JP4098220B2 - Magnetic information processing method and apparatus using the same - Google Patents

Description

  The present invention relates to a magnetic information processing method for reading and analyzing magnetic information from a magnetic stripe such as a bankbook or a card, and a magnetic information processing apparatus using the method.

The present applicant has disclosed the following invention in Patent Document 1. FIG. 8 is a block diagram schematically showing the configuration of the magnetic information processing apparatus disclosed in Patent Document 1, and FIG. 9 is a flowchart of analysis processing by the apparatus.
JP 2002-56349 A

［ただし、Ｘ１，Ｘ２…Ｘｎはサンプリング番号、Ｙ１，Ｙ２…Ｙｎはサンプリング時点の電圧値。］を用いて、電圧値の±ピーク位置情報を作成するピーク情報作成処理、（Ｐ１）、ピーク情報作成処理で得られたピーク間の距離データを隣り合うピークのサンプリング番号の差から求めるピーク間距離データ作成処理（Ｐ２）、その処理により得られたピーク間距離データを磁気データの“０”“１”に対応するピーク間距離データの基準値と対比して、当該ピーク間距離データが“０”か“１”かを判定する２値判定処理（Ｐ３）、その他の処理（Ｐ４〜Ｐ６）を行うことを特徴とする磁気情報の解析方法である。 The invention described in Patent Document 1 is affected even if there is a slight level of fluctuation in the read magnetic information due to a decrease in magnetic force due to the passage of the medium usage period, the influence of an external magnetic field, the performance deterioration of the magnetic head, etc. The magnetic information can be reliably coded, and even when fluctuations occur in the recording period of the magnetic stripe, the fluctuations are automatically corrected and correct analysis can be performed. In order to achieve the purpose of providing an analysis method capable of dealing with the recording density, the magnetic information processing is divided into (a) reading processing and (b) analysis processing. The magnetic information recorded in the magnetic stripe is read by the magnetic head, and the amplified analog signal is given a predetermined conversion period (sampling period) by the AD (analog / digital) converter. AD conversion is performed, and AD conversion data including the sampling number and the voltage value obtained by the AD conversion is stored in the storage unit. In the analysis process (b), the following processes, that is, FIG. As shown, the following equation 8 is applied to the AD conversion data stored in the storage unit.

[Where X1, X2... Xn are sampling numbers, and Y1, Y2... Yn are voltage values at the time of sampling. ], The peak information creation process for creating ± peak position information of the voltage value, (P1), the distance between the peaks obtained from the difference between the sampling numbers of the adjacent peaks, the distance data between the peaks obtained by the peak information creation process The distance data creation process (P2) is compared with the reference value of the distance data between peaks corresponding to the magnetic data “0” and “1”. This is a magnetic information analysis method characterized by performing a binary determination process (P3) for determining whether it is “0” or “1” and other processes (P4 to P6).

  However, in the magnetic information analysis method (hereinafter referred to as the prior art) according to Patent Document 1, as shown in FIG. 8, an analog signal output from the magnetic head 1 is amplified by an amplifier 2 as shown in FIG. All of the data (AD conversion data) obtained by AD conversion by the conversion unit 3 is stored in the memory (RAM) 5 via the peak analysis unit 4, and in the step of peak information creation processing (P1) in the analysis processing, The AD conversion data is calculated in software by the CPU 7 using the above equation 8 stored in the ROM 6. First, the accuracy of the required peak position is not necessarily high enough. Thirdly, there is a problem that the memory capacity is required, and thirdly, the calculation takes time and the output of the data determination result becomes slow.

  Therefore, the present invention has been made to solve the above-mentioned problems, and its problem is to improve the peak information creation processing (P1) part of FIG. It is an object of the present invention to provide an improved magnetic information processing method and a magnetic information processing apparatus using the same method, which can reduce the calculation time and the calculation time.

