it was designed to detect.

After determining all of the results of the first subsystem the object is moved into the second subsystem and the scanning process above is repeated. This continues until all N subsystems have scanned the object by two, three, four, five, six or more detection methods. If any two detection methods in any of the subsystems have resulted in a positive detection then an alarm is raised and the object needs to be searched or quarantined. Some embodiments are directed to improving the detection of threats in objects by sensing the presence of a component of a threat by one detection method and then improving the detection of the threat by selecting one, two, three, four, five, six or more detection methods which target the detection of different components of the threat and then combining those two, three, four, five, six or more detection methods into the one detector to achieve superior detection. An example here may be in the detection of IED's. The first detection method identifies that the IED can be detected by detecting the explosive in the IED. It is also recognised that the IED can also be detected by detecting two other components of the IED: a battery and an electronic circuit.

Figure 2 shows a flow diagram of the process in identifying a new detection method and combining it into the detector. In the first step 31 a detection technique is identified and added to a detector. In the second step 32, another component of the threat is identified and then a new detection technique is added to the detector. Preferably, this new technique will be independent of the first technique and broaden the scope of the components the detector can detect. In step 37 the value of the new technique is evaluated. If it improves the detection rate it is retained 36 otherwise it is discarded 38.

As a matter of convenience, the remainder of the embodiments are described in relation to the detection of IED's in an object, however embodiments not limited to this application and an IED is only one threat that could be detected by the embodiments. Other threats that could be detected include, for example: IBD's; IND's; ICD's; IRD's; IID's; WMD's; unwanted contaminants, such as a powered mobile phone which has fallen into the object; contraband, such as a mobile phone which a perpetrator is trying to smuggle into a prison or a courthouse or any object that can produce an electromagnetic field when subject to or not subject to an electromagnetic field. The object which contains threat could be cargo, luggage, packages, mail, container, circuit boards, manufactured item, clothing, handbag, sachet, briefcase, shoes, minerals, food, manufactured pharmaceuticals or any item that needs to be screened for threats or anomalies. Some embodiments include at least one subsystem which can detect IED's within large volume objects that are scanned whilst on a conveyor belt by sensing the presence of electronic circuits.

Figure 3 shows an example of embodiments in which the subsystem sensing the presence of electronic circuits is a Non Linear Junction Detector (NLJD). The item to be scanned is placed on the conveyor belt 1 and is scanned by the series of transmit and receive coils 3, 5, 7, 10 and then exits the scanner. This design will be further explained below.

Most IED devices need a switch to trigger the device to explode. This switch can come in various forms; however the electronic element that tends to be common in these devices is a Non Linear Junction (NLJ). A typical example of an NLJ is a semiconductor diode. These NLJ's can be detected by stimulating the NLJ at one frequency and then receiving signals from the NLJ at the 2nd or 3rd harmonic of the original transmit signal.

The NLJ detector can also detect ferrous metals.

To scan large objects moving along a conveyor belt, a new and unique coil design was built as compared to the standard small handheld NLJD detectors commercially available. This coil design includes two separate coil types:

(i) Axially oriented magnetic field producing balanced coil; and

(ii) Vertical oriented magnetic field producing double saddle transmit coil and a quadruple saddle receiver coil.

As shown in Figure 4, the axial balanced coil design produces magnetic field lines 4 which are parallel to the conveyor belt 1. The transmit coil 3 is a single loop of copper and is resonated at a particular frequency by placing high Q capacitors between the gap at the top of the coil (not shown). The optimum transmit frequency for detection of NLJ's has been found to be approximately 9MHz, but other values could also be used. The receiver coil 5 also produces substantially horizontal and axial like field lines 4. This coil is unconnected to the transmit coil and is resonated at 2 times the resonant frequency of the transmit coil. This coil could be described as being two single loops of copper connected together in series, such that when the transmit coil generates a magnetic field, the current induced on both receiver coils is balanced because the currents will be travelling in opposite directions. If an IED enters the transmitter's magnetic field this balance will be affected and as a result there will be an increased current flow in the receiver coil and thus the IED will be detected. Both of the transmitter and receiver coils are housed inside a 2mm thick aluminium shield 14 to screen out unwanted RF interference and prevent the resonant fields from within the coils escaping the localised area. The use of a shield also prevents the subsystem from radiating unwanted noise into the environment and the transmit field from interacting with an object nearby causing unwanted noise. As shown in Figure 5, the saddle coils produce magnetic field lines 9 which are predominantly vertical in nature. The transmitter 7 consists of two copper saddle coils arranged to be back-to-back allowing an object to pass between the two saddle coils along the conveyor belt 1. This coil is resonated by placing capacitors across the gap 8. The receiver coil 10 (Figure 6) consists of quadruple copper saddle coils which are connected in series. One side of this coil produces vertical magnetic field lines 11 pointing upwards and the other side produces magnetic field lines which point downwards 12. Similar to the axial design, this coil system is balanced as the current induced on the two halves of the receiver coil 10 by the transmitter 7 cancel. If an IED disturbs this balance then it can be detected. Preferably, in practice, the transmitter and receiver coils are placed inside the one shield 14.

Both sets of coils have a medium Quality factor, Q, to ensure sufficient power is input into the NLJ' s so they produce a response. A typical value for the Q would be near 50. If the Q was too low it would be difficult to detect the NLJ's. Similarly, if the Q is too high, because this is a balanced system, the system becomes too sensitive to extraneous electrical effects; therefore it is optimal to have a medium Q.

Both sets of coils have been carefully made such that they can be placed into the one shield without contacting each other and will operate alongside each other without any detrimental effects (i.e. they are decoupled from each other). This allows the same volume to be used for the axial and vertical magnetic field producing coils and the two sets of coils produce perpendicular fields to each other. The use of perpendicular fields maximises the possibility of detection of NLJ's by ensuring the NLJ is irradiated with radiation from at least two different directions in case they respond poorly in one particular direction.

