WO2016174036A1

WO2016174036A1 - Sensor comprising capacitive-detection pixels

Info

Publication number: WO2016174036A1
Application number: PCT/EP2016/059292
Authority: WO
Inventors: Josep Segura Puchades
Original assignee: Commissariat A L'energie Atomique Et Aux Energies Alternatives; Safran
Priority date: 2015-04-28
Filing date: 2016-04-26
Publication date: 2016-11-03
Also published as: FR3035724A1

Abstract

Sensor (100) comprising capacitive-detection pixels (102) forming a matrix of N rows and M columns of pixels, each pixel comprising:- a detection capacitor (104); - a pre-charge transistor (106) linked to the detection capacitor and able to apply to the detection capacitor a pre-charge voltage issuing from a pre-charge row or column (108); - a read transistor (112) linked to the detection capacitor and able to carry out a transfer of electric charge stored in the detection capacitor to a read circuit (116); the sensor comprising N‑1 selection rows (126) distinct from the pre-charge row or column, each selection row being linked to the gates of the read transistors of one of the N rows of pixels and to the gates of the pre-charge transistors of one other of the N rows of pixels.

Description

SENSOR COMPRISING CAPACITIVE DETECTION PIXELS

DESCRIPTION

TECHNICAL FIELD AND PRIOR ART The invention relates to a sensor, or a detection device, comprising pixels with capacitive detection. The invention is advantageously applicable for producing a fingerprint sensor, or a palmprint sensor, or a touchpad.

In a sensor comprising pixels with capacitive detection, each pixel delivers, for a given bias voltage, an output signal corresponding to the electrical charges whose value Q is a function of the value C of a capacitance associated with the pixel and in which these charges are stored. The capacitance of each pixel is formed between a first electrode present in the pixel and a second electrode formed by a conductive element to be detected lying on the pixel, for example a part of the hand, a finger, etc. In the case of a fingerprint sensor, since the capacitance is defined between the first electrode and the second electrode formed by the portion of the finger on the pixel, the value of this capacitance varies according to whether this part of the finger is a crest or furrow of the fingerprint.

To determine the value C of the capacity of a pixel in a measure, it is customary to pre-load this capability with voltage V _pr preload eloads known, then read the Q. value of expenses stored in the capacity with a charge amplifier. The value C of the capacitance can then be determined from the equation C = Q / V.

EP 2 178 026 describes such a fingerprint sensor. Each pixel of this sensor comprises a first transistor whose gate is connected to a control line and which is used both to preload two capacitors of the pixel and then to read the electric charge resulting from the read operation. A first of the two capacitors is a fixed capacitance and the second capacitance is formed between a first electrode present in the pixel and a second electrode formed by the finger on the sensor. The first fixed value capacity serves as a reference when the value of the second capacitance is zero (no finger present on the pixel during the reading). Each pixel also has a second transistor connecting the capacitances to the ground to discharge them. This document also describes the possibility of adding, in each pixel, an additional transistor connected to an additional electrical line common to all the pixels of the same column, making it possible to separate the operations of pre-charging and reading.

In a TFT ("thin-film transistor") implementation of such a sensor, the space requirement due to the surface occupied by the transistors, the reference capacitance and the control lines becomes critical. In addition, the control of the pixel matrix is complex and requires dedicated control circuits for such use because commercial "standard" control circuits are designed for LCD applications where only one command per line is available.

The document EP 1 041 356 describes a fingerprint sensor comprising capacitive detection pixels arranged in a matrix. In this sensor, each control line is used to pre-load the pixel capacitances of a first row of the array and to read the charges stored in the pixel capacities of a second row of the array. The structure of the sensor described in this document has the advantage of "mutualizing" the command lines between several rows of pixels, which reduces the size and complexity of the sensor. On the other hand, the reset of each pixel of this sensor is done by means of a diode-mounted transistor, which generates an inaccurate potential on the node to which this transistor, the capacitance and the read transistor are connected. The value of this potential is closely related to the value of the control voltage which must be at least equal to the threshold voltage which makes it possible to turn on the diode-mounted transistor. However, given the possible variations of this threshold voltage due to technological dispersions, the value of this potential can have significant variations from one pixel to another. STATEMENT OF THE INVENTION

An object of the present invention is to provide a sensor with capacitive detection pixels minimizing or eliminating the disadvantages of sensors of the prior art, compact structure and simple to control even when performing sensor TFT technology.

For this, the invention proposes a sensor comprising capacitively sensing pixels arranged forming a matrix of N rows of pixels and M columns of pixels, each pixel comprising at least:

- a detection capability;

a pre-charge transistor connected to the detection capacitor and able to apply on the detection capacitance a pre-charge voltage originating from at least one line or a pre-charge column of the sensor;

a read transistor connected to the detection capacitor and able to carry out a transfer of electric charges stored in the detection capacitor to at least one read circuit of the sensor;

the sensor further comprising at least Nl distinct selection lines of the line or the pre-charge column and such that each selection line is connected to the gates of the pixel reading transistors of one of the N lines of pixels and to the gates pre-charge transistors of the pixels of another of the N lines of pixels.

In this sensor, since each of the selection lines is connected to the gates of the pixel read transistors of a first of the N pixel lines and to the gates of the pre-charge transistors of the pixels of a second of the N lines of pixels, different from the first line of pixels, the voltage applied to each of the selection lines makes it possible at the same time to control the reading of the pixels of this first line and to carry out the pre-charge of the pixels of the second line and which are meant to be read after those in the first line. This configuration therefore makes it possible to pool the sensor control lines (which correspond to the selection lines) and thus to reduce the size of these control lines in the pixels, whatever the technology used for producing the sensor. In addition, because the pre-charge line or column over which the pre-charge voltage is intended to be applied is distinct from the selection lines, the value of this pre-charge voltage is therefore independent of that of the applied voltage as each of the selection lines. This pre-charge voltage can therefore vary according to the desired configuration of the sensor and can therefore be adjusted, for example to optimize or maximize the dynamic output voltages of pixel reading circuits, that is to say, to extend the range possible values of the output voltages of the pixel reading circuits, and / or to compensate for measurement variations between the pixels such as variations due to the non-flatness of a finger whose fingerprint is detected and / or to compensate for variations related to the technological dispersions of the sensor components from one pixel to another.

The sensor may further include a pre-charge circuit capable of varying the value of the pre-charge voltage on the pre-charge line or column. This pre-charge circuit corresponds for example to a digital-to-analog converter.

The sensor may further comprise:

- M read columns such that the reading transistors of the pixels of each of the M columns of pixels are connected to one of the M read columns;

M reading circuits each comprising at least one charge amplifier whose input is connected to one of the M read columns;

N pre-charge lines such that the pre-charge transistors of the pixels of each of the N rows of pixels are connected to one of the N pre-charge lines, or M pre-charge columns such as the pre-charge transistors. charge of the pixels of each of the M columns of pixels are connected to one of the M pre-charge columns.

The charge amplifier of each of the M read circuits may correspond to an operational amplifier.

The sensor may further comprise a control circuit coupled to the precharge circuit and adapted to calculate a nominal value V _pr echarge_nom tension pre-load such that a maximum value V _or t_maxjnit among NxM output voltages Voutjj, with i between 1 and N and j between 1 and M, intended to be delivered by the charge amplifiers of the reading circuits for each of the pixels is substantially equal to, or close to, a saturation voltage of the charge amplifiers of the read circuits.