Ｘaveを波形のピーク位置、Ｙaveを波形のピーク値に比例した値とみなすことにより、前記ピーク情報を取得することを特徴としている。 In order to solve the above problems, the present invention provides a read process for AD-converting a reproduction signal output from a magnetic head at a predetermined sampling period and sequentially outputting AD-converted data consisting of a sampling number and a voltage value, and at least the read Peak information generation processing that generates AD information obtained by processing using predetermined formulas to generate peak information that includes the peak position of the voltage value and the peak value, and peaks from the difference in sampling numbers in the generated peak information Magnetic information for performing inter-peak distance data generation processing for generating inter-peak distance data, and analysis processing including determination processing for comparing the generated inter-peak distance data with a reference value to determine “0” or “1” In the processing method, in order to perform the peak information generation process, the sampling number is an X coordinate value, the voltage value is a Y coordinate value, Define the weight W (Y) corresponding to the coordinate values, a weighted average Yave weighted average value Xave and Y coordinates of the X-coordinate obtained by calculation of the formulas 1 and 2,

The peak information is acquired by regarding Xave as a peak position of the waveform and Yave as a value proportional to the peak value of the waveform.

Ｘ座標の重み付き平均値及びＹ座標の重み付き平均値を求めることを特徴としている。 In the magnetic information processing method according to claim 1, an arithmetic table for obtaining W (Y) and W (Y) · Y with respect to a Y coordinate of a sample is defined, and each time an AD conversion value is input, Calculate equations 3, 4, and 5 using the operation table,
SW ← SW + W (Y) ……………… Formula 3
SWX ← SWX + SW ……………… Formula 4
SWY ← SWY + W (Y) · Y ……………… Formula 5
By calculating Equations 6 and 7 when the waveform leaves the effective area,

It is characterized in that a weighted average value of the X coordinate and a weighted average value of the Y coordinate are obtained.

  In the magnetic information processing method according to the second aspect of the present invention, SW, SWX, and SWY are integrated in hardware.

  In the magnetic information processing method according to claim 1 or 2, the weight {W (Y)} corresponding to each Y coordinate value is a value equal to or larger than the square of the Y coordinate value.

式７を演算してその演算結果Ｖｐｋをピーク値情報として出力する第５演算部と、

を備えたことを特徴としている。 A magnetic information processing apparatus according to the present invention that uses the magnetic information processing method according to claim 2 includes a conversion data accumulation unit including an effective area determination unit and a calculation unit configured by hardware, and the effective area determination unit includes: Each time the AD conversion data is input, it is determined whether the AD conversion data belongs to an effective area that exceeds a noise cut area, and AD conversion data that has entered the effective area is output to the arithmetic unit, The calculation unit calculates a formula (3) to calculate the weight sum SW, a second calculation unit to calculate the formula (4) and multiply the Y coordinate value by the weight, and a second calculation unit And a third calculation unit for obtaining a sum SWX of values obtained by multiplying the X coordinate value by the weight using Equation 5;
SW ← SW + W (Y) ……………… Formula 3
SWY ← SWY + W (Y) · Y ……………… Formula 4
SWX ← SWX + SW ……………… Formula 5
When the area determination unit determines that the AD conversion data has passed through the valid area, the area determination unit includes a software interface unit that passes each calculation result of the calculation unit to software. A fourth calculation unit that outputs the result PK as peak position information;

A fifth calculation unit that calculates Equation 7 and outputs the calculation result Vpk as peak value information;

It is characterized by having.

  According to the method invention of claim 1, in order to perform the peak information generation processing, the weight W (Y) corresponding to the Y coordinate value is defined with the sampling number as the X coordinate value and the voltage value as the Y coordinate value. , X-weighted average value Xave and Y-coordinate weighted average value Yave are obtained by calculation of Equations 1 and 2, and Xave is regarded as a waveform peak position and Yave is regarded as a value proportional to the waveform peak value. Thus, since the peak information is acquired, the accuracy of the required peak position is significantly improved as compared with the conventional case.

