Programmable clock generator for a radiation sensor

DE602021023006T2Active Publication Date: 2024-12-04LYNRED
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
DE602021023006
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Priority Date
2020-11-19
Filing Date
2021-10-19
Publication Date
2024-12-04
Estimated Expiration
2041-10-19

AI Technical Summary

Technical Problem

Conventional programmable phase generators for infrared detectors consume significantly more power than fixed phase generators, especially when dealing with multiple operating modes, due to their complex electronic components and high resource requirements.

Method used

A simplified programmable phase generator design that eliminates data memory, status registers, analog-digital converters, and wired operators, using a microcode interpreter with a minimal set of instructions (up to 7) to generate phase shapes, allowing for efficient control of phase outputs and synchronization, while maintaining low power consumption.

Benefits of technology

This approach reduces power consumption to levels comparable to fixed phase generators, making it advantageous for detectors with numerous operating modes, and allows for optimal phase generation considering parasitic elements and detector limitations.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader
Need to check novelty before this filing date? Find Prior Art

Description

DOMAINE DE L'INVENTION

[0001] The present invention relates to the field of the detection of electromagnetic radiation and in particular, to the detection of infrared radiation.

[0002] The invention relates more particularly to a programmable phase generator for such a detector, that is to say an electronic circuit making it possible to control the operation of the detector.

[0003] The invention finds a particularly advantageous application for controlling a detector comprising several distinct operating modes. ETAT ANTERIEUR DE LA TECHNIQUE

[0004] Detectors for thermal, or infrared, imaging are typically made as a matrix of elementary detectors, forming image points or pixels, in one or two dimensions. To ensure thermal insulation of the detectors, they are suspended above a substrate via support arms.

[0005] The substrate usually includes means for sequentially addressing the elementary detectors and means for electrical excitation and pre-processing of the electrical signals generated from these elementary detectors. This substrate and the integrated means are commonly referred to as the "reading circuit".

[0006] To control such a reading circuit, it is known to use a generator to provide the signals necessary for the operation of the circuit. These signals are called "phases". Conventionally, the phases are used to control switches present at the level of the reading circuit and each elementary detector. For example, these switches can control the exchange of information between a photodiode and a reading circuit, the exchange of data between different capacitors and amplifiers, as well as the electrical transmissions between the elementary detectors.

[0007] With these phases, an infrared detector can be controlled to integrate different gains or to perform complex functions, such as obtaining a bi-spectral image.

[0008] For the purposes of the invention, the "gain" corresponds to a conversion factor between a photon and a voltage. This conversion factor depends on the integration carried out, for example the integration capacity used, and the dynamics of the phases.

[0009] To achieve complex functions, document WO2019 / 166465 illustrates, for example, the shape of the phases expected to obtain an operating mode of an infrared detector in which the detector has a high reading dynamic range. American patent application US 2014 / 247382 also describes a programmable phase generator for driving a pixel array.

[0010] The phases generated by a phase generator therefore make it possible to place an infrared detector in a specific operating mode by controlling the integration time, activation and / or inhibition of certain components of the reading circuit or of the elementary detectors.

[0011] A phase generator is classically produced, based on phase shapes defined before the infrared detector is produced, by implementing logic functions to create the predefined phase shapes. This type of phase generator is therefore fixed or " hardcoded » in the Anglo-Saxon literature. Of course, these phases can vary depending on a control signal to activate one or other of the operating modes of the infrared detector.

[0012] To satisfy user needs as much as possible, manufacturers are integrating more and more functions into the reading circuits and elementary detectors, so that it becomes possible to obtain the same gain by generating distinct phases, that is to say by modifying the sequencing and / or the binary value of the phases. Thus, after the implementation of the infrared detector, it is often observed that the shape of the phases allowing to obtain a specific gain is not optimal due to the parasitic elements and the limitations of the infrared detector, observable only after its realization.

