Polarization of electronic modules, in particular analog radio frequency modules

Abstract

A method for biasing at least one first electronic module from a first continuous bias current, said at least one first module (MDi) being supplied by a supply voltage (Vdd) having actual values ​​dispersed around a nominal value, said at least one first module having at least one first physical parameter whose value can undergo a variation resulting from said voltage value dispersion, the method comprising open-loop compensation of said voltage value dispersion including the generation of a first corrected current (IREF_corr) from a reference current and a first correction factor (GFF) determined from the variation of the value of said at least one first physical parameter, resulting from said voltage value dispersion, and the generation of the first bias current (IREF_i) from the first corrected current. Figure for the abstract: Fig 3

Description

Title of the invention: Polarization of electronic modules, in particular radiofrequency analog modules

[0001] Embodiments and implementations relate to the polarization of electronic modules with continuous polarization currents, in particular analog modules produced in BiCMOS or CMOS technology and used in radiofrequency (RF) integrated circuits, for example RF amplifiers, mixers, phase shifters.

[0002] The performance of these modules depends heavily on these bias currents.

[0003] However, variations in the supply voltage are a significant cause of variations in these polarization currents, thus leading to a degradation in the performance of these modules, particularly in terms of gain.

[0004] In general, to overcome variations in the power supply, either the current sources are cascoded or the entire integrated circuit is equipped with a low dropout voltage regulator ("LDO: Low Drop Out") delivering the power supply voltage.

[0005] However, these two solutions have the disadvantage of reducing the value of the supply voltage delivered to the various components of the integrated circuit, which can prove very restrictive or even unacceptable when the supply voltage supplying the regulator has a low value.

[0006] There is therefore a need to reduce the impact of variations in the supply voltage on the bias currents across the entire integrated circuit, without imposing constraints on the supply voltage.

[0007] According to one aspect, an integrated circuit is provided, comprising at least a first electronic module configured to be powered by a supply voltage and to be biased from a first direct bias current.

[0008] The supply voltage may have actual values ​​scattered around a nominal value.

[0009] Said at least one first module has at least one first physical parameter, for example a gain, the value of which may undergo a variation (or dispersion) resulting from said dispersion of voltage values.

[0010] The integrated circuit according to this aspect comprises centralized biasing means comprising first compensation means configured to perform open-loop compensation of said dispersion of the voltage values.

[0011] This open-loop compensation includes the development of a first current corrected from a reference current and a first correction coefficient determined from the variation (or dispersion) of the value of said at least one first physical parameter resulting from said dispersion of the voltage values.

[0012] Thus the first correction coefficient is advantageously adjusted as a function of the sensitivity presented by said at least one first module to variations in voltage, this sensitivity being reflected by a variation in the first physical parameter, for example the gain, of the first module.

[0013] For example, in the case where the first physical parameter is the gain of the first module, a first module which would have a gain variation of 0.5dB when the voltage varies from 1.7V to 1.9V for example would be assigned a higher first coefficient than a first module which would only have a variation of 0.2dB for the same variations (or dispersion) of voltage values.

[0014] The centralized polarization means also comprise first current copying means, for example of the current mirror type, configured to produce the first polarization current from the first corrected current.

[0015] Centralized open-loop pre-compensation, and therefore centralized pre-correction, of the bias current and possibly of all the bias currents (when there are several first electronic modules), makes it possible to obtain less sensitivity of the bias current(s) to variations in the supply voltage.

[0016] According to one embodiment, the centralized polarization means comprise a reference current source configured to deliver the reference current.

[0017] Furthermore, the first compensation means are configured to -determine a voltage difference between the actual value of the supply voltage and its nominal value, -transform this voltage difference into a first correction current using the first correction coefficient and -inject this first correction current into the reference current source to obtain the first correction current.

[0018] According to one embodiment, the integrated circuit comprises a reference voltage source, for example a “band gap” type voltage source configured to deliver a reference voltage.

[0019] And the first means of compensation include for example: -a first transconductance operational amplifier having a first input connected to the reference voltage source and an output looped back to a second input of this first transconductance operational amplifier, - a first current copying block connected to the output of the first transconductance operational amplifier and having a copying factor adjustable according to the first correction coefficient, - a resistive voltage divider bridge having an input to receive the supply voltage and an output.