  According to the method invention of claim 2, equations 3, 4, and 5 are calculated using the calculation table every time an AD conversion value is input, and equations 6 and 7 are calculated when the waveform leaves the effective area. Thus, the weighted average value of the X coordinate and the weighted average value of the Y coordinate are obtained, so that processing that takes time such as multiplication / division is performed on the portion that is calculated each time AD conversion is performed, that is, the portion that is repeatedly calculated Since the processing is simple and only the reference and addition of the operation table are performed, a great effect of shortening the operation processing time can be obtained.

  According to the invention of claim 3, by integrating SW, SWX, and SWY in hardware, it is possible to reduce the memory consumption and further improve the processing speed.

  According to the invention of claim 4, in the magnetic information processing method of claim 1 or 2, since a value equal to or greater than the square of the Y coordinate value is used as the weight {W (Y)} corresponding to each Y coordinate value, The accuracy of the peak position is improved, and the influence of noise can be avoided reliably.

  According to the fifth aspect of the present invention, SW, SWY, SWX is obtained every time valid AD conversion data is input, and when the calculation for one waveform is completed, the data is transferred to the memory. Since it is not necessary to store all AD conversion data as in the prior art, it is possible to reduce the memory capacity. In addition, since peak information is obtained by weighted average of AD conversion data, the peak value of an analog waveform can be obtained accurately, and even if minute noise enters a position outside the waveform in the effective area, the noise is peaked. Judging by value is avoided. In addition, the amount of calculation for obtaining SW, SWY, SWX from AD conversion data is enormous, and a simple calculation is executed by a calculation unit having a hardware configuration capable of high-speed processing, and the peak information is obtained using SW, SWY, SWX data. Since only this is executed by software, a significant effect of shortening the calculation time can be obtained.

Next, the best mode for carrying out the present invention will be described with reference to the drawings. As the first embodiment, all of the calculations of Formulas 1 and 2 can be processed in software by the CPU. However, in the preferred second embodiment, the operations of equations 3, 4, and 5 are processed in hardware, and equations 6 and 7 are processed in software. Hereinafter, an example of the second embodiment will be described.
FIG. 1 is a block diagram showing a configuration of an apparatus using a magnetic information processing method according to the present invention, and FIG. 2 is a waveform diagram schematically showing a relationship between an analog waveform inputted to an AD converter and AD conversion data corresponding thereto. FIG. 3 is a block diagram mainly showing the configuration of the conversion data integrating unit, FIG. 4 is an explanatory diagram for explaining the technical idea of the present invention for obtaining a peak point by weighted average for each waveform, and FIG. 5 is weighted. FIG. 6 is a schematic diagram illustrating an example of the contents of the calculation table stored in the calculation table, and FIG. 7 is a flowchart illustrating the operation of the conversion data integrating unit. is there.

  As shown in FIG. 1, the magnetic information processing apparatus reads magnetic information from a magnetic stripe of a medium such as a passbook or a card conveyed at a predetermined speed and converts it into an electric signal, and the magnetic head 1 An amplifier 2 that amplifies an analog signal to be output, an AD conversion unit 3 that performs AD conversion on the amplifier output, a ROM 6 that stores various programs P and a calculation table T described later, and an AD output from the AD conversion unit 3 A conversion data integration unit 4A that performs integration processing as a preparation for inputting conversion data and obtaining peak information using an operation table, a RAM 5 that stores data output from the conversion data integration unit 4A, and an AD conversion unit 3 The CPU 7A is connected to the conversion data integration unit 4A, the RAM 5 and the ROM 6 and controls them, and the data output from the conversion data integration unit 4A It has a DMA transfer processing unit 8 that stores data in RAM using the MA (direct memory access) method, and a peak analysis unit 4B that is built in the CPU 7A and performs peak analysis in software using the data transferred to the RAM. ing.

  The programs stored in the ROM 6 include a first program for setting an AD conversion cycle, a second program for reading magnetic information, a third program for analysis, and other system programs. Calculation tables added to the ROM 6 as a new configuration for carrying out the analysis method of the present invention include a calculation table T1 and a calculation table T2, which will be described later. Further, the working data is stored in the RAM 5, but the calculation result obtained by the conversion data integrating unit 4A as described later is stored directly in the RAM 5 via the DMA transfer processing unit 8 independently of the operation of the CPU. It is configured to be.