[0013] Typically, parasitic elements may appear due to manufacturing tolerances of the capacitors made at the elementary detectors. In addition, if several capacitors are made at each elementary detector so as to obtain distinct gains, it can be investigated which capacitors must be activated to obtain each gain optimally following the measurement of the actual values ​​of the capacitors made.

[0014] In the most recent infrared detectors, it is not uncommon to integrate more than 15 operating modes, for example 18 operating modes. With such a number of operating modes, the implementation of a fixed phase generator becomes particularly complex, because it must integrate all the electronic components needed to generate the shape of each phase of each operating mode. In addition, with the multiplicity of electronic components, the phase generator generates a consumption much higher than the consumption of a fixed phase generator of an infrared detector with only a few operating modes.

[0015] To partially address these problems and enable the generation of optimal phases, it is possible to use a programmable phase generator. Such a generator is similar to a microprocessor; it integrates a program memory in which the instructions to be carried out to generate the phases are encoded. From this program memory, a processor interprets the instructions encoded in the program memory. Typically, a microcontroller also integrates a data memory, a working or status register, analog-to-digital and digital-to-analog converters, and hard-wired operators for performing arithmetic operations, conditional operations, and speculative code executions.

[0016] The major drawback of these programmable phase generators lies in their consumption, typically at least ten times higher than that of a fixed phase generator.

[0017] The technical problem that the invention aims to solve is therefore to obtain a programmable phase generator, that is to say one that makes it possible to control the production of several distinct phases according to a program code, with consumption comparable to that of a fixed phase generator. STATEMENT OF THE INVENTION

[0018] To address this technical problem, the invention proposes a programmable phase generator without data memory, working or state register, analog-to-digital and digital-to-analog converter, and hard-wired operator allowing arithmetic operations, conditional operations and speculative executions of the code to be performed.

[0019] Indeed, the invention is the result of a search for the simplest instructions for generating the specific phases of an infrared detector. It was observed that with a maximum of seven instructions, it is possible to generate all the forms of phases conventionally used in this specific field.

[0020] Thus, it is possible to form a very simple programmable phase generator, that is, capable of interpreting only instructions coded on 3 bits.

[0021] To this end, the invention relates to a phase generator of a detector, the generator integrating at least one elementary machine for interpreting a microcode stored in a register, each elementary machine comprising: ▪ at least one control input comprising a logic level change detector; ▪ at least one phase output comprising a controlled switch, making it possible to define the logic level of the phase output and a controlled inverter making it possible to invert the logic level of the phase output; ▪ at least one clock signal associated with a counter; ▪ a module for loading the instructions and arguments stored in the register, the instructions being coded on 3 bits and making it possible to carry out at least: an application of a logic level on at least one phase output; an inversion of the logic level of at least one phase output; a synchronization with a logic change of a control input; a wait of a predetermined duration; and a restart of a predetermined sequence; ▪ a member for inhibiting the loading module;and ▪ a module for executing the instructions loaded by the loading module, the execution module being configured to: control the position of at least one controlled switch following an application instruction; control the activation of at least one controlled inverter following an inversion instruction; detect a logical change following a synchronization instruction; control the inhibition of the inhibition member for a duration measured by the counter following a wait instruction; and control the position of a pointer of the loading module following a restart instruction.

[0022] The invention allows, with these few simple and easy-to-implement instructions, to program the phases of a detector, for example an infrared detector. In other words, the search for the simplest instructions for generating the specific phases of a detector has allowed a drastic reduction in the possibilities of a conventional programmable phase generator to limit its complexity.

[0023] The invention thus makes it possible to obtain a programmable phase generator with a consumption comparable to that of a fixed phase generator. The greater the number of operating modes of the detector, the more advantageous the use of the programmable phase generator of the invention becomes compared to a fixed phase generator. Typically, for a detector having more than 15 operating modes, for example 18 operating modes, the programmable phase generator of the invention is more advantageous than a fixed phase generator because they both have equivalent consumption. However, the programmable phase generator makes it possible to obtain optimal phases by taking into account the parasitic elements and the limitations of the infrared detector, observable after its production.