The calculation of this nominal value of the pre-charge voltage makes it possible to maximize the dynamics of the output voltages of the reading circuits by judicious choice of the value of the pre-charge voltage applied to the pre-charge lines or columns. of the sensor, this choice being made such that the maximum value among the output voltages of the reading circuits is substantially equal to, or close to, the saturation voltage (high or low according to the sign of the input signal) of the amplifiers. read circuit charges (the range of values of the output voltage of a charge amplifier extends between the value of a low saturation voltage, generally referred to as -Vsat, and the value of a high saturation voltage , usually called + Vs _a t).

Thus, for a given range of values of the detection capabilities, the range of values of the output voltages of the read circuits for these values of the detection capabilities is extended. For example, for two different values of the detection capacitors, a larger difference between the values of the corresponding output voltages is obtained by having previously determined the nominal value of the pre-charge voltage. The use of this nominal value of the pre-charge voltage thus makes it possible to better distinguish the values of the output voltages of the read circuits from each other for given values different from the detection capacitors.

In this case, the control circuit may be adapted to perform the calculation of the nominal value V _pr echarge_nom of the precharge voltage by implementing the steps of:

- applying an initial pre-charge voltage V _pr echargejnit on lines or pre-load column;

reading of the output voltages V ₀ ut_u in the presence of a reference element on the pixels of the sensor;

calculating the maximum value V _or t_maxjnit among the values of the output voltages V _or tjj obtained during the previous reading; - adjusting the value of the voltage pre-load to the nominal value V _pr echarge_nom as V _or t_maxjnit is substantially equal to the saturation voltage of the read circuits of the charge amplifiers.

The nominal value V _pr echarge_nom of the precharge voltage can be

+ γ

calculated as V _{precharge name} -V _{precharge init} -.

out max init

The value of the initial pre-charge voltage V _pr echargejnit can be chosen arbitrarily.

The reference element may correspond to a finger model, a metal plate, or any element able to maximize the values of the detection capabilities.

For the calculation of the nominal value V _pr echarge_nom of the precharge voltage, the control circuit may comprise:

means for applying an initial pre-charge voltage Vprechargejnit to the pre-charge lines or columns;

means for calculating the maximum value V _or t_maxjnit among the values of the output voltages Voutjj;

- means for adjusting the value of the voltage pre-load to the nominal value V _pr echarge_nom as V _or t_maxjnit is substantially equal to the saturation voltage of the charge amplifiers of the read circuits.

The control circuit may comprise means for calculating the nominal value V _pr echarge_nom of the voltage pre-charge such as

+ Sat

prech arg e _ name = V p_rech arg e _ init

out _ max_init

The control circuit may be able to control the precharging circuit such that N x M pre-charge voltages V _pre- charge_u distinct are successively applied to the rows or columns of pre-charge during a reading of the pixels. In this configuration, the value of the pre-charge voltage can be adapted and optimized independently for each of the pixels. Each pixel may furthermore comprise a reference capacity coupled to the detection capacitance, to the pre-charge transistor and to the reading transistor of the pixel, and the control circuit may be able to perform, after the calculation of the nominal value V _P recharge_name of the pre-charge voltage, an empty calibration of the sensor by calculating the values of the N x M pre-charge voltages Vp cnargejj via the implementation of the steps of:

- application of the pre-charge voltage of nominal value V _P recharge_name on pre-charge lines or columns;

reading of the output voltages V ₀ ut_u in the absence of element on the pixels of the sensor;

calculating a mean V _or t_moy or a minimum value V _or t_min name of the output voltages V ₀ ut_u obtained during the previous reading;

calculating a coefficient Ky associated with each pixel such that

Kij = Vout moy / VoutJJ OR such that Ky = Vout min noun / VoutJJ;

- calculation of N x M pre-load voltages V _pre loadjj such that V _pre loadjj

= Kij.V

With these steps, which correspond to a vacuum calibration of the sensor, the pre-charge voltage applied for each of the pixels is adapted in order to compensate for any measurement differences due to the technological dispersions of the pixel elements, making it possible in particular to compensate for uniformity of the measurements due to the technological dispersions of the reference capacities of the pixels. When the minimum value V _or t_min_nom is taken into account for the calculation of the coefficients Ky, this vacuum calibration of the sensor also makes it possible to increase the dynamics of the output voltages of the reading circuits.

Several variants of implementation of the vacuum calibration of the sensor are possible. For example, instead of calculating a coefficient Ky associated with each pixel, it is possible to calculate N coefficients K, each associated with one of the N rows of pixels and to calculate from average values of the output voltages of each pixel line. In this case, the pre-charge voltages calculated from these coefficients can be similar for all the pixels belonging to the same line. A similar variant can be implemented by considering each of the columns of pixels. For the calculation of the values of the N x M preload voltages Vprecharge, the control circuit may comprise:

- means for applying the voltage pre-charge nominal value V _pr echarge_nom on lines or pre-load column;

means for calculating a mean V _or t_moy or a minimum value V _or t min for the output voltages V _or tjj;

means for calculating a coefficient Ky associated with each pixel such that

Kij = Vout moy / VoutJJ OR such that Ky = Vout min noun / VoutJJ;

means for calculating the N x M pre-charge voltages Vprecharge, such as VprechargeJJ-Ky .Vprecharge nom-

It is possible that the vacuum calibration operation of the sensor is implemented without the nominal value of the pre-charge voltage making it possible to maximize the dynamics of the output voltages of the amplifiers has been calculated beforehand. In this case, each pixel may furthermore comprise a reference capacitance coupled to the detection capacitance, to the pre-charge transistor and to the reading transistor of the pixel, and the control circuit may be able to control the pre-charge circuit. load as NxM preload V _pr echargejj separate voltages are successively applied on lines or pre-load column during a read of the pixels, and be capable of performing a vacuum calibration of the sensor by calculating the values N x M preload voltages Vprechargejj via the implementation of the steps of:

reading the output voltages V _or tjj in the absence of an element on the pixels of the sensor;

calculating a mean V _or t_moy or a minimum value V _or t_minjnit of the output voltages V _or tjj obtained during the previous reading;

calculating a coefficient Ky associated with each pixel such that Ky = V ₀ ut_moy / Voutjj or such that Ky = V _or t_minjnit / Voutjj;

calculation of N x M pre-load voltages Vprechargejj such that VprechargeJJ ⁼ Ky. Vprechargejnit- The control circuit can be adapted to perform, after the sensor empty calibration, an equalization operation in reading the sensor by implementing the steps of:

- reading of the output voltages V _or t_u for all the pixels by successively applying on pre-charge lines or columns the pre-charge voltages

VprechargeJJ, ^'

calculating a mean V _or t_mi of the output voltages V _or tjj for

M

each of the N lig ones of p r ~ ixels such that V-xt, _mi. = - / V out, ι. _j. ; '

M _{J =} i

calculating a maximum value V _or t_mi_max or an average V _or t_mi_moy of the values V _or t_mi;

calculating a coefficient K, for each of the N rows of pixels such that K, ⁼ Vout mi max / v out mi or such that K, = V _or t_mix av / v out mi,

- adjustment of the precharge voltages V _pr echargejj by multiplying the values of the pre-charge voltages Vprechargejj of each of the N rows of pixels by the coefficient K, associated with each of the N pixel lines.