[0020] This divider bridge is also sized to deliver at its output a voltage equal to the reference voltage when the supply voltage present at its input has its nominal value.

[0021] According to this embodiment, the first compensation means also comprise: - a second transconductance operational amplifier having a first input connected to the output of the divider bridge and an output looped back to a second input of this second transconductance operational amplifier, - a second current copying block connected to the output of the second transconductance operational amplifier and having said copying factor adjustable as a function of the first correction coefficient.

[0022] The node connected to the respective outputs of the two current copy blocks is capable of delivering the first correction current.

[0023] The reference current source and the first current copying means (which deliver the first polarization current(s)) are connected to said node.

[0024] Although not essential, the first current copying block may comprise two first cascoded current copying stages and the second current copying block may comprise two second cascoded current copying stages.

[0025] The use of cascoded current copying (or mirroring) stages makes it possible to make the polarization current(s) even less sensitive to variations in the supply voltage.

[0026] The first correction parameter is advantageously programmable.

[0027] It can be stored in a register for example and being able to modify it makes it possible to adapt to different types of electronic modules and / or applications.

[0028] As indicated above, the first physical parameter is for example the gain of the first module.

[0029] The integrated circuit may comprise several first modules configured to be powered by the supply voltage and to be biased from respective first DC bias currents.

[0030] And the first current copying means are configured to produce these respective first polarization currents from said corrected current.

[0031] The first modules can for example form a radiofrequency transmission or reception chain and the first physical parameter is for example the gain of the transmission or reception chain.

[0032] According to one embodiment, the integrated circuit further comprises several second modules configured to be powered by the supply voltage and to be biased from respective second continuous bias currents.

[0033] The first modules form, for example, a radiofrequency transmission chain and the second modules form, for example, a radiofrequency reception chain having a gain forming a second physical parameter, the value of which may undergo a variation resulting from said dispersion of voltage values.

[0034] The centralized polarization means comprise -second compensation means, similar to the first compensation means, configured to carry out open-loop compensation of said dispersion of the voltage values ​​comprising a development of a second corrected current from the reference current and a second correction coefficient determined from the variation of the value of said second physical parameter resulting from said dispersion of the voltage values, and - second current copying means configured to produce the second polarization currents from the second corrected current.

[0035] The second compensation means and the second current copying means may be distinct from the first compensation means and the first current copying means, respectively.

[0036] Alternatively, as the transmission and reception chains operate in practice with time division multiplexing, the second compensation means and the second current copying means could be combined with the first compensation means and the first current copying means, to form only one and the same compensation means and one and the same current copying means, implemented sequentially by sequentially using the first correction coefficient and the second coefficient, as a function of the time division multiplexing.

[0037] According to another aspect, there is provided a method of polarizing at least one first module from a first continuous polarization current, said at least one first module being supplied by a supply voltage having real values ​​dispersed around a nominal value, said at least one first module having at least one first physical parameter whose value can undergo a variation resulting from said dispersion of voltage values, the method comprising - an open-loop compensation of said dispersion of the voltage values ​​comprising a development of a first corrected current from a reference current and a first correction coefficient determined from the variation of the value of said at least one first physical parameter resulting from said dispersion of the voltage values, and -a development of the first polarization current from the first corrected current.

[0038] According to one embodiment, the method comprises a delivery of a reference current by a reference current source and said compensation comprises -a determination of a voltage difference between the actual value of the supply voltage and its nominal value, - a transformation of this voltage difference into a first correction current using the first correction coefficient and -an injection of this first correction current into the reference current source to obtain the first correction current.

[0039] According to another aspect, there is provided a method for determining the value of the first correction coefficient used in the integrated circuit as defined above or in the method as defined above, comprising, after the manufacture of the integrated circuit, a measurement of the variation of the first physical parameter caused by the dispersion of the values ​​of the supply voltage, for different values ​​of the first correction coefficient, and a selection of the value of the first correction coefficient corresponding to a desired value for the variation of the first physical parameter.