To pass an object through the coil system, a conveyor belt 1 runs just above the bottom edge of the coils. This conveyor belt is driven by a motor 2 located well back from the coils to prevent RF interference. The surrounding shield 14 has open ends which allow passage of objects through the scanner while providing shielding to external magnetic fields. Figure 7 shows the block diagram for NLJ detector. To scan an object, it is placed onto the moving conveyor 18 until it breaks the optical fence 19. The PC Control 20 then initiates the RF signal generator 30 to begin generating RF pulses at the resonant frequency of the axial transmit coil 110. These pulses are then filtered 70, 90; amplified 80 and sent to the axial transmit coil 1 10. While these pulses are being sent to the axial transmit coil, the axial receiver coil 160 receives any induced RF signal at the second harmonic of the transmit frequency. This signal is then filtered 120; transformed into quadrature signals 120; demodulated 50 from MHz signals down to kHz signals and then both quadrature signals are sampled at 0.25MHz 40 to create digital signals and stored in the DSP's memory 40. The sampling occurs for 560μs generating 140 points of data. It should be appreciated that the preferred transmit frequencies may vary depending upon the application. For a small carry-on size scanner, an exemplary transmit frequency is at or around 4.4 MHz, with a separation between the axial and vertical transmit frequencies being approximately 100kHz.

This process is repeated for the vertical field producing coils, with the main exception being that the transmit frequency is slightly different to the axial coils to avoid any coupling effects between the two sets of coils. This process of sampling and switching between the axial and vertical field producing coils is further repeated a number of times (typically 480) until the object has passed through the coils and is well clear of the scan area. At this point the two lots of 480 datasets of 140 points of data are sent to the computer 20 for analysis. It may not be necessary to scan the object in so many repetitive steps as the object passes through the detector for this NLJD and the following active electronic detector and the metal detector. As it is possible to perform only two measurements, the first being a baseline with the object outside the coils and then use the conveyor belt to move the object into the detection volume, stop the conveyor and perform a second measurement. Thereby only two measurements are performed on each object to determine if an NLJD is present. This method may be useful if the conveyor belt motor generates a large amount of RF noise.

To process the digital signals from both the axial and vertical receive coils they are initially analysed by subtracting off the mean of each dataset. The data is then smoothed by taking the median value of every eighth point to remove any random spikes occurring in the data. This reduces the data size down to two sets of 60 datasets instead of 480 datasets. These 60 datasets are then Fast Fourier Transformed (FFT). The complex FFT signals in the demodulated data which would correspond to 2 times the original transmit frequency are recorded for each of the 60 datasets. The sixty peak FFT complex data points are background corrected by performing a polynomial fit to these data and subtracting off the resulting fit. This correction is performed to eliminate drift that can occur in the data. The polynomial fit has a low order, typically 9, so it can remove the slow drift but does not remove the typical narrow NLJ signal. Figures 8a and 8b show the final result of processing for a NLJ signal for the axial coil system. Typically in Figure 8a the magnitude occurs as two narrow peaks centered around where the IED was located. The second plot, Figure 8b, shows the complex and real components of the signal in Figure 8a. After performing this correction, the maximum FFT signal is then used to determine the final result of the NLJD processing. The most reliable result was found by taking the logical OR of the axial and the vertical coil maximum peak heights recorded for the sixty data points and comparing this result against a previously set threshold. If the ORed result exceeds the threshold, then the NLJD subsystem has recorded a positive detection, otherwise the object is passed as being clear of NLJDs.

Further embodiments are shown in Figure 9. Such embodiments relate to an Active Electronics Detector (AED). Active electronics are those electronics which are powered (active) and may be present in an IED. The detection of AE relies on detecting time varying magnetic and/or electric fields. Changing voltages and currents within a circuit will cause electromagnetic fields to be generated. In this example magnetic field detection is shown.

The AED is a tuned receiver which passively listens to electronic noise emanating from an object. If it detects above average RF noise this may indicate the presence of an" IED.

The active electronics detector consists of up to 4 separate coils in this example. The format of two antenna systems is shown, each effectively independent when sensing the object in time. This example displays two preferable independent system designs, which have been arranged in series to increase detection performance over each part. The first system is a single turn sheet coil 200 (Figure 10) which is tuned to a particular frequency and lies within a shield. The coil 200 is resonated by placing high Q capacitors across a gap at the top of the coil (not shown). An optimal frequency for detection of Active Electronics (AE) was found to be between 2-3MHz, as this reduced the influence of external interferences. This particular coil was resonated at 2.3MHz. The first coil receives noise in one orientation 205 at the centre of the system along the axis of the coil. The coil does have high sensitivity in the directions for moving objects since the effective directional field sensitivity is not the same throughout the detection volume. The antenna does have high sensitivity compared to the following coils because of high unloaded Q which makes it useful for detection of low level AE.

The other three-coil system may be adjacently located further along the conveyor belt (see Figure 9) in a separate shield and has magnetic field lines which are oriented in three different directions. The purpose of having three field directions is to ensure detection of AE, some of which may only emit noise in one particular direction. Figure 11 shows the first of these coils 220 which receives substantially vertical magnetic field lines 225 and is resonated at 2MHz. High Q capacitors 222 used to resonate this coil are placed in the gap shown at 221. Figure 12 shows the second of these coils 230 which receives magnetic field lines 235 which are substantially horizontal and parallel with the conveyor belt. This coil is resonated at 2.1MHz. High Q capacitors 232 used to resonate this coil are placed in the gap shown at 231. Figure 13 shows the third of these coils 240 which receives horizontal magnetic field lines 245 which are perpendicular to conveyor belt and is resonated at 2.2MHz. High Q capacitors 242, 243 used to resonate this coil are placed in the gap shown at 241.

Figure 14 shows all three sets of coils 220, 230 & 240 co-located inside a 2mm thick aluminium shield 226. All coils can operate simultaneously within this shield without affecting each other. Figure 15 shows the block diagram for receiving the signal from the single turn sheet coil and the other three coils. An object is placed on a moving conveyor 278 passes through an optical fence. By passing through the optical fence, the system is triggered into acquiring data. It does this by receiving noise on the tuned receivers 250-253. This noise is then amplified 254-257, high pass filtered 258-261, and band pass filtered 262- 265. The high pass filters 258-261 have a cut off frequency at 1.5MHz, which reduces the effects of any noise coming from the radio AM band of frequencies. The data is then processed in two different paths. In one path the signals are then combined together 266, down converted 269,270, digitised in an ADC 271 and sent to the computer 275 for analysis and in the second path the data is sent to what is called a glitch detector to detect anomalies. In the first path, the data is sampled at lMS/s and is narrow band data. In the second path the data is combined into one signal and is wide band data.