This operation of equalizing the reading of the sensor makes it possible to adjust the pre-charge voltages applied for each of the pixels in order to compensate any differences in measurements obtained from one pixel line to the other, for example due to the irregularities of pressure between the different pixel lines of the sensor during a measurement.

To perform the equalization operation in reading of the sensor, the control circuit may comprise:

means that successively apply the pre-charge voltages Vprechargejj to the pre-charge lines or columns;

means for calculating a mean V _or t_mi of the output voltages

M

Voutjj for each of the N rows of pixels such that V _{out ni} = -ΣV _{0Ut j} ;

M _{j =} i

means for calculating a maximum value V _or t_mi_max or an average V ₀ ut_mi_mo _y of values V _or t_mi; means for calculating a coefficient K, for each of the N rows of pixels such that K, = V _or t_max max / V ₀ ut_mi or such that Ki = V _or t_mix av / v out mi,

means for adjusting the pre-charge voltages Vp chargejj multiplying the values of the pre-charge voltages Vp chargejj of each of the N lines of pixels by that of the coefficient K, associated with each of the N lines of pixels.

As a variant, the reading equalization of the sensor can be implemented by adjusting the pre-charge voltages applied for each of the pixels in order to compensate for possible differences in measurements obtained from one column of pixels to another, for example due to pressure irregularities between the different pixel columns of the sensor during a measurement. In this case, the steps described above are adapted such that the average of the output voltages for each of the M columns of pixels is calculated, then adjustment coefficients are calculated for each of the M columns of pixels, and the voltages of pre-charge are adjusted from these adjustment factors.

Other variants of the sensor read equalization can be implemented, realizing for example an adjustment of pre-charge voltages not pixel by pixel, but by line of pixels or by column of pixels.

Each pixel may further comprise a reference capacitor coupled to the detection capacitance, the pre-charge transistor and the reading transistor of the pixel.

The pre-charge and pixel-reading transistors may advantageously be of the TFT type.

The various operations of adjusting the pre-charge voltage, notably making it possible to maximize the maximum values of the output voltages of the amplifiers and / or a sensor empty calibration and / or a sensor read equalization, may be implemented by a sensor having no Nl distinct selection lines of the line or the pre-charge column and such that each selection line is connected to the gates of the pixel reading transistors of one of the N lines of pixels and to the gates of the pixel pre-charge transistors of another of the N pixel lines. It is therefore also described a sensor carrying capacitively sensing pixels arranged forming a matrix of N rows of pixels and M columns of pixels, each pixel comprising at least:

- a detection capability;

at least one transistor connected to the detection capacitor, capable of applying to the detection capacitance a pre-charge voltage originating from at least one pre-charge line or column of the sensor, and capable of carrying out a transfer of electrical charges stored in the detection capability to at least one sensor reading circuit;

a pre-charging circuit able to vary the value of the pre-charge voltage on the pre-charge line or column.

The transistor of each pixel can be adapted to carry out the transfer of electric charges stored in the detection capacitor to the sensor reading circuit via the pre-charge line or column of the sensor, called in this case a pre-charge line or column. charge / reading of the sensor.

The sensor may further comprise:

- M pre-load / read columns such that the pixel transistors of each of the M columns of pixels are connected to one of the M pre-load / read columns;

- M read circuits each comprising at least one charge amplifier whose input is connected to one of the M pre-charge / read columns.

The sensor may further comprise a control circuit coupled to the precharge circuit and adapted to calculate a nominal value V _pr echarge_nom tension pre-load such that a maximum value V _or t_maxjnit among NxM output voltages Voutjj, with i between 1 and N and j between 1 and M, intended to be delivered by the charge amplifiers of the read circuits for each of the pixels is substantially equal to a saturation voltage of the charge amplifiers of the read circuits .

In this case, the control circuit may be adapted to perform the calculation of the nominal value V _pr echarge_nom of the precharge voltage by implementing the steps of: - applying an initial pre-charge voltage V _pr echargejnit on the column precharge / read;

reading of the output voltages Voutjj in the presence of a reference element on the pixels of the sensor;

calculating the maximum value V _or t_maxjnit among the values of the output voltages Voutjj obtained during the previous reading;

- adjusting the value of the voltage pre-load to the nominal value V _pr echarge_nom as V _or t_maxjnit is substantially equal to the saturation voltage of the read circuits of the charge amplifiers.

The control circuit may be able to control the precharging circuit such that N x M pre-charge voltages Vp chargejj are successively applied to the columns of pre-charge / read during a reading of the pixels.

Each pixel may further comprise a reference capacitor coupled to the sensing capacitance and the transistor of the pixel, and the control circuit may be adapted to perform, after the calculation of the nominal value V _pr echarge_nom of the precharge voltage, a sensor empty calibration by calculating the values of the N x M precharge voltages Vp chargejj via the implementation of the steps of:

- application of the pre-charge voltage of nominal value V _P recharge_name on the pre-charge / read columns;

calculating a mean V _or t_moy or a minimum value V _or t_min, the name of the output voltages V _or tjj obtained during the previous reading;

calculating a coefficient Ky associated with each pixel such that

Kij = Vout moy / Voutjj OR such that K = Vout min name / Voutjj;

calculation of the N x M pre-load voltages Vp loadjj such that the load loadJJ ⁼ ij.Vprecharge name.

Each pixel may furthermore comprise a reference capacitor coupled to the detection capacitance and to the pixel transistor, the control circuit may be able to control the pre-charge circuit such that N × M pre-charge voltages Vp chargejj the columns are successively applied to the columns of pre-charge / reading during a reading of the pixels, and the control circuit can be adapted to perform a vacuum calibration of the sensor in ca lcula nt the values of N x M pre-load voltages -load Vprechargejj via the implementation of the steps of:

- applying an initial pre-charge voltage V _pr echargejnit on the column precharge / read;

reading of the output voltages Voutjj in the absence of element on the pixels of the sensor;

calculating a mean V _or t_moy or a minimum value V _or t_minjnit of the output voltages Voutjj obtained during the previous reading;

calculating a coefficient Ky associated with each pixel such that

Kij = Vout moy / VoutJJ OR such that K = Vout min init / VoutJJ;

calculation of N x M pre-load voltages Vprechargejj such that VprechargeJJ ⁼ Kij. Vprechargejnit-

The control circuit can be adapted to perform, after the sensor empty calibration, an equalization operation in reading the sensor by implementing the steps of:

reading out the output voltages V _or tjj for all the pixels by applying the pre-charge voltages successively to the pre-charge / read columns;

VprechargeJJ, ^'

calculating a mean V _or t_mi of the output voltages V _or tjj for

M

M _{J =} i

- adjustment of the precharge voltages V _pr echargejj by multiplying the values of the pre-charge voltages Vprechargejj of each of the N rows of pixels by the coefficient K, associated with each of the N pixel lines. In a variant, each pixel may furthermore comprise a reference capacitor coupled to the detection capacitance and to the transistor of the pixel.

The pixel transistors may be of the TFT type.

Each pixel may furthermore comprise a second transistor capable of connecting, at a reset of the pixel, at least the detection capacitance (and the reference capacitance when the pixels comprise reference capacitors), to a line connected to a reference potential.