[0024] The minimal instructions identified to form the different classical phases of a detector are: ▪ applying a logic level to at least one phase output; ▪ inverting the logic level of at least one phase output; ▪ synchronizing with at least one control input; ▪ waiting for a predetermined duration; and ▪ restarting a predetermined sequence.

[0025] The application of a logic level is implemented by a simple controlled switch associated with each output.

[0026] To do this, each controlled switch is wired between an output and logic level “1” or logic level “0” depending on the command applied to the controlled switch.

[0027] Similarly, a simple controlled inverter is placed at each output to allow the logic level of an output to be inverted.

[0028] To synchronize the execution of the microcode with the reception of an external signal, a detector is used to detect a logical change of a control input.

[0029] While waiting for this synchronization signal, the operation of the loading module must be inhibited by the inhibition device. To do this, it is possible to use an inhibition signal generated by the execution module.

[0030] Preferably, said inhibiting member of the loading module is interposed between a clock signal of the elementary machine and a clock signal transmitted to the loading module. Thus, it is possible to easily inhibit the operation of the loading module by blocking its clock signal used to regulate its operations.

[0031] The wait instruction also requires inhibiting the operation of the load module while the execution module uses the counter to reach the predetermined wait time. Preferably, the instructions perform: ▪ a short wait; and ▪ a long wait; the length of the long wait argument being greater than the length of the short wait argument.

[0032] Preferably, the short wait argument is coded on 4 bits. This short wait thus makes it possible to represent the operation of all the phases with high variation with the optimum number of bits to limit the resources of the elementary machine.

[0033] Preferably, the long wait argument is coded on 14 bits. This long wait makes it possible to represent the operation of all low variation phases with the optimum number of bits to take into account the maximum variation duration observed on the phases of a detector.

[0034] In addition to these instructions, another instruction preferentially allows the elementary machine to be placed in a low-power mode, in which the execution module inhibits the operation of several components of the elementary machine.

[0035] Furthermore, it is possible to generate several phases with a single elementary machine or several juxtaposed elementary machines. For example, the phase generator may comprise at least two elementary machines. This embodiment makes it possible to increase the number of phases that can be generated simultaneously.

[0036] Preferably, the phase generator comprises a programmable multiplexer arranged at the output of the elementary machine. This embodiment also makes it easier to encode an operating mode of the detector by modifying the routing of the outputs of said elementary machine. BRIEF DESCRIPTION OF THE FIGURES

[0037] The invention will be clearly understood on reading the following description, the details of which are given solely by way of example, and developed in relation to the appended figures, in which identical references refer to identical elements: There figure 1 illustrates a programming file and the associated binary code of a register of an elementary machine of a programmable phase generator according to an embodiment of the invention; The figure 2 illustrates an elementary machine of a programmable phase generator according to an embodiment of the invention; The figure 3 illustrates a loading module of the elementary machine of the figure 2 ; There figure 4 illustrates the sequencing of operations between the loading module and an execution module of the elementary machine of the figure 2 ; There figure 5 illustrates a programmable phase generator integrating two elementary machines of the figure 2 ; and The figure 6 illustrates an example of phases generated by the programmable phase generator of the figure 5 . DETAILED DESCRIPTION OF THE INVENTION

[0038] The programmable phase generator 10 of the invention is capable of generating several phase shapes according to a microcode.

[0039] This microcode is formed by a set of bits stored in a register 12 of an elementary machine 14 of the programmable phase generator 10. In the following description, a “computer function” corresponds to a command of a computer programming language making it possible to formulate an algorithm. Furthermore, an “instruction” corresponds to a physical implementation of a computer function in the register 12 of an elementary machine 14. All of these instructions stored in the register 12 form the “microcode”.

[0040] To obtain this microcode, a programming machine 28 transforms a programming file 11 into microcode, as illustrated in the figure 1 . The programming file 11 can be coded with a programming language specifically adapted to the programmable phase generator 10 or by using a generic programming language, such as assembly language or C language. Of course, the functions conventionally used by these languages ​​cannot all be implemented in the register 12, because at most seven Inst instructions can be recognized by the programming machine 28 to be integrated into the register 12. Indeed, to limit the complexity and consumption of the elementary machine 14, the Inst instructions are coded on 3 bits, so that only 7 computer functions can be recognized by the programming machine 28.

