ASYNCHRONOUS METHOD FOR SAMPLING SIGNALS IN METAL DETECTORS

Abstract

This invention is related to the method providing computation of the signal frequency components in an acceptable accuracy in contravention of the shifts in the phase and the magnitude information caused by asynchronous sampling of the signals in the process of asynchronous sampling of metal detectors wherein the received signal by the receiver unit (4) divided into time intervals, say timing values those are far shorter than the sampling period and correspond to nearest probable sampling of the ADC (6); providing the computation of the sine and cosine coefficients or exponents of time constant coefficients of the said timing value from previously located or dynamically generated coefficient table; resulting the elimination of the requirement of synchronous sampling and the requirement of the signal period is multiple of the sampling period.

Claims

1. A method for obtaining information related to a target presence and/or a target kind for metal detectors, the method comprising: sampling a signal by at least one ADC which is received through a receiver unit wherein said signal is generated as a result of a reaction of said target to a magnetic field transmitted by a transmitter unit; by a controller, utilizing a DFT based transformation technique consisting of summing of samples where each sample is multiplied by coefficients corresponding to a timing within a period in order to be expressed in terms of components of phase/magnitude or time constants for obtaining said information; by the controller, providing the transmitter unit to transmit a reference signal where said reference signal is divided into time intervals where said time intervals are indexed by reference timing values; by the controller, providing the receiver unit to receive and ADC to sample said reference signal; by the controller, detecting a phase shift for each sampling of ADC according to a reference timing value; by the controller, updating a transformation table, the transformation table defined for the timing intervals that are smaller than or equal to sampling intervals of the ADC and multiplying coefficients corresponding to a defined time with acquired samples, by the controller, providing transmitter unit to transmit a signal for detection, by the controller, providing receiver unit to receive and ADC to sample said signal, by the controller, defining nearest timing of each acquired sample from a table/array by using said the transformation table/array and determining corresponding coefficients for each sample; and by the controller, obtaining phase/magnitude and/or time constant components of targets by using the sum of the asynchronously acquired samples, each multiplied by the corresponding coefficient.

2. (canceled)

3. The method according to claim 1, wherein the said metal detector operates any of single frequency, multifrequency or time-domain methods.

4. (canceled)

5. The method according to the claim 1, wherein a table on an internal or external memory of the processor is used for locating the sine and cosine coefficients corresponding to each timing value or determined by calculation, dynamically.

6. The method according to the claim 5, wherein the time constants coefficients are located in the table in the memory when using time-domain, say pulse induction method.

7. The method according to the claim 1, wherein the computing, timing or signal conversion processes are accomplished by at least one microcontroller, at least a digital signal processor (DSP) and/or at least one FPGA.

Description

DESCRIPTION OF THE FIGURES

[0030] FIG. 1: Sample block diagram of a digital metal detector.

[0031] FIG. 2: The graphical representation of the sample timing of the received signal. The (t.sub.n) timings point to the samples synchronously acquired whereas (t′.sub.n) point to the timing of asynchronous samples. The numbers 17, 18, 19, 20 . . . 36 depict the counter values of the timing.

DESCRIPTION OF REFERENCE NUMBERS

[0032]

TABLE-US-00001 NO PART NAME 1 Transmitter Coil 2 Receiver Coil 3 Transmitter Unit 4 Receiver Unit 5 Switching Component 6 ADC 7 Controller

DESCRIPTION OF THE INVENTION

[0033] This invention is related to the metal detector containing transmitter unit (3) generating a single frequency, multifrequency of pulsed magnetic field and providing the electrical reception of the magnetic field by the receiver unit (4) which is created at the target as a result of the field time-varying magnetic field generated by the transmitter unit (3) wherein; the signal received by the receiver unit (4) is sampled asynchronously relative to the transmitter unit by at least one ADC (6) (Analog to Digital Converter) and analyzed by the vectors of the sinusoidal components are obtained in this way.

[0034] Within the metal detector which the sampling is accomplished asynchronously; the switched signal generated by the controller (7) containing the multifrequency components is applied to the switching component (5). The voltage switched by the switching component (5) is applied to the transmitter unit (3) and a magnetic field is generated at the transmit coil (1) or around the transmit coil (1) which is proportional to the electric current flow. The magnetic field creates a current flow within the target and as a result, the current flow at the target generates a magnetic field and induces a voltage on the receiver coil (2) to be transferred to the ADC (6) through the receiver unit (4). The in-phase and quadrature components of the signal that is sampled asynchronously by the ADC (6) are computed by the controller (7). Same process is valid for the time-domain metal detectors such as pulse induction detectors were the phase components replaced with the exponential time constants.

[0035] The period of the sampled signal is not the multiple of the sampling period utilizing asynchronous digitization by the ADC (6). Accordingly, it is not predictable to determine the point that ADC (6) will sample the signal. Therefore, it is not useful that a predetermined sine and cosine table located in the controller (7) memory. Conversely, the greatest advantage of the asynchronous configuration is providing an opportunity to use a faster and high resolution ADC (6) at a bearable cost. Albeit, it is required to predetermine the time intervals between the asynchronous sampling and that of synchronized version in order to the asynchronous configuration can be applied (in order to be applied of the asynchronous configuration).