As a variant of the various configurations above, in which each pixel comprises a transistor for pre-charging and reading the pixel and connected to one of the M pre-charge / read columns, it is possible for the sensor to comprise N lines. pre-load / read circuit such that the pixel transistors of each of the N lines of pixels are connected to one of the N pre-charge / read lines, the sensor comprising in this case N read circuits each comprising at least one amplifier loads whose input is connected to one of N pre-charge / read lines.

As a variant, each pixel may furthermore comprise a reading transistor of the pixel connected to the detection capacitance and able to carry out a transfer of electric charges stored in the detection capacitor to at least one reading circuit of the sensor. In this case, the other transistors can form the pre-charge transistors of the pixels. In this configuration, the sensor comprises reading lines or columns through which the electrical charges read are intended to be transferred, these read lines or columns being distinct from the pre-charge lines or columns on which the pre-charge voltages are intended to be applied.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention will be better understood on reading the description of exemplary embodiments given purely by way of indication and in no way limiting, with reference to the appended drawings in which:

- Figure 1 shows a portion of a sensor object of the present invention, according to a particular embodiment; FIG. 2 represents the modification made to the output characteristic obtained during a reading of a pixel by calculating the nominal value of the pre-charge voltage;

FIG. 3 represents the modification made to the output characteristic obtained during a reading of a pixel by means of a vacuum calibration of the sensor;

FIG. 4 represents the modification made to the output characteristic obtained during a reading of a pixel thanks to a reading equalization of the sensor;

- Figures 5 and 6 show parts of a sensor object of the present invention, according to alternative embodiments.

Identical, similar or equivalent parts of the different figures described below bear the same numerical references so as to facilitate the passage from one figure to another.

The different parts shown in the figures are not necessarily in a uniform scale, to make the figures more readable.

The different possibilities (variants and embodiments) must be understood as not being exclusive of each other and can be combined with one another.

DETAILED DESCRIPTION OF PARTICULAR EMBODIMENTS Referring firstly to FIG. 1, which represents a portion of a sensor 100 according to a particular embodiment. The sensor 100 here corresponds to a fingerprint detector.

The sensor 100 comprises a matrix of pixels 102 comprising N lines and M columns, with N and M integers greater than or equal to 2. In FIG. 1, two pixels 102.1 and 102.2 arranged on two adjacent lines and on one and the same column of the matrix are represented.

Each pixel 102 has a detection capability 104, referenced 104.1 and 104.2 in FIG. 1 for each of the pixels 102.1 and 102.2. Each detection capacitance 104 comprises a first electrode disposed in the pixel 102. A second electrode of the detection capacitor 104 is formed by the part of the finger intended to face the first electrode during a fingerprint measurement. Thus, the value of each detection capacitance 104, called Cdigt, varies as a function of the distance between the first electrode of this detection capacitor 104 and the part of the finger lying opposite the first electrode, this distance being different depending on whether this part of the finger corresponds to a ridge or a groove.

Each pixel 102 also includes a pre-charge transistor 106, referenced 106.1 and 106.2 in FIG. 1 for each of the pixels 102.1 and 102.2. In the example of FIG. 1, the pre-charge transistors 106 are NMOS transistors each comprising a first source / drain electrode electrically connected to a pre-charge column 108 common to the pixels 102 of the column and on which a voltage precharge V _pr eloads is applied. A second source / drain electrode of each pre-charge transistor 106 is electrically connected to the first electrode of the detection capacitance 104 of the pixel 102.

Each pixel 102 also comprises a reference capacitor 110, referenced 110.1 and 110.2 in FIG. 1 for each of the pixels 102.1 and 102.2. Each reference capacitor 110 comprises a first electrode electrically connected to the first electrode of the detection capacitor 104 and the second source / drain electrode of the pre-charge transistor 106 of the pixel 102, and a second electrode electrically connected to a potential electrical reference, for example to ground. Unlike the detection capabilities 104, the values of the reference capacitors 110, called Cno, are fixed and do not change as a function of the portion of the finger on the pixel 102. Thus, the reference capacitors 110 provide measurement stability to the sensor 100 even when no finger is present on the pixels 102 during a measurement. The reference capacitors 110 may be formed by capacitors and / or parasitic elements present in each pixel 102.

The pre-charge transistors 106 make it possible to pre-charge the capacitors 104 and 110, that is to say to apply an initial voltage across these capacitors prior to a read operation. Each pixel 102 also comprises a read transistor 112, referenced 112.1 and 112.2 in FIG. 1 for each of the pixels 102.1 and 102.2. In the example of FIG. 1, the reading transistors 112 are NMOS transistors each comprising a first source / drain electrode electrically connected to the first electrodes of the capacitors 104 and 110 of the pixel 102. A second source / drain electrode of each read transistor 112 of the pixels of the same column of the matrix is connected to a read column 114 on which the charges stored in the capacitors 104 and 110 are intended to be transferred when the read transistor 112 is conducting, when a read operation of the pixel 102.

The sensor 100 also comprises a pre-charge circuit 115 able to apply the desired pre-charge voltage (s) to the pre-charge column 108 for each of the pixel columns 102. The pre-charge circuit 115 corresponds, for example to a digital to analog converter. The pre-charge circuit 115 is controlled by a control circuit 117 whose role and functions are detailed below.

Each read column 114 is connected, at the bottom or at the top of the column, to a read circuit 116. Each read circuit 116 comprises a charge amplifier 118, corresponding here to an operational amplifier, whose inverting input is electrically connected. 114. A compensating MOS transistor 120 is present on each read column 114 and has its source electrodes / drains short-circuited by this reading column 114. The compensation transistors 120 serve to compensate for the injection. charges occurring on the read columns 114 during a pixel read. A reference potential V _ref is applied to the noninverting input of the charge amplifier 118. The output of the charge amplifier 118 is connected to its noninverting input via a capacitance 122 value A switch 124 is connected in parallel with the capacitor 122.

The sensor 100 also has N selection lines 126, referenced 126.1 to 126.3 in FIG. 1. Each of the N selection lines 126 is associated with one of the N rows of pixels 102 such that the gates of the reading transistors 112 of the pixels 102 of FIG. said one of the N rows of pixels are electrically connected to this selection line 126. In FIG. 1, the gate of transistor 112.1 is electrically connected to the line 126.1, and the gate of transistor 112.2 is electrically connected to the selection line 126.2. In addition, although not visible in FIG. 1, the gates of the reading transistors 112 of the pixels of the line arranged under that including the pixel 102.2 are electrically connected to the selection line 126.3. During a reading of the pixels 102 of one of the N rows of pixels, a reading voltage Viectu is applied to the corresponding selection line 126 such that the reading transistors 112 of this row of the matrix become on and the loads stored in the capacitors 104, 110 of the pixels 102 of this row of the matrix are transferred to the read columns 114.

Typically, the compensation MOS transistors 120 receive on their gate a control signal corresponding to the inverse of the read voltage Viectu, in order to compensate for the coupling caused by the on-state of the reading transistors 112 on the amplifiers. 118.

In addition to the role of selection of pixels during a reading of these pixels, each selection line 126 is also used to perform a pre-charge of the pixels 102 of a row of the matrix other than that comprising the pixels whose reading transistors 112 have their gate connected to this selection line 126. For this, the gates of the precharging transistors 106 of the pixels 102 of this other line of the matrix are connected to the selection line 126. In the example of FIG. the gate of the pre-charge transistor of the pixel above the pixel 102.1 is electrically connected to the selection line 126.1. The gate of the pre-charge transistor 106.1 of the pixel 102.1 is connected to the selection line 126.2. The gate of the pre-charge transistor 106.2 of the pixel 102.2 is connected to the selection line 126.3.