As AE objects do not always emit RF noise there is a need to collect as much data as possible otherwise the emissions may not be detected. However, continually streaming data is probably impractical; hence the data was sampled at high speed.

The glitches were detected by using an analogue circuit. This circuit counts how many times the signal's voltage exceeds a preset threshold. This glitch detector also looks at the rate of increase to determine whether the signal is a glitch, i.e. slow increases in voltage are ignored and rapid increases are counted. If it exceeds the threshold then it is counted as a glitch. This process is repeated a number of times as the object passes through the scanner. The number of glitches is recorded and is used as an indicator of an IED. For the other method, the data is processed within the computer. If the maximum of digital time data exceeds a threshold then the AED has detected an object which possibly contains an IED.

The computer collects the results from these two processing methods and combines them together using a logical "OR". If either of the processes produces a positive result then possibly an IED has been detected, otherwise the object is clear of IED's. If an IED has been detected then the computer either displays an alarm on the computer monitor or signals an audio or visual alarm separate from the computer. In a third example of this embodiment, a Quadrupole Resonance (QR) detector is used to detect AE and Stimulated Emissions (SE). In a baggage scanning application, a QR detector is built to only detect explosives and interferences generated by other objects are generally unwanted as they increase the false alarm rate and make reliable detection difficult. In this current application it is desirable to not only detect the explosives but also detect the AE and SE as these may be an indicator that an IED is present. The same objects that would be detected by the AED described previously should also be detected by the QR detector. However, because the QR detector is transmitting a pulse of high energy into the object it can also stimulate electronics to begin emitting RF signals which then can be received by the QR detector. The AED because it is a passive unit, will not receive any signals from these devices.

QR systems have been built for some time now and have been shown to exhibit high detection rates and low false rates in certain circumstances. The QR scanner operates by sending high power transmit pulses into a single turn sheet coil which excites and aligns the QR nuclei. After the transmit pulse has been switched off, the nuclei relax back to their starting positions and in the process radiate small RF signals which can be detected by inducing a small voltage on a high Q coil.

Figure 16 shows the high Q single sheet coil 600 within a shield. This coil is resonated by placing high Q capacitors along the gap at the top of the coil (not shown). The shield 14 has open ends which allow objects to travel through the scanner on a conveyor belt 628, and at the same time screens out external interferences.

Figure 17 shows a block diagram of the QR system. QR frequencies are dependent on the temperature, hence preferably at the beginning of the measurement process, the computer 620 measures the ambient temperature via the probe 644 and memorizes the transmit frequency such that it corresponds to the resonant frequency of the explosive at the measured ambient temperature. At the start up of the conveyor belt 641 an object is carried on the belt past an optical fence (not shown). The breaking of the fence signals to the computer 620 that an object is present that needs to be scanned. The computer 620 preferably moves the belt 641 forward until the object lies approximately in the centre of the coil 629.

The process of placing an object inside the resonant coil system can cause the resonant frequency of the tank circuit to change. An example here would be a metallic object which would typically lower the inductance of the coil and require the capacitance to be preferably changed to bring the coil's resonant frequency back to the QR resonant frequency previously determined after measuring the temperature. This is achieved by switching in capacitor values 632 into the circuit such that the resonant frequency of the circuit moves from the lowest possible value to the highest possible value, and after each capacitor is switched into the circuit transmitting a small signal through the transmit circuit (the operation of which will be explained further shortly) and then sensing the resultant magnetic field on the tune loop 645. After this process has been the completed the capacitor value at which the maximum signal was received on the tune loop is determined and then the coil is retuned back to the QR resonant frequency by calculating how much capacitance needs to added or subtracted from the circuit to realign it with the QR frequency. Once this number has been calculated this capacitance is added or subtracted out of the circuit.

After the tank circuit has been correctly tuned, the computer 620 then initiates the scanning sequence, which begins by generating a series of small amplitude pulses in the Pulse Generator Controller 623. These pulses consist of oscillating sinusoidal voltages near the QR resonant frequency (between 0.5-5MHz). Each pulse is sent to the RF power amplifier 624, the transmit isolator 625, and low pass filter 626. Then a switch 627 is used to switch the transmit pulses between explosives, this is done to switch in a different matching circuit 628 to ensure the power is transferred efficiently. After leaving the matching unit 628, the pulse is transferred to the resonant probe coil 629.

Once a pulse has finished the circuit disconnects from the transmitter via the transmit isolator 625 and becomes a receiver circuit. As stated previously the high power pulses tip and align the nuclei. Once the high power pulse has been removed the nuclei relax back to their equilibrium position and in the process release a small varying magnetic field. The same single sheet coil and resonant circuit used for transmitting the pulses receives this small varying magnetic field from any QR material present in the scan volume. This small magnetic field induces a voltage on the coil which passes through the receive isolators 633,634, Pre amplifiers 635,636 and to another switch 637 which selects which isolator and preamplifier is used for each explosive measured. The signal is then band pass filtered 639 to remove out of band unwanted noise and down converted 640 from MHz signals down to kHz signals. The down conversion also converts the signal into quadrature signals. The Digitiser 622 samples the two quadrature signals from the receiver at approximately IMHz and stores these digital signals in memory.

This process is preferably repeated up to thousands of pulses to accumulate more blocks of data. After the thousands of pulses have been transmitted and the data accumulated, these blocks of data are all averaged together to form the final data which is analysed in the computer. The averaging process is performed due to the fact that the QR signal is quite weak and therefore averaging is required to increase the signal to noise ratio.