[0041] In the example of the figure 1 , a typical programming file 11 for phase generation is described. In this programming file 11, a first computer function Set_GPO_Init makes it possible to apply a logic level to the different phase outputs INT, XRINT1, XRINT2 of the programmable phase generator 10. This is an initialization instruction for the programmable phase generator 10. When transforming the programming file 11 into the register 12, this first computer function Set_GPO_Init is coded 000 and the argument of this computer function Set_GPO_Init is placed after code 000 in register 12.

[0042] For the purposes of the invention, “argument” means the variable of a computer function.

[0043] The second computer function Set_Remap of the programming file 11 specifies a position of a programmable multiplexer 26 of the programmable phase generator 10.

[0044] The third computer function Seq in programming file 11 indicates a tag from which the program will be executed in a loop. This tag therefore makes it possible to program an infinite loop between the computer function Restart and the computer function Seq.

[0045] The fourth computer function Thread of the programming file 11 allows you to specify the elementary machine 14 on which the microcode must be executed.

[0046] Using the argument 0 after the computer function Thread, the first elementary machine 14 is targeted for the loop described between the computer function Seq and the computer function Restart. In the example of the figure 1 , a single elementary machine 14 is programmed. Alternatively, several elementary machines 14 can be implemented in the programmable phase generator 10 and the programming file 11 can integrate computer functions for several distinct elementary machines 14 using several computer functions Thread with distinct arguments.

[0047] The first line of the infinite loop contains the synchronization computer function Sync allowing synchronization with at least one control input GPI_0, GPI_1 of the elementary machine 14. More specifically, this instruction is coded 001 and aims to detect a rising edge on the internal signal IntOR of the elementary machine 14. The argument of this function is 00 and it allows to define on which internal signal this function must be synchronized. Thus, to implement this computer function Sync IntOR, register 12 integrates a second line with instruction 001, corresponding to a computer synchronization function Sync, and with the argument 00 corresponding to a detection of a rising edge on the internal signal IntOR. The following bits are represented by an X because they are random and do not influence the operation of the elementary machine 14.

[0048] The third instruction coded in register 12 concerns the computer function of waiting Wait of a predetermined duration whose argument is 8. In the example of the figure 1 , this wait instruction is coded 010 in register 12 and the argument 8 is coded on four bits with a binary value of 1000. In the example of the figure 1 , only one type of wait instruction is described, and is coded on four bits. Alternatively, two wait instructions can be implemented: a short wait instruction and a long wait instruction.

[0049] The short wait instruction then presents a 4-bit argument, as shown in the figure 1 whereas the long wait instruction may include an argument encoded on a larger number of bits, for example 14 bits, to describe the variation of very little used phases.

[0050] Following this wait instruction, the infinite loop includes a computer reversal function Toggle of the logic level of at least one phase output GPO_0-GPO_9.

[0051] This computer function of inversion Toggle is used with a 5-bit argument to specify one or more phase outputs GPO_0-GPO_9 to reverse. This computer function of reversal Toggle is encoded as instruction 100 in the register 12 and, following this code of instruction Inst, the register 12 presents the value of the argument 00001.

[0052] A new computer synchronization function Sync is then encoded in the register 12. This computer synchronization function Sync is similar to the one embedded in the second line of the register 12 and it is also followed by a computer waiting function Wait also similar to the one previously used.

[0053] Following this second waiting computer function Wait, a new computer function of inversion Toggle is used. This second computer function of inversion Toggle presents a different argument from the first computer inverse function Toggle, since the argument of the second inverse computer function Toggle is 00010.