Thus, during a reading of the pixels of one of the N rows of pixels 102, a pre-charge of the pixels 102 of another of the N rows of pixels 102 is performed simultaneously. In the example of FIG. 1, during a reading of the pixels 102 of the pixel line to which the pixel 102.2 belongs, the voltage applied on the selection line 126.2 turns on the reading transistors 112 (whose transistor 112.2) which are connected to this selection line 126.2. This voltage also turns on the precharging transistors 106 which are connected to this selection line 126.2, that is to say the transistors of pre-charge 106 (including the pre-charge transistor 106.1) pixels 102 of the line above that whose pixels are read.

In the example of FIG. 1, the gates of the pre-charge transistors 106 of the pixels of a line "a" of the matrix are connected to the selection line 126 to which the gates of the pixel reading transistors are connected. a line "b" of the matrix which is arranged under the line "a", because in the example of Figure 1, the rows of pixels are read in the direction from the bottom up of the matrix. If the pixel lines of the matrix were read in opposite directions, i.e. from top to bottom, then the gates of the pre-charge transistors 106 of the pixels of a line "a" of the matrix would be connected to the selection line 126 to which would also be connected the gates of the read transistors 112 of the pixels of a line "b" of the matrix which is arranged above the line "a". It would thus be possible to have the gate of the pre-charge transistor 106.2 electrically connected to the selection line 126.1 to which would also be connected the gate of the read transistor 112.1. Thus, the gates of the pre-charge transistors 106 of the pixels of a first line of the matrix are electrically connected to a selection line 126 to which are also connected the gates of the pixel reading transistors of a second line of the matrix that is adjacent to the first line. The signal flowing on the selection line 126 of the pixels of the first line of the matrix is duplicated so that it is applied to the gates of the pre-charge transistors 106 of the last pixel line of the matrix.

During a reading of the pixels 102 of one of the rows of the matrix, a reading voltage Viectu is applied to the corresponding selection line 126 such that the reading transistors 112 of the pixels 102 of this row of the matrix become on and off. that the charges stored in the capacitors 104, 110 of the pixels 102 of this row of the matrix are transferred to the read columns 114. When reading the value of a pixel 102, that is to say the reading of the sum of the cumulative charges Qsignai in the capacitors 104 and 110 of the pixel 102, the switch 124 of the read circuit 116 associated with the column to which this pixel 102 belongs is in the open position and the charges are then transferred to the capacitor 122. The charge amplifier 118 and the capacitor 122 form an integrator assembly and the reference potential V _re f is found to be applied. on the capacities 104 and 110 that are read. The voltage V _or t obtained on the output of the charge amplifier 118 is equal to:

nz signal ('γ pre-arg arg-y ref /) ^* ' finger + C 110 /) _j

ouï ref ref

When the read operation is completed, the switch 124 of each read circuit 116 is switched to the closed position.

The value of the precharge voltage V _pr eloads applied to the pre-load column 108 may be the same for all pixels of lines 102 during a read of the pixels 102 of the matrix.

Each charge amplifier 118 delivers an output voltage V _or t whose value is between that of a low saturation voltage -Vsat and that of a high saturation voltage + Vs _at t. The value of the reference potential V _re f can be chosen to be equal to that of the low saturation voltage -Vsat.

Advantageously, in order to obtain output voltages V _or t from the charge amplifiers 118 which are as representative as possible of the measurement made by the pixels 102, the value of the pre-charge voltage is chosen such that the values maximum output voltages V _or t obtained for the different read circuits 116 are as close as possible to that of the high saturation voltage (or low if the sign of the input signal is reversed at the input of the amplifiers 118) of the amplifiers 118 which is lower (or higher in the case of the low saturation voltage) than the supply voltage of the amplifiers 118, for example between about -5V and + 5V.

However, the structure of the sensor 100 uses pre-charge columns 108 distinct from the selection lines 126, which has the advantage of allowing the value of the pre-charge voltage to be adjusted independently of the read reading voltages applied. on the selection lines 126. By choosing the appropriate value of the pre-charge voltage such that the maximum values of the output voltages V _or t obtained are as close as possible to the saturation voltage of the charge amplifiers 118, it is possible to maximize the dynamics of the output voltages of the amplifiers 118. Thus, for a given range of values of the detection capabilities 104, the range of values of the output voltages of the amplifiers 118 for these values of the detection capabilities 104 is extended.

It is also possible to compensate for any variations in capacitance to be measured which strongly depend on the dielectric and / or protective layer thicknesses of the surface of the pixel array 102, and thus maintain the value of Qsignal at optimal values.

This adjustment of the value of the pre-charge voltage can be achieved either manually or automatically for example by an external control loop.

This possibility of adjusting the value of the precharge voltage V _pr eloads notably allows to integrate the sensor 100 of the preexisting reading circuits 116 which are not optimized specifically for the sensor 100.

To be able to modify the value of the pre-charge voltage, it is for example possible to connect the output of a digital-to-analog converter (corresponding to the pre-charging circuit 115) to the pre-charge columns 108, the value of the pre-charge voltage being regulated by the converter. It is also possible to use other solutions, such as a multiplexer or a variable voltage source.

In order for the value of the pre-charge voltage to be chosen such that the maximum values of the output voltages V _or t are as close as possible to that of the saturation voltage of the charge amplifiers 118, it is possible to calculate a value V _pr echarge_nom nominal voltage precharge.

For this, a voltage of initial preload V _pr echargejnit is applied to the pre-load column 108. The value of the voltage of initial preload Vprechargejnit is arbitrarily chosen.

The minimum values of the output voltages are obtained when no element is present on the sensor, which corresponds to Cdigt = 0. By taking the previous definition of V _or t, this minimum value called V _or t_minjnit is then such that :

r

v ouï _ min_init = v ref + ('V prech arg e_in init - V ref /)' ^ ¹¹⁰ For the calculation of the nominal value V _pr echarge_nom, a reference element, for example corresponding to a finger model or a metal plate is disposed on the pixels 102 of the sensor 100 so that the value of the Cdoigt capacity is at its maximum Cdoi _g t_max.

The output voltages V ₀ ut_u are then read for all the pixels 102, with i lying between 1 and N and corresponding to the line of the considered pixel, and j lying between 1 and M and corresponding to the column of the pixel in question, in the presence of the pixel reference element 102.

A maximum value V _or t_maxjnit among all the values of the output voltages V ₀ ut_u obtained during the previous reading is then calculated.

Using the previous definition of V _or t, this maximum value V ₀ ut_max_init can be expressed according to the equation:

The value of the pre-charging voltage is then adjusted to the nominal value V _pr echarge_nom as V _or t_maxjnit is substantially equal to the voltage Vs + high saturation _has been amplifiers 118.