[0040] Other advantages and characteristics of the invention will appear on examining the detailed description of the embodiment and implementation, which is in no way limiting, and the appended drawings in which:

[0041] [Fig.l] ;

[0042] [Fig.2] ;

[0043] [Fig.3] ;

[0044] [Fig.4] ;

[0045] [Fig.5] ;

[0046] [Fig.6] ; and

[0047] [Fig.7] ;

[0048] schematically illustrate embodiments and implementations of the invention.

[0049] In [Fig.l], the reference IC designates an integrated circuit comprising here a chain of first modules MD1-MD5 forming for example a radiofrequency transmission chain.

[0050] The input signal of the chain is referenced RF^ and the output signal RFOut-

[0051] Each first MDi module is supplied by a supply voltage Vdd which can have real values ​​dispersed around a nominal value.

[0052] For example, the nominal value may be equal to 1.8 volts and the actual values ​​may be between 1.7 volts and 1.9 volts for example.

[0053] Furthermore, each first MDi module is intended to be polarized from a first direct bias current IREFj.

[0054] The polarization currents Irefj are delivered by centralized polarization means CBIAS, the structure and function of which will be discussed in more detail below.

[0055] The CHT transmission chain is here affected by a first physical parameter, for example the gain, i.e. the RFOut / RFin ratio, the value of which may undergo a variation or dispersion resulting from the dispersion of the voltage values ​​of the supply voltage Vdd.

[0056] However, we can already indicate that these centralized CBIAS polarization means include: - first compensation means configured to carry out open-loop compensation of the dispersion of the voltage values, this open-loop compensation comprising the development of a first corrected current from a reference current and a first correction coefficient GFF determined from the variation in the value of the first physical parameter, in this case the gain of the CHT transmission chain, resulting from said dispersion of the voltage values.

[0057] The centralized polarization means also comprise first current copying means configured to develop each polarization current IREFj from the first corrected current.

[0058] This first correction coefficient GFF, for example stored in a register RG, is constant but can be programmable by control means MC, for example a microcontroller of the integrated circuit IC using a digital control word Gff-ADJ.

[0059] Although the bias current Irefj may actually be the bias current of the MDi module, it may be, as illustrated in [Fig.2], a reference current from which the effective bias current of the MDi module will be developed.

[0060] Indeed, in general, and as is particularly the case for radiofrequency modules, an MDi module may comprise, as illustrated in [Fig.2], a local polarization means LBS comprising current copying means based on NMOS and / or PMOS transistors, intended to deliver the effective polarization current IBIAS from the current Irefj, to the radiofrequency stage RTSG of this MDi module, this radiofrequency stage RTSG being powered by the supply voltage Vdd and receiving the current Idd_BLOCK.

[0061] In this case, the current Irefj can be considered as a reference current.

[0062] That being said, in the rest of the text, and for the purpose of homogenizing the terms, we will continue to refer to it as the “first polarization current”.

[0063] Reference is now made more particularly to Figures 3 and 4 to describe in more detail details an example of the structure of the first compensation means MCP1 and the first current copying means MIR1.

[0064] [Fig.3] also illustrates a mode of implementation of a method of polarizing each first module MDi from the first corresponding direct polarization current Irefj.

[0065] The open-loop compensation of the dispersion of the values ​​of the supply voltage Vdd comprises, as indicated above, - a development by the first compensation means MCP1, of a first corrected current IRFF_coir from a reference current IRFF0 and the first correction coefficient GFF, and - an elaboration by the current copying means MIR1 of each first polarization current IRFFJ from the first corrected current lREF_con-

[0066] The reference current Ij^po is delivered by a reference current source SCR.

[0067] Open loop compensation includes a determination of a voltage difference VFF (schematically symbolized by a subtractor STR) between the actual value of the supply voltage Vdd and its nominal value VddTYp.

[0068] This voltage difference VFF is transformed into a first correction current I corr using the first correction coefficient GFF.

[0069] And, this first correction current ICorr is injected into the reference current source SCR to obtain the first corrected current

[0070] This first corrected current is delivered to the input IN of the first current copying means MIR1 and the different first polarization currents IRFFJ are delivered to the outputs OUT of these first current copying means MIR1.