For explosive types such as RDX which produce a strong QR signal for a selected quantity, the conveyor belt does not need to be stopped for scanning. The procedure in the case where it is to be used with another similar technology such as an NLJD may interlace the two scan processes. Optionally, the scanning processes can be overlapped.

Other coils that are described as "switched out" could be preferably switched in parallel to create a more uniform magnetic field within the system and a more uniform receive signal intensity within the scan volume. Optionally, the phases to the coils can be controlled to produce a particular polarisation to the RF field, such a circular polarisation. The degree of signal degradation caused by a less favourable tuning point may be more acceptable than the increase in machine complexity/cost to including tuning circuitry and algorithmic procedures. It is possible to find a satisfactory tuning point in most cases where the object has low coupling to the QR coil. This is the case for a class of objects transported as organic cargo.

The final averaged data are then Fast Fourier Transformed and if the peak signal that lies within a small frequency window lies above a nominated threshold for either explosive material then an explosive has possibly been detected. If either the electronics detecting subunit or the explosives detector subunit returns a positive result then the object is flagged as having a potential IED. If an IED has been detected then the computer either displays an alarm on the computer monitor or signals an audio or visual alarm separate from the computer. Various pulse sequences can be used in different situations, and the description of the variety of pulse sequences is beyond the scope of the present invention but discussions on these sequences can be found in U.S. Patent 7,355,400 and U.S. Patent 7,282,913, the disclosures of which are hereby incorporated by reference. In some embodiments, any two or three of the previously described detectors (NLJD, AED, QR) are placed alongside or co-located with each other and the object to be scanned is passed or placed in the subsystems to detect IED's. The scan procedure for each subsystem is as previously described in this embodiment and the results of the subsystems are combined by taking a logical "OR". If any of the subsystems generate an alarm then an IED has been potentially been detected.

Embodiments shown in Figure 18 are substantially the same as some previously described embodiments except that as well as having at least one subsystem which can detect the presence of IED's by sensing the presence of electronic circuits, there is at least one subsystem which can detect the presence of metal objects, which may be batteries, timers, connectors, wires, controllers, initiators, hand held phones, closed circuit loops, metallically shielded NLJ's, AE's, electronics which produce SE's or explosives. One of the most important points about this metal detector is that it is able to detect the obvious countermeasure of shielding the threat such as an IED surrounded by metal. As a result, it becomes increasingly difficult for a terrorist to generate a method which allows the IED to pass through the system undetected.

The object to be scanned is moved along the conveyor 730 into the subsystem which detects the presence of metallic objects 700 and then after that subsystem has finished scanning the object it is moved into the second subunit 710 which determines if the object contains any electronic circuits. The order of the units can be optimised to some principle, for example, requiring the less invasive to precede a more invasive technique, to for example reduce the (if any) chance of damage to scanned articles during scanning. The large volume NLJD or AE detectors described previously are two examples of subsystems which could be used to detect IED's and the following metal detector could be an example of a metal detector to detect the presence of the batteries, metal in the circuit components or shielded object.

The metal detector is designed to detect metal objects within the scanned object. An example is a power source within an IED which can be a battery or a series of batteries.

Preferably, the metal detector is tri-axial as the batteries and shielded NLJ's, AE's and explosives may have any shape and orientation and thus may be difficult to detect if only one set of coils is used. The tri-axial system consists of the axial, vertical and horizontal magnetic field producing elements. For the axial field producing coil (Figure 19), a narrow single turn coil 300 serves as the transmit coil and the receive coil 310 consists of two single loops which are connected in series either side of the transmit coil 300. This design is identical in form to the axial magnetic field producing coil for the NLJD. When the transmitter 300 generates a magnetic field it induces opposing currents in the two loops which cancel. If a metal object enters the magnetic field it reduces the current induced on one of the coils and thus it can be detected by the change in current circulating in the receiver coil 310. The horizontal transmit/receive system (Figure 20) consists of two large transmit coils left hand side 400 and right hand side 410 and within those coils are each located two smaller receive coils 420,430,440,450. During operation the transmit coil 400 transmits first and induces currents on receiver coils 440,450. The size of the current induced is measured and stored. If a metal object enters the magnetic field, the current induced on the receiver coils changes and thus the metal object can be detected. The same operation occurs with the right hand side transmit coil 410 and the associated receiver coils 420,430.

The vertical transmit/receive system (Figure 21) is similar to the horizontal system except that the coils are located top and bottom of the scan volume. Again there a two transmit coils 500,510 with two smaller receiver coils 520,530,540,550 located within them. The operation of the transmit/receive coils is identical to those in the previous horizontal section. To operate the system, the conveyor belt is set into motion and an object is placed onto it. This object passes through an optical fence which triggers the system to begin transmitting and receiving signals. As shown in Figure 22, the transmit signal for all of the coil designs is generated in a DSP 561, amplified 562-564, and sent to each transmit coil 565-569, in turn. The transmit signal typically contains three frequencies (1 kHz, 3 kHz and 10 kHz) to maximise the detection potential of various targets. The use of three frequencies also enables some discrimination of the type of metal detected to provide more information to the user. During transmission of the transmit signal a voltage is induced on the receiver coils 570-578. This signal is then filtered 579, amplified 580 and digitised 561 and sent to the computer 560 for processing. Each dataset is created by sampling the signal at 100kHz for 41ms which creates an array of 4096 points. After one transmit coil has been excited it is switched off and then the next transmit coil is excited until all coils have been excited. In the case of the horizontal and vertical field producing coils the signals from the two smaller coils on the opposite side of the scan volume are recorded simultaneously through the above procedure. Once all the coils have been excited the cycle repeats until the object has cleared the scan area. The number of cycles is typically around 17 steps before this occurs.

During the processing, there are 17 datasets from each of the 9 receiver coils each containing 4096 points. Every dataset is first Fast Fourier Transformed into frequency space and the complex peak heights at the three frequencies are recorded. This reduces the data to 17 datasets x 9 receivers x 3 frequency peaks. Then background complex FFT peak signals are subtracted off these signals (this background signal was collected prior to the object passing through the scan area and processed identically up to this point). To further reduce the data each set of 3 frequencies is combined to make one frequency peak by weighting the lowest frequency by the mean peak signals of the three frequencies divided by the median. If this weight ratio is greater than one then it is inverted before the weighting occurs. The purpose of doing this weighting is to reduce any variations caused by noise in the system.