[0054] A new computer synchronization function Sync is then programmed with an argument to detect a rising edge on the internal signal Int0F . This detection of a rising edge on the internal signal Int0F is coded 01. Thus, the register line 12 associated with this computer synchronization function Sync presents bits 00101, with code 001 to describe the synchronization instruction and argument 01 to describe the argument of this instruction.

[0055] Following this computer synchronization function Sync, a new waiting computer function Wait is coded with a wait time of 12. Thus, the argument of this computer synchronization function Sync is coded 1100. The last instruction of the infinite loop corresponds to a new computer inverse function Toggle with the binary code 0001 as argument.

[0056] Finally, the computer restart function Restart is coded in the register 12 with code 111. This instruction 111 has the argument 0010 which refers to the second line of the register 12, because the computer restart function Restart echoes the tag Seq placed after the IT function Set_GPO_Init.

[0057] To interpret these different instructions with their respective arguments, an elementary machine 14 can be developed based on the model of the figure 2 . This elementary machine 14 has two control inputs GPI_0, GPI_1 allowing the operation of the elementary machine to be controlled during a synchronization instruction. To do this, each control input has a detector 19 allowing to detect a logical change, i.e. a rising edge or a falling edge on each control input GPI_0, GPI_1. At the detector output 19associated with each order entry GPI_0, GPI_1, the elementary machine 14 has internal signals indicating the detection of rising edges Int0F And IntIF or falling edges IntOR And IntOR of the two control inputs GPI_0, GPI_1. The elementary machine 14 also features ten phase outputs GPO_0-GPO_9 intended to control a detector. To implement the application instructions, each phase output GPO_0-GP0_9 has a controlled switch 20. Each controlled switch 20 is wired between an output GPO _ 0-GPO_9 and the high logic level, called Vhigh, or the low logic level, called Vlow, depending on the order Cmd1 applied to the controlled switch 20. In addition to this controlled switch 20, each exit GPO _ 0-GPO_9 also includes a logic inverter 21,allowing the instruction to invert the logic level of the output to be applied GPO_0-GPO_9 when applying a control signal Cmd2.

[0058] Furthermore, the elementary machine 14 also has an entrance MCK for a clock signal allowing the timing of the execution of instructions. This clock signal MCK is connected to an inhibition organ 17 and a counter 18.

[0059] The elementary machine 14 has the function of decoding the microcode loaded into the register 12. To do this, the elementary machine 14 includes a loading module 15 configured to detect the instruction Inst to be carried out as well as its argument Arg. This loading module 15 is connected to an execution module 16 configured to execute instructions Inst transmitted by the loading module15. The loading module 15 is more particularly detailed on the figure 3 . This loading module 15 includes an instruction decoder 23 connected to an argument decoder 25. For each line of the register 12, the loading module 15 uses a pointer Point to extract a particular line.

[0060] In the example of the figure 3 , a line is copied into an internal register 22 to make it easier to understand how to extract the useful bits. Preferably, extracting the data from the register 12 can be done directly on the register 12 without requiring a copy into an internal register 22.

[0061] From this extracted line, the first three bits are passed to an instruction register in the instruction decoder 23.Again, the instruction register, like the internal register 22, aims to illustrate the process of extracting useful bits and is not necessary in practice.

[0062] When the bits corresponding to the instruction are extracted from the register 12, they can be directly passed to the execution module 16 by the signal Inst. Additionally, these bits are also directed to a selector 24 whose inputs correspond to the different possible argument sizes of the different instructions L_arg0 , L_arg1 , L_arg2 ... L_argX. So, depending on the detected instruction, the size of its argument L_arg is delivered at the exit of the selector 24. This argument size L_arg is passed to the argument decoder 25which copies the number of bits corresponding to the size of the argument into an argument register. Again, the argument register, like the internal register 22, is intended to illustrate the process of extracting useful bits and is not necessary in practice. The bits corresponding to the argument can then be passed to the execution module 16 by the signal Arg.

[0063] As illustrated on the figure 3 , the loading module 15 is preferentially clocked by a clock signal Clk1 . To do this, the instruction decoder 23 can be configured to perform decoding on each rising edge of the clock signal Clk1 . Inside the instruction decoder 23, the selector 24 can also be inhibited until an end of execution signal is obtained FinExec of the execution module 16.