[0071] In [Fig.4], we see that the integrated circuit comprises a reference voltage source BG, for example a bandgap voltage source configured to deliver a reference voltage Vbg, typically of the order of 1.2 volts.

[0072] The first compensation means MCP1 comprise in this embodiment, a first operational transconductance amplifier, of conventional structure and known per se, referenced OTAI.

[0073] This first OTAI transconductance operational amplifier has - a first input + connected to the reference voltage source BG to receive the reference voltage Vbg, and - an output looped back to a second input - of this first operational amplifier of the OTAI transconductance.

[0074] The output of this first transconductance operational amplifier OTAI is also connected to ground GND by a resistor R31.

[0075] Furthermore, the first compensation means MCP1 comprise a first ETM1 current copying block whose input is connected to the output of the OTAI transconductance amplifier and having an adjustable copying factor f(GFF) as a function of the first GFF correction coefficient.

[0076] In practice, depending on the value of this first correction coefficient GFF, more or fewer PMOS transistors will be used in the first current copying block ETM1.

[0077] The current delivered by this first block ETM1 is copied into another current copying block ETM3 so as to draw a current i_pull to ground GND.

[0078] The first compensation means MCP1 also comprise a resistive voltage divider bridge RI, R2 having an input for receiving the supply voltage Vdd and an output (common node between the resistors RI and R2).

[0079] This voltage divider bridge is sized to deliver at its output a voltage equal to the reference voltage Vbg when the supply voltage Vdd present at its input has its nominal value VddTYP (1.8 volts for example).

[0080] The first compensation means MCP1 also comprise a second operational transconductance amplifier OTA2.

[0081] This second operational transconductance amplifier OTA2 has -a first input + connected to the output of the divider bridge RI, R2, -and an output looped back to a second input - of this second OTA2 transconductance operational amplifier.

[0082] Furthermore, the output of this second amplifier OTA2 is also connected to ground GND via a resistor R32.

[0083] The first compensation means MCP1 also comprise a second current copying block ETM2 connected to the output of the second transconductance operational amplifier OTA2 and having the same copying factor as that of the first block ETM1, a copying factor which is again adjustable from the first correction coefficient GFF.

[0084] This ETM2 block delivers the current i_push.

[0085] And, the MD node connected to the respective outputs of the two current copy blocks ETM1 and ETM2, delivers the first correction current ICorr which will be injected into the reference current source SCR.

[0086] This first correction current ICorr is equal to i_push- i_pull.

[0087] The first current copying means MIR1 copy the corrected current I REF_coRR to, with variable copying factors, deliver the first polarization currents Iref_i...iref_5.

[0088] Although this is not essential, it is preferable, in order to further reduce the sensitivity of the polarization currents to variations in the supply voltage, that the first current copying block ETM1 comprises two first cascoded current copying stages ETM10, ETM11 ([Fig.5]).

[0089] Similarly, it is particularly advantageous that the second current copying block also comprises two second cascoded current copying stages ETM20, ETM21.

[0090] Reference is now made more particularly to [Fig.6] to illustrate an example of a method for determining the value of the first correction coefficient GFF used in the integrated circuit as described above or in the method as described above.

[0091] This determination method comprises, after the manufacture of the integrated circuit, a measurement, in the laboratory for example, of the variation of the first physical parameter Gp (here the gain of the CHT transmission chain for example) caused by the dispersion of the supply voltage value, for different values ​​of the first correction coefficient.

[0092] Thus, in this example, it is considered that the supply voltage can have values ​​between 1.7 volts and 1.9 volts for a nominal voltage of 1.8 volts.

[0093] We will therefore measure (step ST60) the difference between the gain Gp of the transmission chain at 1.9 volts and the gain Gp of this transmission chain at 1.7 volts, and this for different values ​​3, 4, 5 and 6 of the first correction coefficient GFF.

[0094] We first note that when there is no compensation, we obtain a variation in dB of 2.16 which corresponds to a large variation or dispersion of the gain.

[0095] On the other hand, as soon as compensation is applied, this variation decreases.

[0096] This gain variation is for example equal to 0.83 for a GFF coefficient equal to 3.

[0097] It is equal to 0.57 dB for a GFF coefficient equal to 4.