Once this weighting has been completed each receiver signal for the 14 steps is compared to a corresponding threshold. If any of the peak signal strengths exceed the threshold then a positive detection has occurred which may indicate the presence of a metal substance.

If either the electronic circuit sensing subunit or this metal detector produces a positive detection then the object is deemed to be containing a potential IED. If an IED has been detected then the computer either displays an alarm on the computer monitor or signals an audio or visual alarm separate from the computer.

Another example of this embodiment would be the NLJ detector by itself. As previously indicated the NLJD can detect electronic circuits. However, an NLJ detector can also detect ferrous materials, and in particular ferrous metals, because these materials generate harmonic signals. Hence, the NLJD can be used for detecting electronic circuits and ferrous metals. Ferrous metals are typically used as the outer surface of alkaline batteries. In Figure 23, the object to be scanned is transported along a conveyor 730 into an NLJD detector 700. Using the method and block diagram previously discussed in the first embodiment, the object is scanned for both NLJ's and ferrous metal objects. If either characteristic is detected then either the computer displays an alarm on the computer monitor or signals an audio or visual alarm separate from the computer.

Embodiments shown in Figure 24 are substantially the same as some previously described embodiments except that as well as having at least one subsystem which can detect the presence of IED' s by sensing the presence of electronic circuits 700, there is at least one subsystem which can detect the presence of explosives 715. The large volume NLJD or AE detectors described previously are two examples of subsystems which could be used to detect IED 's and the previously described Quadrupole Resonance (QR) detector could be an example of a detector to detect the presence of explosives. Other techniques which could be used to detect explosives are X-rays, Neutrons, Nuclear Magnetic Resonance (NMR), gamma rays, trace detection and Terahertz imaging. If any of the subsystems produce a positive result then potentially an IED has been detected. If an IED has been detected then the computer either displays an alarm on the computer monitor or signals an audio or visual alarm separate from the computer.

Further embodiments are substantially the same as some previous embodiments except that as well as having at least one subsystem which can detect the presence of IED's by sensing the presence of electronic circuits, there is at least one subsystem which can detect the presence of large metal objects and there is at least one subsystem which can detect the presence of explosives. The large volume NLJD or AE detectors described previously are two examples of subsystems which could be used to detect electronic circuits; the previously described MD could be used to detect metal objects and the previously described Quadrupole Resonance (QR) detector could be an example of a detector to detect the presence of explosives. Other techniques which could be used to detect explosives are X-rays, Neutrons, Nuclear Magnetic Resonance (NMR), gamma rays, trace detection and Terahertz imaging. Some of the subsystems can detect more than one component of an IED. For instance, as previously mentioned, the NLJD can detect ferrous metals which tend to be the main outer layer of batteries and the QR system can detect magnetoacoustically ringing objects; stimulated electronics, and active electronics. The QR system can also detect piezoelectric materials. This can be important if the explosive cannot be detect by QR but is in powdered form and generates a piezoelectric signal.

In Figure 25 the object is moved through all three subsystems 700,715,725 on top of a conveyor belt 730. If any of the subsystems produce a positive result then potentially an IED has been detected. If an IED has been detected then the computer either displays an alarm on the computer monitor or signals an audio or visual alarm separate from the computer. Further embodiments are substantially the same as some previous embodiments except that there are two subsystems which can detect the presence of IED 's by sensing the presence of electronic circuits. There is also at least one subsystem which can detect the presence of large metal objects and there is at least one subsystem which can detect the presence of explosives. The large volume NLJD and AE detectors described previously are two examples of subsystems which could be used to detect electronic circuits; the previously described MD could be used to detect metal objects and the previously described Quadrupole Resonance (QR) detector could be an example of a detector to detect the presence of explosives. Other techniques which could be used to detect explosives are X-rays, Neutrons, Nuclear Magnetic Resonance (NMR), gamma rays, trace detection and Terahertz imaging. In Figure 26 the object is moved through all four subsystems 700,701,715,725 on top of a conveyor belt 730. If any of the subsystems produce a positive result then potentially an IED has been detected. If an IED has been detected then the computer either displays an alarm on the computer monitor or signals an audio or visual alarm separate from the computer.

Further embodiments comprise at least one subsystem which can detect the presence of IED's by sensing the presence of electronic circuits by two separate methods. The large volume NLJD described previously is capable of detecting IED's by sensing the presence of electronic circuits by detecting NLJ' s and passively listening for AE. This design reduces the complexity of the device and also reduces the footprint of the instrument. Reduction of the footprint of the threat detector can be important in spatially constrained areas, such as airports.

Figure 27 shows the block diagram for combined NLJ and AE detector. To scan an object it is placed onto the moving conveyor 1018 until it breaks the optical fence 1019.

The PC Control 1020 then initiates the RF signal generator 1030 to begin generating RF pulses at the resonant frequency of the axial transmit coil 1110. These pulses are then filtered 1070, 1090; amplified 1080 and sent to the axial transmit coil 1110. While these pulses are being sent to the axial transmit coil, the axial receiver coil 1160 receives any induced RF signal at the second harmonic of the transmit frequency. This signal is then filtered 1 120; transformed into quadrature signals 1120; demodulated 1050 from MHz signals down to kHz signals and then both quadrature signals are sampled at 0.25MHz 1040 to create digital signals and stored in the DSP's memory 1040. The sampling occurs for 560μs generating 140 points of data. This process is repeated for the vertical field producing coils, with the main exception being that the transmit frequency is slightly different to the axial coils to avoid any coupling effects between the two sets of coils.

After the NLJ data has been collected from both coils the AE data is collected. It follows the same process described above except no transmission occurs and the subsystem only listens for noise. This process of sampling and switching between the axial and vertical field producing coils and the NLJ and AE detection is further repeated a number of times (typically 480) until the object has passed through the coils and is well clear of the scan area. At this point the four lots of 480 datasets of 140 points of data are sent to the computer 1020 for analysis.