In the case where V _re f = 0, this nominal value can be expressed by the following equation:

+ V

prech arg e _ name prech arg e _ init '

Hearing

The minimum output voltage V _or t_min_name obtained using the nominal pre-charge voltage is then greater than the value V _or t_minjnit and such that:

r

V ï _ _ min_ name = v ref + ('V prech arg e _ name - V ref /)' ^ ¹¹⁰

2 shows the value of the output voltage of an amplifier 118 according to the Cdoigt value of a detection capabilities 104. When the initial precharge voltage V _pr echargejnit is applied to the precharge column 108, the excursion of the output voltage V t _or the amplifier 118 is between V and V _or t_min_init _or t_maxjnit for Cdoigt between 0 and C _d0 igt_max (characteristic referenced 10 in Figure 2). When the pre-charge voltage is adjusted to its nominal value Vprecharge nom, the excursion of the output voltage V _or t of the amplifier 118 is then between V _or t_min_nom and + V _Sa t for Cdigt between 0 and Cdoi _g t_max (ca ractéristique referenced 12 in Figure 2), which is more important than when the initial pre-charge voltage Vp is applied hargejnit _{the rec.}

Advantageously, the value of the pre-charge voltage may be different for different groups of pixels 102, and in particular for each of the rows of pixels 102 or each of the columns of the pixels, or for each of the pixels 102. value of the pre-charge voltage is variable during the sequential reading of the different rows or columns of pixels 102 of the matrix or for each pixel 102.

The adjustment of the pre-charge voltage may also be used to realize, after the calculation of the nominal value of the voltage V _pr preload echarge_nom, a vacuum calibration of the sensor 100. This vacuum calibration aims to define a preload voltage preferably adapted for each pixel 102 (but which can also be adapted for each row or each column of pixels) and makes it possible to compensate for the measurement distortions due to technological dispersions, in particular reference capacitors 110. For this purpose, the control circuit 117 and the pre-charging circuit 115 are adapted to successively apply one of the N x M pre-charge voltages Vp chargejj for each of the N x M pixels. This vacuum calibration is performed in the absence of element on the pixels 102 of the sensor 100.

The pre-charge voltage of nominal value V _pr echarge_nom is applied on the column precharge 108, then the output voltages V ₀ ut_u are read. Considering V _re f = 0, these output voltages can be expressed as:

Where CHOJJ are the values of each reference capacitor 110 which are different from each other due to the technological dispersions. The average value V _or t_moy of the output voltages V ₀ ut_u obtained during the previous reading is then calculated. This average value is equal to

N M

v out moy = - ~ \ -Y Yv out ι. . .

NM f -

This average value can also be expressed according to the equation

V prech arg e name C 110 a

V _{out moy} = - -, where Cno_moy is the average value of all abilities

C _f

reference 110.

A coefficient Ky is then calculated for each pixel 102 such that

NxM precharge voltages V _pr echargejj are then calculated such that Vprechargejj = Kij.V _P recharge_nom. As shown by the equation below, the output voltages V _or tjj thus become, in the absence of element on the pixels 102 of the sensor, equal to the average of these output voltages:

V prech arg e _ i _ j ^* C 110 _ _ j V prech arg e _ name c \ \ _ _ av

out î _ _ j ^ ^or - ^m ° y

Thus, during a reading of the pixels 102 with an element to be detected present on the pixels, the differences between the values V _or tjj due to the technological non-uniformities between the reference capacitors 110 of the pixels are eliminated.

3 shows the value of the output voltage of an amplifier 118 according to the Cdoigt value of the detection capability 104 when the voltage pre-charge nominal value V _pr echarge_nom is applied to the column pre charge 108. The various points designated by the reference 14 correspond to different output voltages obtained in the absence of element on the pixels 102, these differences being due to the variations on the value of the reference capacitors 110. output all have the same value for the same capacitance value Cdigt, the above method corrects these values so that they take a value equal to the average value Vout avg. As a variant of the vacuum calibration method described above, it is possible to use a minimum value V _or t_min_nom, corresponding to the minimum value among the output voltages V _or t_u obtained by using the nominal value V _P recharge_nom the pre-charge voltage, instead of the average value V _or t_moy. This also makes it possible to improve the output dynamics of the amplifiers 118.

Alternatively, it is possible that this vacuum calibration is performed not by calculating a coefficient K for each pixel, but by calculating a coefficient K for each line or each column of pixels. In this case, the coefficients K may correspond to the ratio of the average value V _or t_moy, or the minimum value Vout_min_nom of the output voltages, to the average value of the output voltages of the pixels of the corresponding line or column.

As an alternative to the vacuum calibration method of the sensor described above, it is possible to subtract, at each output voltage, the difference between its nominal value and the average V _or t_moy in order to equalize it to the latter, c ' that is, such that VoutJJ ⁼ VoutJJ + (Vout moy ^- Vout ij) ⁼ Vout avg.

The adjustment of the pre-charge voltage can also be used to carry out, after the empty calibration of the sensor 100, an equalization operation in reading of the sensor. This operation makes it possible, for example, to optimize the values of the output voltages Vout for non-uniform capacitance values according to the line or column of pixels 102 considered.

For example, for a fingerprint capture, the finger pressure on the sensor 100 is generally not the same over the entire area of the pixel array 102. On the center area of the pixel array 102, the value average of the measured capacitances is often greater than in the upper and lower end regions of the pixel matrix 102 because the finger-electrode distance (the first electrode of the detection capacitance 104) is lower at this central zone than in the other regions because of greater finger pressure pressure at this central area. In order to compensate for this, it is possible to implement a method of equalizing the reading of the sensor 100 comprising the steps below. The output voltages V _or t_u for all the pixels 102 are read by successively applying to the pre-charge columns 108 the pre-charge voltages Vp chargejj previously determined during the empty calibration of the sensor 100. For each pixel 102, the output voltage obtained is (considering V _re f = 0):

(r + C)

V.

A calculation of a mean V _or t_mi of the output voltages V ₀ ut_u for each of the N rows of pixels 102 is then performed such that V,. = - / V,. . .

M

A maximum value V _or t_mi_max corresponding to the maximum value among the set of average values V _or t_mi previously calculated, is then determined. The calculation of this maximum value V _or t_mi_max makes it possible to identify the line of pixels 102 where the pressure of the finger is the strongest. As a variant, it is possible to calculate a mean V _or t_mi_moy of the values V _or t_mi.

A coefficient K, is then calculated for each of the N rows of pixels such that K, = V _or t_ mi max / v out mi or such that K, = V _or t_ mi moy / v out mi.

Finally, the pre-charge voltages Vp chargejj intended to be successively applied to the pre-charging columns 108 during a reading of each of the N rows of the pixels 102 are then adjusted by multiplying the values of the pre-charge voltages Vpchargejj of each of the N rows of pixels by that of the coefficient K, associated with each of the N rows of pixels, that is to say such that

Adjusted JJ Charge ⁼ i.JJ Charge.

The different adjusted values of the pre-charge voltages Vp-charge are then applied successively to the pre-charge columns 108 during the successive reading of the different pixels 102. In this way, the average of the output voltages of the pixels of each line becomes close. of V ₀ ut_mi_max or V ₀ ut_mi_moy regardless of the pressure exerted by the finger. The implementation of these steps therefore has the effect of compensating the pressure differences on different regions of the pixel matrix 102, which simulates the support of a "flat" finger that would apply a uniform pressure on all pixel lines 102 of the matrix. Figure 4 illustrates the read equalization performed by implementing the steps described above. The points designated by the reference 16 correspond to the output voltages obtained for a first pixel line, and those designated by the reference 18 correspond to the output voltages obtained for a second pixel line. Those designated by the reference 20 correspond to those obtained for the line of pixels whose average corresponds to V _or t_mi_max. By applying the pre-charge voltages adjusted by the previously described read equalization, the output voltages designated by the references 16 and 18 are brought to the same level as those whose average is maximum.