[0064] To execute the various instructions, the execution module 16 therefore receives the instruction Inst and the argument Arg of the function to be performed.

[0065] When the instruction Inst to be executed corresponds to a computer application function of a logical level Set_GPO_Init, the execution module 16 directly controls the controlled switches 20 by means of the signal Cmd1 depending on the argument Arg transmitted by the loading module 15. Similarly, to execute a reverse instruction, the execution module 16 controls the inverters 21 by means of the signal Cmd2 depending on the argument Arg transmitted.

[0066] These apply and reverse functions can be performed synchronously with the decoding of instructions performed by the load module 15.For other standby or synchronization functions, it is necessary to inhibit the operation of the loading module 15 for a duration. To do this, the clock signal Clk1 transmitted to the loading module is preferentially generated by the inhibition module 17 controlled by the control signal Cmd3 of the execution module 16. So, for a synchronization instruction, the execution module 16 waits for the detection of the programmed synchronization signal before lifting the inhibition of the inhibition module 17 and allow the loading module 15 to continue loading a new instruction Inst and his argument Arg. Similarly, the inhibition module 17 is also used during a wait instruction. In this case, the execution module 16 use the meter 18to count a waiting time corresponding to the argument Arg programmed before lifting the inhibition of the inhibition module 17.

[0067] Instead of using an inhibition module 17 directly positioned on the clock signal of the loading module 15, an end of execution signal FinExec can be transmitted to the loading module 15 by the execution module 16, when execution is complete and the execution module 16 can receive a new instruction Inst.

[0068] Additionally, the wait instruction can be coded as two separate instructions: a long wait and a short wait with arguments of different lengths. In both cases, the execution module 16 is able to interpret these two expectations by means of the meter 18.

[0069] As for the program restart instruction, this instruction only aims to change the position of the pointer Point of the loading module 15. To do this, the execution module 16 uses the control signal Cmd4 to move the pointer Point on the position corresponding to the argument Arg program.

[0070] In addition to these six instructions, it is possible to implement a seventh instruction with a 3-bit coding for these different instructions. Preferably, the seventh instruction corresponds to the placement of the elementary machine 14 in a low power consumption mode. In this mode, the execution module can command the deactivation of part of the electronic components of the elementary machine 14 to limit its consumption.

[0071] As illustrated on the figure 4 , the relationships between the loading module15 and the execution module 16 are sometimes regular and sometimes irregular. For apply or reverse instructions, on the first clock cycle, the load module 15 proceeds to read the instruction Inst, then passes the argument Arg of this instruction Inst to the execution module 16. In the second clock cycle, the loading module 15 loads the second instruction Inst while the execution module 16 executes the first instruction Inst.

[0072] The second instruction requires an inhibition of the loading module 15, for example for synchronizing a signal on a control input GPI_0, GPI_1 of the elementary machine 14. So, during the third clock cycle, the loading module 15 proceeds to read the third instruction Inst, while the execution module 16 executes the second instruction while inhibiting the operation of the load module 15. When the second instruction is executed, for example when the signal from a detector 19 is received by the execution module 16, the operation of the loading module 15 is reactivated, and it can load a new instruction Inst while the execution module 16 executes the instruction Inst previously loaded.

[0073] This operation between the loading module 15 and the execution module 16 allows you to quickly and efficiently generate the different phases of a detector 10. Thus, it is possible to form a phase generator 10 of a detector with a single elementary machine 14.

[0074] To improve the number of phases that can be generated, the phase generator 10 may comprise several elementary machines 14 juxtaposed, as illustrated in the figure 5 .

[0075] In the example of the figure 5 , two elementary machines 14 are used and each elementary machine 14 integrates its own register 12. In addition, the outputs of the elementary machines 14 are introduced into a programmable multiplexer 26 including a register 27 allows you to specify the routing topology.