[0098] It is equal to 0.33 dB for a GFF coefficient equal to 5.

[0099] It is equal to 0.05 dB for a GFF coefficient equal to 6.

[0100] We therefore see that we obtain a small variation in the gain for an equal GFF coefficient at 5 and a very low variation in gain for a GFF coefficient equal to 6.

[0101] It would therefore be possible to choose the value 6 for the GFF correction coefficient.

[0102] However, the inventor observed that in certain cases it was preferable to choose the value 5 for this GFF coefficient, that is to say to accept a gain variation a little higher than the minimum variation, which makes it possible to be less sensitive to dynamic variations (“ripple”) of the supply voltage.

[0103] As illustrated in [Fig.7], the integrated circuit IC may comprise both a radio frequency transmission chain CHT and a radio frequency reception chain CHR.

[0104] The reception chain CHR comprises second radiofrequency modules MD21-MD25 configured to be powered by the supply voltage Vdd and to be biased from respective second continuous bias currents 12^;.

[0105] The CHR reception chain then has a gain forming a second physical parameter whose value can also undergo a variation resulting from the dispersion Vdd voltage values.

[0106] The centralized CBIAS polarization means can then comprise - second compensation means MCP2, similar to the first compensation means MCP1, configured to perform open-loop compensation of the dispersion of the voltage values, this compensation comprising the development of a second corrected current 12^ COIT from the reference current IRFFO and a second correction coefficient G2FF determined from the variation in the value of the gain of the reception chain CHR, resulting from said dispersion of the voltage values.

[0107] The centralized polarization means then comprise second current copying means MIR2 configured to produce the second polarization currents I2RFF intended to polarize the modules of the reception chain CHR from the second corrected current I2RFF coir.

[0108] And, of course, here again, the GFF and G2FF correction coefficients can be programmable and adjusted.

[0109] In the embodiment which has just been described, the second compensation means MCP2 and the second current copying means MIR2 are distinct from the first compensation means MCP1 and the first current copying means MIR1.

[0110] Alternatively, since the transmission and reception chains operate with time division multiplexing, the second compensation means MCP2 and the second current copying means MIR2 could be combined with the first compensation means MCP1 and the first current copying means MIR1, to form only one and the same compensation means and one and the same current copying means, implemented sequentially by sequentially using the first correction coefficient GFF and the second coefficient G2FF, as a function of the time division multiplexing.

[0111] The invention makes it possible in particular to reduce the sensitivity to variations in the supply voltage, of the continuous polarization currents of the different modules of the integrated circuit while retaining, when necessary, small local LBS polarization structures ([Fig.2]) which makes it possible to reduce the size on silicon.

[0112] Programmable centralized compensation leads to improved performance compared to prior art solutions.

[0113] Furthermore, this centralized open-loop compensation also reduces the maximum current consumption when the supply voltage has its maximum value, which therefore leads to a reduction in the maximum power consumption.

Claims

Demands

1. Integrated circuit, comprising at least one first electronic module (MDi) configured to be powered by a supply voltage (Vdd) and to be biased from a first continuous bias current (IREFj), the supply voltage being able to have actual values ​​dispersed around a nominal value, said at least one first module having at least one first physical parameter (Gp) the value of which may undergo a variation resulting from said dispersion of voltage values, and centralized biasing means (CBIAS) comprising -first compensation means (MCP1) configured to perform open-loop compensation of said dispersion of voltage values ​​comprising the development of a first corrected current (IREF_coir) from a reference current (Irefo) and a first correction coefficient (GFF) determined from the variation of the value of said at least one first physical parameter,resulting from said dispersion of voltage values, and -the first current copying means (MIR1) configured to derive the first bias current (IREFj) from the first corrected current.

2. Integrated circuit according to claim 1, wherein the centralized biasing means (CBIAS) comprise a reference current source (SCR) configured to deliver the reference current and the first compensation means (MCP1) are configured to -determine a voltage difference (VFF) between the actual value of the supply voltage and its nominal value, -transform this voltage difference into a first correction current (Icorr) using the first correction coefficient and -inject this first correction current into the reference current source to obtain the first corrected current.