Several variants of the sensor read equalization previously described can be envisaged. During the equalization, it is for example possible to calculate not the average per line of the output voltages, but the average per column of pixels. The coefficients K can also be calculated such that each coefficient K is associated with a column of pixels and not with a row of pixels. Thus, it is the pressure differences between the pixel columns that are compensated by the read equalization.

It is also possible to consider calculating a coefficient K for each pixel.

It is also possible to combine the read equalization of the pixel columns and the read equalization of the pixel lines to compensate for the horizontal and vertical pressure differences on the sensor 100.

In addition, it is also possible that the pre-charge voltages are not calculated for each pixel, but they are calculated such that a single pre-charge voltage is associated with each row or column of pixels.

In the description above, a first operation is performed by calculating the nominal value V _pr echarge_nom maximizing the maximum values of the output voltages of the charge amplifiers 118, and a vacuum sensor calibration is carried out by using of the V _pr echarge_nom value to determine the preload chargejj Vp values, and finally a sensor reading equalization is performed by making use of preload Vp chargejj values and the value V _pr echarge_nom to adjust the values of precharge voltages Vp chargejj of.

Alternatively, it is possible to implement the operation for calculating the nominal value V _pr echarge_nom and equalization sensor reading, without performing the vacuum sensor calibration. It is also conceivable embodiment of the vacuum sensor calibration and / or sensor reading equalization without precalculated nominal value V _pr echarge_nom voltage precharge. In this case, in the steps described above using this nominal value, the initial value V _pr echargejnit the pre-charge voltage determined arbitrarily is used.

The steps of these operations can be performed by the control circuit 117, or computer, external to the pixel array 102, which controls the pre-charge circuit 115 applying the different pre-charge voltages to the pre-charge columns. 108 (for example a digital-to-analog converter).

The various operations described above can be implemented by considering not each row or each column of pixels individually, but considering groups of rows or columns of pixels 102. It is also possible that these operations are not implemented. implemented for all the pixels. Finally, it is also possible that this method is repeated and applied separately for different regions of the pixel matrix, by implementing the process steps several times for different rows or columns of pixels 102.

In the sensor 100 previously described, the pre-charge voltages are applied to pre-charge columns 108 each connected to a column of pixels. Alternatively, these pre-charge voltages can be applied to precharge lines which are each connected to a pixel line 102.

The sensor 100 is advantageously made of TFT technology from amorphous silicon or polysilicon, which makes it possible to produce important detection surfaces at relatively low costs. This technology does not allow the integration of a large number of transistors in the surface of a pixel (typically of dimensions equal to about 50 x 50 μιη ² for a digital fingerprint sensor) as important as in technology. CMOS. The sensor 100 may especially be made of OTFT ("Organic Thin-Film Transistor") or PTFT ("Polymer Thin-Film Transistor") technology, also known as organic or printed electronics, implementing printing techniques that have the advantage of having very low manufacturing costs, including for very large surfaces. The consequence is the limitation of the size of the pixels (dimensions generally greater than about 500 μιη) when it is desired to include at least two transistors in a pixel. On the other hand, its low cost per unit of surface is an interesting economic alternative for the realization of a sensor with capacitive detection having a large capture area requiring only a medium resolution, as for example to realize a complete detection of the hand (detection palmar). Such a device is for example used to perform medium security access controls as a redundant and / or additional identification device to a digital access code.

The compensation method described above advantageously applies to correct the non-uniformity inherent in printing techniques.

The sensor 100 can however be realized in CMOS technology. The strong integration capability of CMOS technology allows other high performance reading techniques (eg RF, etc.) but at much higher costs than TFT technology.

The different operations of adjusting the pre-charge voltage

(Maximization of the maximum values of the output voltages of the amplifiers, vacuum calibration of the sensor, sensor read equalization) advantageously apply to the sensor 100 described in connection with FIG. 1 in which each selection line 126 is connected to the gates of the read transistors 112 of the pixels 102 of one of the N rows of pixels 102 and the gates of the pre-charge transistors 106 of pixels 102 of another of the N rows of pixels 102. However, these voltage adjustment operations pre-charge may also apply to a sensor other than the sensor 100 previously described, that is to say not having at least Nl separate selection lines of the line or the pre-charge column and such that each selection line is linked to the grids of the transistors for reading pixels of one of the N pixel lines and the gates of the pixel pre-charge transistors of another of the N pixel lines.

Figure 5 shows schematically a portion of a sensor 200, here corresponding to a fingerprint detector. In this figure, only one pixel 202 is shown.

Each pixel 202 of this sensor 200 comprises a first transistor 204 whose gate is connected to a selection line 206. Each pixel 202 includes a detection capacitor 208. The detection capacitor 208 comprises a first electrode disposed in the pixel 202. A second electrode 210, shown symbolically in FIG. 5, is formed by the portion of the finger intended to face the first electrode of the detection capacitor 208 during a fingerprint measurement. Each pixel 202 also comprises a reference capacitor 212. The capacitors 208 and 212 operate in a manner similar to the capacitors 104 and 110 previously described for the sensor 100. Each pixel 202 also comprises a second transistor 214 enabling connection, during a reset at zero of the pixel 202, the capacitors 208 and 212 at the line 216 which is connected to ground. The second transistor 214 comprises its gate connected to a reset column 218 on which a control signal of the transistor 214 flows.

In this sensor 200, the selection line 206 is used both to preload the two capacitors 208, 212 of the pixel 202 and then read the resulting electrical charge. One of the source / drain electrodes of the first transistor 204 is connected to a pre-charge / read column 220 on which a pre-charge voltage is to be applied during a pre-charge of the capacitors 208, 212 , and on which the charges stored in the capacitors 208, 212 are intended to be transferred during a reading of the pixel 202.

The pre-charge / read columns 220 associated with the pixel columns 202 of the sensor 200 are connected to a switching circuit 222 to which the pre-charging circuit 115 and the control circuit 117, similar to those of the sensor 100, are connected. as well as the read circuits 116 also similar to those previously described for the sensor 100. Unlike the sensor 100 previously described in which the reading of one of the lines of pixels is performed simultaneously with the pre-loading of the capacitances of the pixels of another line, the sensor 200 operates by first realizing the precharging of the pixel 202. sending the pre-charge voltage to the pre-charge / read column 220, and then reading that pixel 202 via the same pre-charge / read column 220.

The pre-charge voltage applied can be adjusted as previously described for the sensor 100 by performing a first adjustment via the calculation of a nominal value of the pre-charge voltage making it possible to maximize the maximum values of the output voltages of the amplifiers. loads of the read circuits 116, and / or a vacuum calibration of the sensor 200 making it possible to eliminate the variations due to the technological dispersions between the reference capacitors 212 of the pixels 202, and / or a read equalization of the sensor 200 enabling the compensation of the measurement differences between rows and / or columns of pixels. For this, the various steps previously described for the sensor 100 apply to the sensor 200 in a similar manner.