The NLJD data is processed as described in the first embodiment.

The AE data is processed slightly differently. The time series AE data is processed by simply determining if any of the time series data exceeds a threshold. If it has then potentially an IED has been detected.

The most reliable method for combining the results of the NLJ and AE data is by taking the logical OR of the final results. If the ORed result is positive, then the combined NLJD and AE subsystem has potentially detected an IED and an alarm is signalled.

Still further embodiments are similar to some previous embodiments except that they comprise at least one subsystem which can detect the presence of IED 's by sensing the presence of electronic circuits and metal objects. The large volume NLJD described previously is capable of detecting IED's by sensing the presence of electronic circuits by detecting NLJ's. Using retuning of the coils it is also capable of detecting metal objects. This design reduces the complexity of the subsystem and also reduces the footprint of the instrument. Reduction of the footprint can be important in some spatially constrained situations. Figure 28 shows the block diagram for combined NLJ and MD detector. To scan an object, it is placed onto the moving conveyor 2018 until it breaks the optical fence 2019. The PC Control 2020 then initiates the RF signal generator 2030 to begin generating RF pulses at the resonant frequency of the axial transmit coil 2110. These pulses are then filtered 2070, 2090; amplified 2080 and sent to the axial transmit coil 2110. While these pulses are being sent to the axial transmit coil, the axial receiver coil 2160 receives any induced RF signal at the second harmonic of the transmit frequency. This signal is then filtered 2120; transformed into quadrature signals 2120; demodulated 2050 from MHz signals down to kHz signals and then both quadrature signals are sampled at 0.25MHz 2040 to create digital signals and stored in the DSP's memory 2040. The sampling occurs for 560μs generating 140 points of data.

This process is repeated for the vertical field producing coils, with the main exception being that the transmit frequency is slightly different to the axial coils to avoid any coupling effects between the two sets of coils.

After the NLJ data has been collected from both coils, the metal detection data is collected. However, just prior to this data being collected the receiver coils are retuned to have the same frequency as the transmit coils by the Matching/Tuning Box 2150. Once this step has been completed the process of transmitting and receiving described above is repeated for both the horizontal and vertical coils. The entire process of sampling and switching between the axial and vertical field producing coils and the NLJ and MD detection is further repeated a number of times (typically 480) until the object has passed through the coils and is well clear of the scan area. At this point the four lots of 480 datasets of 140 points of data are sent to the computer 2020 for analysis.

The NLJD data is processed as described in the first embodiment.

The MD data is processed slightly differently. The data is then Fourier transformed into complex frequency data and then a background signal is subtracted off the data. This background data was collected and Fourier transformed while the object was not present and stored in the computer's memory. If the peak signal at the frequency which corresponds to the transmit frequency lies above a threshold for any of the 480 steps, then potentially an IED has been detected. The most reliable method for combining the results of the NLJ and MD data is by taking the logical OR of the final results. If the ORed result is positive, then the combined NLJD and AE subsystem has potentially detected an IED and an alarm is signalled. Further embodiments comprise at least one subsystem which can detect the presence of IED's by sensing the presence of electronic circuits and explosives. The large volume NLJD described previously is capable of detecting IED's by sensing the presence of electronic circuits by detecting NLJ's. Using retuning of the coils it is also capable of detecting explosives. This design reduces the complexity of the subsystem and also reduces the footprint of the instrument. Reduction of the footprint can be important in some spatially constrained situations.

Figure 29 shows the block diagram for combined NLJ and explosives detector. To scan an object it is placed onto the moving conveyor 3018 until it breaks the optical fence 3019. The PC Control 3020 then initiates the RF signal generator 3030 to begin generating RF pulses at the resonant frequency of the axial transmit coil 31 10. These pulses are then filtered 3070, 3090; amplified 3080 and sent to the axial transmit coil 3110. While these pulses are being sent to the axial transmit coil, the axial receiver coil 3160 receives any induced RF signal at the second harmonic of the transmit frequency. This signal is then filtered 3120; transformed into quadrature signals 3120; demodulated 3050 from MHz signals down to kHz signals and then both quadrature signals are sampled at 0.25MHz 3040 to create digital signals and stored in the DSP's memory 3040. The sampling occurs for 560μs generating 140 points of data. This process is repeated for the vertical field producing coils, with the main exception being that the transmit frequency is slightly different to the axial coils to avoid any coupling effects between the two sets of coils.

This process of transmitting on the vertical and axial coils is repeated a number of times until the conveyor belt is stopped to perform the quadrupole resonance measurements. The conveyor is preferably stopped when the object is substantially centered within the coils. The axial transmit coil is the only coil used to detect quadrupole resonance signals, the other coils are preferably switched to be open circuit and thereby have no effect upon the QR measurement. Just prior to this QR data being collected the axial transmit coil is preferably retuned by the Matching/Tuning Box 3100 to have the same frequency as the quadrupole resonant frequency at the temperature measured by the temperature probe 3125. The Matching/Tuning box 3100 also retunes the probe using the previously described retuning procedure in the first embodiment. Once this step has been completed the QR measurements can begin. The process of transmitting and receiving described above is repeated for the QR signals. The QR signals are stored within the computer for subsequent analysis.

Once the QR measurements have been completed the NLJ measurements continue until the object is well clear of the coils. The NLJ and QR data are processed as discussed in the first embodiment.

For explosive types such as RDX which produce a strong QR signal for a selected quantity, the conveyor belt does not need to be stopped for scanning. The procedure in the case where it is to be used with another similar technology such as NLJ, may interlace the two scan processes for NLJ and QR. Optionally, a single sequence can be constructed to deliver both a QR result and a NLJ result. Here, the coils would be multiply tuned for high signal to noise close to the fundamental for QR and multiple harmonics for NLJ.