  The AD conversion unit 3 is a known device called a successive approximation AD converter that is actually built in the CPU 7A, and converts analog data input from the amplifier 2 into digital data in response to an AD conversion activation trigger. Is. In the present invention, when the AD converter 3 inputs an AD conversion start trigger at a constant time interval, the absolute value (voltage value) of the AD converted digital data from the center point (0 V) of the AD conversion range is expressed as the Y coordinate. The number of times of AD conversion (sampling number) counted in the reverse direction from the time when the waveform leaves the effective area is output as an X coordinate to the subsequent conversion data integration unit 4A.

  As a result, as shown in FIG. 2, the amplified waveform (a) of the magnetic head output inputted from the amplifier 2 is shown as an image with dots at each sample time as shown in FIG. The converted AD data is output in time series and input to the converted data integrating unit 4A. FIG. 2A shows an analog waveform output from the amplifier 2, and a dotted line in FIG. 2B schematically shows AD conversion data when sampled at a predetermined cycle. Xi and Yi of each point output in the range (effective area) exceeding the noise level L1 in FIG. 2B are AD conversion data (sampling number and voltage value). In FIG. 2, Xa1 and Ya1; Xa2, Ya2;... Xan, Yan; Xb1, Yb1 to Xbn, Xb2, Yb2 to Xbn, Ybn.

The conversion data integration unit 4A analyzes the AD conversion data input corresponding to the analog waveform from the AD conversion unit 3 and obtains peak information. Therefore, as shown in FIG. The calculation unit 42 includes a calculation unit 42 for calculating Expressions 3 to 5 using the calculation table T1 and the second calculation table T2. The calculation unit 42 calculates the first calculation unit 42A for calculating the expression 3, the second calculation unit 42B for calculating the expression 4, and the expression 5 to execute the calculation of the expressions 3 to 5, respectively. And a third arithmetic unit 42C for processing. These are all configured by hardware by a gate array FPGA.
SW ← SW + W (Y) ……………… Formula 3
SWY ← SWY + W (Y) · Y ……………… Formula 4
SWX ← SWX + SW ……………… Formula 5
In the above equation, SW is the sum of weights, SWY is the sum of values obtained by multiplying the Y coordinate values by weights, and SWX is the sum of values obtained by multiplying the X coordinate values by weights.

  In order to obtain peak information as accurate as possible by calculation, the present invention employs a method for obtaining peak information by weighted average of effective samples. More specifically, FIG. 4 is a graph conceptually showing a part of the analog waveform of the amplifier 2 and AD conversion data (sampling number and output value) obtained by the AD conversion. In FIG. 4, Vcent is the center level of the waveform, and the outside of the noise cut area between Vcp and Vcn is the effective area. In the sample in the effective area, the output value of the i-th (i = 1, 2,... N) sample from the sample immediately before leaving the effective area toward the old direction (left direction) is Vi, and each effective sample The distance from the Vcent of the waveform of Y is Yi (= | Vi−Vcent |), and the number of samples up to the sample immediately after leaving the effective area is Xi (= i). The degree of involvement (weight) with respect to is W. In this case, the larger the Y coordinate value is, the greater the weight for the peak information is, and W is determined as a function of Yi (ie, weight = W (Y)).

  When the peak information (peak position and peak value) is obtained by the weighted average, even if the analog waveform from the magnetic head is not symmetrical with respect to the peak position as illustrated in FIG. 5, it is limited to the actual peak position Vp. There is an advantage that a close point can be recognized as a peak position.