According to a variant shown in FIG. 6, each pixel 202 of the sensor 200 further comprises a third transistor 224 forming a reading transistor of the pixel 202. The gate of the third transistor 224 is connected to a read control line 226 on which is applied the playback control signal. Thus, only a pre-charge control signal is sent on the selection line 206, and the columns 220 are dedicated solely to pre-charging the pixels 202. One of the source / drain electrodes of the third transistor 224 is connected to the capacitors 208, 212 and the other source / drain electrode of the third transistor 224 is connected to a read column 228 common to all the pixels arranged on the same column, on which the charges stored in the capacitors 208 are transferred, 212 during a reading. This variant makes it possible to separate the pre-load and read operations. The switching circuit 222 is not present in this configuration. The various adjustment operations of the pre-charge voltages applied to the pre-charge columns load 220 previously described can also be performed for the sensor 200 according to this variant.

According to a variant of the configuration shown in FIG. 6, the pixels 202 of the sensor 200 may not comprise the second transistors 214 or the reset columns 218.

Claims

A sensor (100) having capacitively sensing pixels (102) arranged forming a matrix of N rows of pixels (102) and M columns of pixels (102), each pixel (102) comprising at least:

a detection capability (104);

a pre-charge transistor (106) connected to the detection capacitor (104) and able to apply to the detection capacitor (104) a pre-charge voltage originating from at least one pre-charge line or column; load (108) of the sensor (100);

- a read transistor (112) connected to the detection capacitor (104) and capable of transferring electric charges stored in the detection capacitor (104) to at least one read circuit (116) of the sensor (100) ;

the sensor (100) further comprising at least N1 distinct selection lines (126) of the pre-charge line or column (108) and such that each selection line (126) is connected to the gates of the read transistors ( 112) pixels (102) of one of the N pixel lines (102) and the gates of the pre-charge transistors (106) of the pixels (102) of another of the N pixel lines (102).

2. Sensor (100) according to claim 1, further comprising a pre-charge circuit (115) adapted to vary the value of the pre-charge voltage on the pre-charge line or column (108).

3. Sensor (100) according to one of the preceding claims, further comprising:

- M read columns (114) such that the reading transistors (112) of the pixels (102) of each of the M columns of pixels (102) are connected to one of the M read columns (114);

- M read circuits (116) each comprising at least one charge amplifier (118) having an input connected to one of the M read columns (114); N pre-charge lines such that the pre-charge transistors (106) of the pixels (102) of each of the N rows of pixels (102) are connected to one of the N pre-charge lines, or M pre-charge columns. charge (108) such that the pre-charge transistors (106) of the pixels (102) of each of the M pixel columns (102) are connected to one of the M pre-charge columns (108).

4. The sensor (100) according to claims 2 and 3, further comprising a control circuit (117) coupled to the precharge circuit (115) and adapted to calculate a nominal value V _pr echarge_nom voltage precharge such that a maximum value V ₀ ut_max_init among N x M output voltages Voutjj, with i between 1 and N and j between 1 and M, to be delivered by the charge amplifiers (118) of the read circuits ( 116) for each of the pixels (102) is substantially equal to a saturation voltage of the charge amplifiers (118) of the read circuits (116).

5. The sensor (100) according to claim 4, wherein the control circuit (117) is adapted to perform the calculation of the nominal value V _pr echarge_nom voltage precharge implementing the steps of:

- applying an initial pre-charge voltage V _pr echargejnit on lines or pre-load column (108);

reading the output voltages Voutjj in the presence of a reference element on the pixels (102) of the sensor (100);

calculating the maximum value V _or t_maxjnit among the values of the output voltages V _or tjj obtained during the previous reading;

- adjusting the value of the voltage pre-load to the nominal value V _pr echarge_nom as V _or t_maxjnit is substantially equal to the saturation voltage of the charge amplifiers (118) read circuits (116).

6. Sensor (100) according to one of claims 4 or 5, wherein the control circuit (117) is adapted to control the pre-charge circuit (115) such that N x M distinct pre-charge voltages Vprechargejj are successively applied to the pre-charge lines or columns (108) during a reading of the pixels (102).

The sensor (100) according to claims 5 and 6, wherein each pixel (102) further comprises a reference capacitor (110) coupled to the detection capacitance (104), the pre-charge transistor (106) and the read transistor (112) of the pixel (102), and wherein the control circuit (117) is adapted to perform, after the calculation of the nominal value V _pr echarge_nom of the voltage pre-charge, a vacuum calibration of the sensor (100) by calculating the values of the N x M preload voltages Vprechargejj via the implementation of the steps of:

- application of the pre-charge voltage of nominal value V _P recharge_nom on lines or pre-charge columns (108);

- Reading the output voltages Voutjj in the absence of element on the pixels (102) of the sensor (100);

calculating a mean V _or t_moy or a minimum value V _or t_min_nom of the output voltages Voutjj obtained during the previous reading;

calculating a coefficient Ky associated with each pixel (102) such that Kij = Vout moy / VoutJJ or such that Ky = V _or t_minname / v out y,

- calculation of N x M pre-load voltages V _pre loadjj such that VprechargeJJ ⁼ Ky .Vprecharge nom-

The sensor (100) according to claims 2 and 3, wherein each pixel (102) further comprises a reference capacitor (110) coupled to the detection capacitance (104), the pre-charge transistor (106) and to the read transistor (112) of the pixel (102), wherein the control circuit (117) is adapted to control the precharging circuit (115) such that N x M distinct pre-charge voltages Vprechargejj are successively applied to the pre-charge lines or columns (108) during a reading of the pixels (102), and wherein the control circuit (117) is adapted to perform a vacuum calibration of the sensor (100) by calculating the values of the N x M preload voltages Vprechargejj via the implementation of the steps of: - applying an initial pre-charge voltage V _pr echargejnit on lines or pre-load column (108);

calculating an average V ₀ ut_moy or a minimum value V _or t_min init output voltages Voutjj obtained during the previous reading;

calculating a coefficient Ky associated with each pixel (102) such that

Kij = Vout moy / VoutJJ or such that Ky = V _or t_min init / VoutJJ, '

calculation of the N x M preload voltages V _pre loadjj such that VprechargeJJ-Kij. Vprechargejnit-

9. Sensor (100) according to one of claims 7 or 8, wherein the control circuit (117) is adapted to perform, after the vacuum calibration of the sensor (100), an equalization operation read the sensor (100) implementing the steps of:

reading the output voltages V _or tjj for all the pixels (102) by successively applying on the pre-charge lines or columns (108) the pre-charge voltages V _pr ehar _g ejj;

calculating a mean V _or t_mi of the output voltages V _or tjj for

M

each of N pixel lines (102) such that V _{out ni} = -ΣV _{0Ut t} ;

M _{J =} i

calculation of a maximum value Vout mi max or of an average V _or t_ mi average of the values V _or t mi,

calculating a coefficient K, for each of the N rows of pixels (102) such that Ki = V ₀ ut_max max / Vout mi or such that K, = V _or t_mix av / Vout mi, '

adjusting the pre-charge voltages Vprecharge by multiplying the values of the pre-charge voltages V charge of each of the N rows of pixels by the coefficient K associated with each of the N rows of pixels.

The sensor (100) according to one of claims 1 to 6, wherein each pixel (102) further comprises a reference capacitor (110) coupled to the detection capacitor (104), to the pre-charge transistor ( 106) and the read transistor (112) of the pixel (102).

11. The sensor (100) according to one of the preceding claims, wherein the pre-charge transistors (106) and reading (112) of the pixels (102) are of the TFT type.