3. An integrated circuit according to claim 2, comprising a reference voltage source (BG) configured to deliver a reference voltage (Vbg), and wherein the first compensation means (MCP1) comprise a first transconductance operational amplifier (OTAI) having a first input connected to the reference voltage source and an output fed back to a second input of the first transconductance operational amplifier, -a first current feedback block (ETM1) connected to the output of the first transconductance operational amplifier and having a feedback factor (f(GFF)) adjustable according to the first correction coefficient (GFF), -a resistive voltage divider bridge (RI, R2) having an input to receive the supply voltage and an output, and sized to deliver on its output a voltage equal to the reference voltage when the supply voltage present on its input has its nominal value, -a second transconductance operational amplifier (0TA2) having a first input connected to the output of the divider bridge and an output fed back to a second input of the second transconductance operational amplifier,- a second current copying block (ETM2) connected to the output of the second transconductance operational amplifier and having said copying factor (f(GFF)) adjustable according to the first correction coefficient; - a node (ND) connected to the respective outputs of the two current copying blocks and capable of delivering the first correction current (I corr); - the reference current source (SCR) and the first current copying means (MIR1) being connected to said node (ND).

4. Integrated circuit according to claim 3, wherein the first current copying block comprises two first cascoded current copying stages (ETM10, ETM11) and the second current copying block comprises two second cascoded current copying stages (ETM20, ETM21).

5. Integrated circuit according to any one of the preceding claims, wherein the first correction parameter (GFF) is programmable.

6. Integrated circuit according to any one of the preceding claims, wherein the first physical parameter is the gain of the first module (MDi).

7. Integrated circuit according to any one of the preceding claims, comprising several first modules (MDi) configured to be powered by a supply voltage and to be biased from respective first DC bias currents, and the first current copying means are configured to elaborate these respective first bias currents from said corrected current.

8. Integrated circuit according to claim 7, wherein the first modules form a radio frequency transmission (CHT) or reception chain and the first physical parameter is the gain (Gp) of the transmission or reception chain.

9. Integrated circuit according to claim 8, further comprising several second electronic modules (MD21-MD25) configured to be powered by the supply voltage and to be biased from respective second continuous biasing currents (I2REF i), wherein the first modules form a radio frequency transmitting chain (CHT) and the second modules form a radio frequency receiving chain (CHR) having a gain forming a second physical parameter, the value of which may undergo variation resulting from said voltage value dispersion, and the centralized biasing means (CBIAS) include -second compensation means (MCP2), analogous to the first compensation means, configured to perform open-loop compensation of said voltage value dispersion, comprising the development of a second corrected current (12^,^) from the reference current and a second correction coefficient (G2FF) determined from the variation of the value of said second physical parameter, resulting from said voltage value dispersion, and -second current copying means (MIR2) configured to develop the second bias currents from the second corrected current.

10. A method for biasing at least one first electronic module from a first continuous bias current, said at least one first module (MDi) being supplied by a supply voltage (Vdd) having actual values ​​dispersed around a nominal value, said at least one first module having at least one first physical parameter whose value can undergo a variation resulting from said dispersion of voltage values, the method comprising - open-loop compensation of said dispersion of voltage values ​​comprising the development of a first corrected current (iref_coit) from a reference current and a first correction coefficient (Gff) determined from the variation of the value of said at least one first physical parameter, resulting from said dispersion of voltage values, and -a development of the first polarization current (Irefj) from the First corrected current.

11. A method according to claim 10, comprising the delivery of a reference current (IREfo) by a reference current source and wherein said compensation comprises - a determination of a voltage difference (VFF) between the actual value of the supply voltage and its nominal value, - a transformation of this voltage difference into a first correction current using the first correction coefficient and - an injection of this first correction current into the reference current source to obtain the first corrected current.

12. A method for determining the value of the first correction factor used in the integrated circuit according to any one of claims 1 to 9 or in the method according to claim 10 or 11, comprising, after the manufacture of the integrated circuit, a measurement (ST60) of the variation of the first physical parameter caused by the dispersion of the supply voltage values, for different values ​​of the first correction factor, and a selection of the value of the first correction factor corresponding to a desired value for the variation of the first physical parameter.

Images