As shown in the flowchart of FIG. 7, the calculation of SW, SWX, SWY of the conversion data integration unit 4A when reading magnetic information from the magnetic head 1 is performed as follows. That is,
When the magnetic information reader is activated, the RAM 5 storing SW, SWX, and SWY is cleared in advance before AD conversion data is input (S1; S1 is a step number; the same applies hereinafter). After clearing, input of AD conversion data is awaited (S2). When AD conversion data is input, it is determined whether or not the Y coordinate value is within an effective area outside the noise cut area between Vcp and Vcn (S3). If it is determined that the area is within the valid area, the RAM for storing SW, SWX, and SWY is validated (S4), the arithmetic processings of Equation 1 and Equation 2 are executed in parallel, and Equation 3 is executed (S5).
That is, as shown in FIG.
(A) The Y coordinate value is obtained by the effective area determination unit 41.
(B) W (Y) and W (Y) · Y are obtained. That is, on the one hand, the weight {W (Y)} corresponding to the Y coordinate value is read from the first calculation table T1 and given to the first calculation unit 42A, and on the other hand, it corresponds to the Y coordinate value from the second calculation table T2. A value obtained by multiplying the weight by the Y coordinate value (W (Y) · Y) is read and provided to the second calculation unit 42B. Then, the first calculation unit 42A executes the calculation process of Expression 3, and adds the weight {W (Y)} read from the first calculation table T1 to the sum (SW) of only the weights, which is the sum { Let SW + W (Y)} be the sum of only new weights (SW). Further, the second calculation unit 42B executes the calculation process of Expression 4, and a value obtained by multiplying the weight read from the second calculation table T2 by the Y coordinate (W (Y) · Y) (Y coordinate value × weight). Is added to the total sum (SWY), and the sum {SWY + W (Y) · Y} is set as a new sum (SWY) of (Y coordinate value × weight). The third calculation unit 42C is given the calculation result SW of Expression 3, adds the calculation result SW to the total SWX of (X coordinate × weight) by the calculation process of Expression 5, and adds the sum (SWX + SW) to the new calculation result SW. The sum SWX of (X coordinate × weight) is assumed.

  The arithmetic processing by the above three formulas is executed every time AD conversion data is input to the conversion data integrating unit 4A, and the obtained arithmetic results SW, SWY, SWX are input to the DMA transfer processing unit 8 (S7). When the effective area determination unit 41 determines that the AD conversion value has left the effective area, the DMA activation signal g to be output is determined based on the input to the DMA transfer processing unit 8, and the data (SW, SWY, SWX) is transferred to the RAM 5 and stored therein. That is, according to the present invention, when data of one SW, SWX, and SWY is obtained for each waveform having one peak, it is stored in the memory (RAM 5) for the first time. The DMA transfer unit process 8 has a transfer counter (not shown), and increments the transfer counter each time data is transferred once. When SW, SWX, SWY are transferred to the RAM 5, the calculation contents of the first to third calculation units are cleared (S8), the process returns to the initial step of the conversion data integration unit 4A, and the next AD conversion data is input. (S2), data creation processing for the next waveform is performed.

Based on the transfer notification signal output to the CPU 7 when the DMA transfer processing unit 8 transfers SW, SWX, and SWY, the peak analysis unit 4B uses the SW and SWX previously stored in the RAM 5 to calculate Equation 6 using software. The peak position is obtained by arithmetic processing, and then the peak value is obtained by arithmetic processing of Equation 7 by software using SW and SWY previously stored in the RAM 5. In this way, peak information (peak position and peak value) corresponding to one waveform is obtained.

  Each time peak information is obtained, it is sequentially stored in the memory (RAM 5). The peak information is then used for the process (P2 in FIG. 9) of measuring the peak-to-peak time width and obtaining the peak-to-peak distance data.

  Each content of peak distance data creation processing (P2), binary determination processing (P3), start code determination processing (P4), code determination processing (P5), and read data check processing (P6) in FIG. 9 is an example. Since it is described in the above-mentioned patent document 1 and is not a main part of the present invention, description thereof is omitted.