[0036] This invention provides the compensation of the probable shifts in measurements resulting from asynchronous sampling by means of the predetermination of the time intervals between asynchronous timing of the ADC (6) and that of synchronous version.

[0037] In the asynchronous signal sampling method of the invention; the controller (7) divides the multifrequency signal period into equal time intervals, based on the required resolution. The period of each said interval has a far less duration than sampling period and a digital record is located for each interval. The timing values can be tracked by modules such as a timer or a counter. A table is constructed as each location holds the sine and cosine coefficients of the corresponding time interval. The steps of the table correspond to the phase (angle) of corresponding time interval. The content of the table is depended to the frequency to be analyzed, therefore the table content is required to be constructed based on that frequency. The tables can be priorly constructed in case the frequencies are also predefined. The major difference to the prior art is, the tables are not defined corresponding to the sampling times but rather defined to probable time intervals that the sampling to be done.

[0038] If the sampling frequency is fs and the time gap is defined as Δt;

Δt=1/fs

[0039] Above expression defines the time gap as the error caused by the sampling frequency. Same equation can be used to express the timing error as the “accurate timing” concept in case of the ADC sampling speed increases. Within this context, in the following expression;

Δt′=1/(Mfs)

[0040] “M” stands for the interval timing precision of the samples acquired by the ADC (6) where M>1. The said timing values are used to track the phase change of the asynchronously sampled signal. Resultingly, a M*N size table that containing the sine and cosine coefficients, is constructed, where N is the total number of samples within a signal period. The timing value can be used to index the sine and cosine coefficients from the table in ratio of the value at the end of the signal period. Finally, the vectoral components are obtained by multiplying the sine and cosine coefficients by the corresponding sampled values. Within this context; the sensitivity of the result of calculations increases with the resolution of sampling.

[0041] The said asynchronous sampling system described by FIG. 2; the ADC (6) samples the received signal at the instants of t′.sub.1, t′.sub.2, t′.sub.3, t′.sub.4, t′.sub.5, t′.sub.6, t′.sub.7 . . . with equal or unequal sampling intervals but non-coherently to the repeating signal. In the asynchronous sampling part of FIG. 2, the ADC (6) samples the 0 degree at t′.sub.1, 111 degree at t′.sub.2, 221 degree at t′.sub.3 and 332 degree at t′.sub.4 at the first period. In the second period, ADC (6) samples 83 degree at t′.sub.5, 193 degree at t′.sub.6 and 304 degree at t′.sub.7 . As seen in the example, the ADC (6) samples the signal asynchronously, by equal intervals but not corresponding to same degrees in following two periods. This case results measurement of the phase of the sampled signal different from the other in each signal period. In the method of the invention; the phase shift is tracked by the controller (7) and used for the computation. The controller (7) takes reference timing values while it applies the reference signal to the switching component (5) generated using the transmitter unit (3). In the method of invention; the signal is divided into smaller time intervals (Δt′). The said time intervals are depicted as 17, 18, . . . 36 in FIG. 2. The reference timing values indexes the time intervals as 0, 1, 2, . . . 19 and provides a calculation of the corresponding phase (angle) values. The timing values can be tracked by modules such as a timer or a counter. The phase (angle) value is calculated in each sampling of the ADC (6) according to the reference timing value. Thanks to the referenced timing; the synchronization shift of the ADC (6) sampling to the received signal is determined as well as the phase shifts because of the asynchronous sampling.

[0042] This invention is a method of; sampling of the signal by at least one ADC's (6) which received through a receiver unit (4), generated as a result of the reaction of the target to the magnetic field generated by a transmitter unit (3); and utilization of a DFT based transformation technique consisting of the summing of the samples each multiplied by the coefficients corresponding to the timing within the period in order to be expressed in terms of components of phase/magnitude or time constants in order to obtain the information for the target presence and/or the target kind for the metal detectors. In addition to the synchronous methods in the state of the art, the invention provides asynchronously sampling by following steps; [0043] Defining the nearest timing of the each acquired sample from the table/array by using the transformation table/array which is defined for the timing intervals that are smaller than or equal to the sampling intervals of the ADC (6) and multiplying the coefficients corresponding to the defined time with the acquired samples. [0044] Phase/magnitude and/or time constant components of targets are obtained in acceptable accuracy by using the sum of the asyncronously acquired samples, each multiplied by the corresponding coefficient.

[0045] In the preferred embodiment; the timing values are generated by internal counters within the processor. The size of the said sine and cosine coefficients table and the speed of the said counters (number of the timing values) varies with the required resolution and frequency spectrum to be analyzed.

[0046] In the preferred embodiment, the detector is a multifrequency VLF continuous wave (CW) detector while all described analogy can be applied to time-domain detectors operating by pulse induction method or any hybrid methods by time-domain analysis with replacement of sine and cosine coefficient tables to corresponding time domain analysis coefficients and expressing them as superposed series of exponentials of time constants.