Other coils that are described as "switched out", could be switched in parallel to create a more uniform magnetic field within the system, and a more uniform receive signal intensity within the scan volume. Optionally, the phases to the coils would be controlled to produce a particular polarisation to the RF field, such a circular polarisation. The degree of signal degradation caused by a less favourable tuning point maybe more acceptable than the increase in machine complexity/cost to including tuning circuitry and algorithmic procedures. It is possible to find a satisfactory tuning point in most cases where the object has low coupling to the QR coil. This is the case for a class of objects transported as organic cargo. The most reliable method for combining the results of the NLJ and QR data is by taking the logical OR of the final results. If the ORed result is positive, then the combined NLJD and QR subsystem has potentially detected an IED and an alarm is signalled.

Further embodiments comprise at least one subsystem which can detect the presence of IED' s by sensing the presence of electronic circuits, metal objects and explosives. The large volume NLJD described previously is capable of detecting IED's by sensing the presence of electronic circuits by detecting NLJ's. By retuning the coils it is also possible to detect metal objects and explosives. This design reduces the complexity of the device and also reduces the footprint of the instrument. Reduction of footprint can be important in some spatially constrained situations.

Figure 30 shows the block diagram for combined NLJ, metal and explosives detector. To scan an object it is placed onto the moving conveyor 3018 until it breaks the optical fence 3019. The PC Control 3020 then initiates the RF signal generator 3030 to begin generating RF pulses at the resonant frequency of the axial transmit coil 3110. These pulses are then filtered 3070, 3090; amplified 3080 and sent to the axial transmit coil 31 10. While these pulses are being sent to the axial transmit coil, the axial receiver coil 3160 receives any induced RF signal at the second harmonic of the transmit frequency. This signal is then filtered 3120; transformed into quadrature signals 3120; demodulated 3050 from MHz signals down to kHz signals and then both quadrature signals are sampled at 0.25MHz 3040 to create digital signals and stored in the DSP's memory 3040. The sampling occurs for 560μs generating 140 points of data.

This process is repeated for the vertical field producing coils, with the main exception being that the transmit frequency is slightly different to the axial coils to avoid any coupling effects between the two sets of coils. After the NLJ data has been collected from both coils the metal detection data is collected. However, just prior to this data being collected the receiver coils are retuned to have the same frequency as the transmit coils by the Matching/Tuning Box 3150. Once this step has been completed the process of transmitting and receiving described above is repeated for both the horizontal and vertical coils. This process of transmitting on the vertical and axial coils for both NLJD and the metal detector is repeated a number of times until the conveyor belt is stopped to perform the quadrupole resonance measurements. The conveyor is stopped when the object is substantially centred within the coils. The axial transmit is the only coil used to detect quadrupole resonance signals, the other coils are switched to be open circuit and thereby have no effect upon the QR measurement. Just prior to this QR data being collected the axial transmit coil is retuned by the Matching/Tuning Box 3100 to have the same frequency as the quadrupole resonant frequency at the temperature measured by the temperature probe 3125. The Matching/Tuning box 3100 also retimes the probe using the previously described retuning procedure in the first embodiment. Once this step has been completed the QR measurements can begin. The process of transmitting and receiving described above is repeated for the QR signals. The QR signals are stored within the computer for subsequent analysis. Once the QR measurements have been completed the NLJ and MD measurements continue until the object is well clear of the coils. The NLJ, MD and QR data are processed as discussed in the first and seventh embodiments.

The most reliable method for combining the results of the NLJ, MD and QR data is by taking the logical OR of the final results. If the ORed result is positive, then the combined NLJD, MD and QR detector has potentially detected an IED and an alarm is signalled.

Further IED detection embodiments are substantially the same as some previously described embodiments, except that the target IED can be located in one, two or three dimensions. All previously described detectors can be configured to repetitively scan an object as it moves through the coil structures. This information can be useful for locating the object in at least one dimension, which is along the conveyor belt.

Furthermore, by using information from the aforementioned metal detector it is possible to further locate metal objects, hence potentially IED's, in three dimensions.

In Figure 31, a metallic object 412 is located approximately half way between the left hand coils 400,420,430 and the right hand coils 410,440,450, but slightly closer to the left hand coils. As the transmit coils 400,410 are identical as are the receive coils 420,430,440,450, then when the first transmit coil 400 generates a field which is received in coils 440,450, then the signal received by these right hand coils will be slightly less than the signal received by the left hand receiver coils 420,430 due to the object being closer to the left hand coils. Therefore if, as previously described, the received voltage signal is transformed into a frequency signal peak via the FFT, then the distance the metal object is across the tunnel can be calculated by:

Across Distance = Total Across Distance * Left Hand Frequency Peak /( Left Hand Frequency Peak + Right Hand Frequency Peak).

Figure 32 shows the signals received from the horizontal receiver coils in the example in Figure 31. As can be seen in Figure 32, the left top coil 420 receives a signal of 2.1 whereas the right top coil 440 receives a signal of 1.8. If the tunnel width is 600mm then the metallic object lies at distance across the tunnel of 323mm. The two lower coils receive much less signal and hence are not used in this calculation. Similarly the vertical height can be calculated by using the above formula and instead using the Tunnel Height, Top and Bottom frequency peaks. The third dimension distance (along the conveyor) can be calculated by determining in which step the peak intensity occurred at the object passed through the coils. In Figure 32, this occurs at step 8. If the belt speed causes the object to travel 1500mm during the 17 steps, at step 8 it will have travelled 8/17* 1500mm = 706mm. Hence, the three dimensional co-ordinates of the objects have been determined. A representation of the metal object is then displayed in a three dimensional plot on the computer monitor. This helps the operator by pointing out the location of the metal object and hence potential IED.

The embodiments where both the NLJD and the AED are implemented, it has been discovered that there are limits as to the frequency that can be used. It becomes increasingly difficult to transmit enough power into the object at low frequencies to illicit a response from a NLJ. By experiment, the lowest preferable limit would lie at

1 OkHz. The upper limit is governed by the penetrability of radio waves into certain materials. Near IGHz water and plastics begin to absorb these frequencies and make detection impossible. Hence preferably the upper limit would lie near 100MHz, to ensure maximum penetrability. Even more preferable would be to operate these systems between 10OkHz and 16MHz as previously described.