1 is a block diagram showing a configuration of a magnetic information processing apparatus according to the present invention. The waveform diagram which shows typically the relationship between the analog waveform input into an AD conversion part, and AD conversion data corresponding to this. The block diagram which mainly shows the structure of the conversion data integration part. Explanatory drawing for demonstrating the technical idea of this invention which calculates | requires a peak point with a weighted average with respect to each waveform. Explanatory drawing explaining the advantage which calculates | requires peak information by a weighted average. The schematic diagram which shows an example of the memory content of the calculation table stored in ROM. The flowchart explaining the effect | action of a conversion data integration part. The block diagram which shows schematically the structure of the conventional magnetic information processing apparatus. The flowchart of the analysis process by the same apparatus.

Explanation of symbols

In FIG.
3 AD Conversion Unit 4A Conversion Data Accumulation Unit 41 Effective Area Determination Unit T1 First Calculation Table T2 Second Calculation Table 42 Calculation Unit 42A First Calculation Unit 42B Second Calculation Unit 42C Third Calculation Unit 5 DMA Transfer Processing Unit

Claims (5)

A read process in which a reproduction signal output from the magnetic head is AD-converted at a predetermined sampling period, and AD conversion data including a sampling number and a voltage value is sequentially output, and at least AD conversion data obtained by the read process is a predetermined formula Peak information generation processing that generates peak information including the peak position of the voltage value and the peak value calculated by calculating the peak value, and inter-peak distance data that generates peak-to-peak distance data from the difference in sampling number in the generated peak information In a magnetic information processing method for performing generation processing, analysis processing including determination processing for determining whether the generated peak-to-peak distance data is “0” or “1” by comparing with a reference value,
In order to perform the peak information generation process, the sampling number is defined as an X coordinate value, the voltage value is defined as a Y coordinate value, a weight W (Y) corresponding to the Y coordinate value is defined, and a weighted average value Xave of the X coordinate is defined. And the weighted average value Yave of the Y coordinates is obtained by the calculation of Equations 1 and 2.

A magnetic information processing method, wherein the peak information is obtained by regarding Xave as a peak position of the waveform and Yave as a value proportional to the peak value of the waveform. The magnetic information processing method of claim 1,
Define a calculation table for obtaining W (Y) and W (Y) · Y for the Y coordinate of the sample. Every time an AD conversion value is input, the expressions 3, 4, and 5 are calculated using the calculation table. ,
SW ← SW + W (Y) ……………… Formula 3
SWY ← SWY + W (Y) · Y ……………… Formula 4
SWX ← SWX + SW ……………… Formula 5
By calculating Equations 6 and 7 when the waveform leaves the effective area,

A magnetic information processing method characterized by obtaining a weighted average value of X coordinates and a weighted average value of Y coordinates.   3. The magnetic information processing method according to claim 2, wherein SW, SWX, and SWY are integrated by hardware.   3. The magnetic information processing method according to claim 1, wherein a value equal to or greater than the square of the Y coordinate value is used as the weight {W (Y)} corresponding to each Y coordinate value. In the magnetic information processing apparatus using the magnetic information processing method of claim 2,
A conversion data integration unit consisting of a hardware effective area determination unit and a calculation unit is provided,
Each time the AD conversion data is input, the effective area determination unit determines whether the AD conversion data belongs to an effective area that exceeds a noise cut area, and sets the AD conversion data that has entered the effective area to the calculation unit. Output to
The calculation unit calculates a weighting sum SW by calculating Equation 3;
SW ← SW + W (Y) ……………… Formula 3
A second calculation unit that calculates Formula 4 to obtain a sum SWY of values obtained by multiplying the Y coordinate value by the weight;
SWY ← SWY + W (Y) · Y ……………… Formula 4
A third arithmetic unit that obtains a sum SWX of values obtained by multiplying the X coordinate value by the weight using Equation 5;
SWX ← SWX + SW ……………… Formula 5
A software interface unit that passes each calculation result of the calculation unit to software when the area determination unit determines that the AD conversion data has passed through an effective area;
A fourth calculation unit configured by the software and calculating Equation 6 and outputting the calculation result PK as peak position information;

A fifth calculation unit that calculates Equation 7 and outputs the calculation result Vpk as peak value information;

A magnetic information processing apparatus comprising:

Images