Many AE objects emit frequencies that are broad band in nature, however the strongest signals occur between 100kHz and 16MHz. In particular, electronic clocks or timers typically have operating frequencies in low MHz range. Both systems are susceptible to external interferences and one of the main interferences that can occur is the AM radio band. This occurs from approximately 500 kHz to 1.6MHz. Even though the system is surrounded by a shield, this AM noise can still be picked up by cables in the system. To overcome this noise a fully enclosed shield could be used or use a band stop filter or other signal processing techniques to filter out the noise generated by particular radio stations.

For the MD, preferably the operating frequency lies between 1 Hz and 100 kHz to ensure ultra low coupling to the object being scanned. Even more preferable would be to operate in the range 1 Hz to 10 kHz as it is possible to determine some information about the type of metal target present in this frequency range. In particular aluminium foil responds differently to solid aluminium in this range. For the QR system the frequencies are fixed and hence cannot be varied. For explosives and many other substances these frequencies typically lie between 500 kHz to 6MHz.

Hence, some embodiments are substantially the same as any of the previous embodiments except that the frequency selected for irradiating or receiving from the object lie within specific frequency ranges. In Figure 33, an object is moved into an NLJD 800 and the transmit frequency is set to lie within the optimal 10 kHz to 100MHz range. The object is then moved by the conveyor 850 into an AED 810 (if present) within which the tuned receiver frequencies lie within a 1 kHz to 100MHz range. After it has been scanned by the AED 810 it is then moved into the MD 820 (if present) where it is scanned in the 1 Hz to 100kHz range. After being scanned by the MD 820 it is then moved by the conveyor into QR detector 830 (if present) where it is scanned for explosives.

The NLJD and AED subsystems may use scanning in the 100kHz-16MHz range and the MD may use scanning in the 1 Hz to 10 kHz range.

In further embodiments, the results of each detector are combined to reduce the false alarm rate of the IED detector. As each subsystem within the IED detector is detecting a different component of an IED, the results from one subsystem can be used as a confirmation sensor for another. For instance, if only the AED produces a positive result but none of the other subsystems produces a positive result then the alarm generated by the AED subsystem is ignored and the object is passed through the system as being clear of IED's. If on the other hand the QR system generates an alarm and at least one other subsystem produces an alarm then the system flags the object as potentially containing an IED.

Figure 34, the method used in this embodiment is shown. In the first instance the object is passed through the various subsystems present in the IED detector 900. Data is collected from each subsystem 910 and processed within a computer and then the results 920 are gathered together. If only one subsystem generates an alarm then the object is passed as clear 940. If however there have been two or more alarms generated then the object is passed as not being clear of IED's 960.

In further embodiments, the results of each detection method derived by scanning the object are used to locate the object in at least one dimension and the positional correlation between detection methods is used to enhance the detection rate or lower the false positive rate.

Referring back to Figure 32, this graph showed the signal on the horizontal coils as a metal object passed through the horizontal field oriented coils of the metal detector. If another detection method, such as an NLJD, performed measurements of the object at the same time interval as the metal detector and produced a similar profile then the two different detection methods will have produced a correlated result. Both detection methods will have detected a potential threat and detected it at the same location, meaning that is highly likely that the object is a threat. Conversely, if the profiles of the signals were indicating differing locations then the object may be passed as clear.

Another example would be the computed location was be found to be not realistic (i.e. outside the volume of the object) and therefore the positive result would be disregarded or used to lower mathematically to reduce signal intensity before the threshold operation.

Preferably information derived of the position of the object and/or threat would be displayed to the user to aid to locating the threat within the object. Further embodiments may be substantially the same as any of the previously described embodiments, with the exception that the object would be placed into at least one subsystem containing at least one of the described detection technologies, but not moved by a conveyor. This embodiment has the advantage of a reduction in complexity of the system and in general size. The object may be placed into each subsystem with at least one permanently open end. The process of scanning in this case for example may be initiated by a delayed timer after an optical beam is crossed by the movement into the scan volume. Equally as effective, the user may initiate the scan process by pushing a button when the object is in the correct position.

The object may be placed into a system whose open ends can be partially or completely closed with doors, which includes hatches, curtains etc. The electromagnetic isolating doors have the significant advantage of reducing the degree of external electrical interference that would induce spurious signal on the detectors. The doors also could have the advantage of reducing any electromagnetic emissions from the detectors of the system to satisfy emissions standards. The doors also could be used to help contain any blast or radiation released while the object or threat in the system.

In an example of such embodiments, in Figure 35, there are located three different subsystems 4000, 4100, 4200. The first subsystem 4000 detects IED's by sensing the presence of electronic circuits, the second 4100 by sensing the presence of metal objects and the third subsystem 4200 senses the presence of explosives. Hence, in the sixteenth embodiment the object is moved by hand, forklift or some other transportation method into the first subsystem 4000 where it is scanned only once. This is different from the situation where there was a conveyor present because many measurements were performed as the object moved through the electronics detection subsystem. After scanning has been completed, it is then moved into the second subsystem 4100 where it is scanned once and then moved into the third subsystem

4200. If a QR subsystem is used for this third subsystem then the object maybe scanned thousands of times as due to the weak QR signal many signals need to be accumulated to achieve enough signal to noise to detect an explosive.

At the end of this scanning, the results are combined together using a logical OR and if any of the subsystems generate a positive result then potentially a threat has been detected and an alarm is raised. Many different embodiments have been described herein, and these embodiments are not to be considered as exhaustive or exclusive. In practice, some modifications or enhancements of the described embodiments could be envisaged and adapted to work equally as well by any one skilled in the art. For example, the basic coil configurations described for the NLJD are meant to represent some useful arrangements and are not intended to be limiting of the invention. As will be apparent to those skilled in the art, a number of arrangements are possible from the described elements of each embodiment, which include the increase or decrease in the number of measurement axes for any of the devices shown.

It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the invention as shown in the specific embodiments without departing from the scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

Throughout the specification, unless the context requires otherwise, the word "comprise" or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated integer or group of integers but not the exclusion of any other integer or group of integers.