[0076] From the programming file 11, the programming machine 28 can thus program both registers 12 of the two elementary machines 14 as well as the register 27 of the programmable multiplexer 26.

[0077] In the example of the figure 5 , order entries GPI_0, GPI_1 of the two elementary machines 14 are associated with signals from a detector FromROIC, while the outputs of the programmable multiplexer 26 form the phases INT, XRINT1, XRINT2 allowing the operation of the detector to be controlled ToROIC. Alternatively, it is possible to use one or more control inputs GPI_0, GPI_1 of an elementary machine 16 to control the phase generation mode using a computer synchronization function on a control input GPI_0, GPI_1.

[0078] There figure 6 illustrates an example of characteristic phases of a detector 10 generated from the programming file 11 of the figure 1 and implemented by an elementary machine 14. As illustrated on the figure 6 , a clock signal MCK is used to clock the generation of the different phases INT, XRINT1, XRINT2 and the transitions between these phases INT, XRINT1, XRINT2 appear with the waiting times, 8 and 12, coded in the programming file 11.

[0079] The invention thus makes it possible to generate the phases INT, XRINT1, XRINT2 of a detector, for example an infrared detector. By reprogramming the phase generator 10, The phase shape can be determined after the detector has been produced, taking into account its technical limitations. Thus, the phase shape can be optimal depending on the desired operating mode.

[0080] In addition, the simplicity of the components used to implement this phase generator 10 allows to limit its consumption, so that the consumption of the phase generator 10 programmable is similar to that of a fixed phase generator integrating more than 15 operating modes.

Claims

1. A generator of phases (10) of an electromagnetic radiations detector, the generator integrating at least one elementary machine (14) for interpreting a microcode stored in a register (12), each elementary machine (14) comprising: ▪ at least one control input (GPI_0, GPI_1) comprising a logic level change detector (19); ▪ at least one phase output (GPO_0-GPO_9) comprising a controlled switch (20), enabling to define the logic level of the phase output (GPO_0-GPO_9), and a controlled inverter (21) enabling to toggle the logic level of the phase output (GPO_0-GPO_9); ▪ at least one clock signal (MCK, Clk1) associated with a counter (18); ▪ a unit for loading (15) the instructions (Inst) and the arguments (Arg) stored in the register (12), the instructions (Inst) being coded over 3 bits and enabling to implement at least: - an application (Set_GPO_Init) of a logic level onto at least one phase output (GPO_0-GPO_9); - a toggling (Toggle) of the logic level of at least one phase output (GPO_0-GPO_9); - a synchronization (Sync) with a logic change of a control input (GPI_0, GPI_1); - a wait (Wait) for a predetermined duration; and - a restarting (Restart) of a predetermined sequence; ▪ means (17) for inhibiting the loading unit (15); and ▪ a unit (16) for executing the instructions (Inst) loaded by the loading unit (15), the execution unit (16) being configured to: - control (Cmd1) the position of at least one controlled switch (20) after an instruction (Inst) of application (Set_GPO_Init); - control (Cmd2) the activation of at least one controlled inverter (21) after an instruction (Inst) to toggle (Toggle); - detect (IntOR, IntOF, Int1R, Int1F) a logic change after an instruction (Inst) of synchronization (Sync); - control (Cmd3) the inhibition by the inhibition means (17) for a duration measured by the counter (18) after a wait instruction (Wait); and - control (Cmd4) the position of a pointer (Point) of the loading unit (15) after a restart instruction (Restart).

2. The phase generator according to claim 1, wherein said means (17) for inhibiting the loading unit (15) are interposed between a clock signal (MCK) of the elementary machine (14) and a clock signal (Clk1) transmitted to the loading unit (15).

3. The phase generator according to claim 1 or 2, wherein the instructions (Inst) enable to implement: ▪ a short wait; and ▪ a long wait; the length of the argument of the long wait being greater than the length of the argument of the short wait.

4. The phase generator according to claim 3, wherein the argument (Arg) of the short wait is coded over 4 bits.

5. The phase generator according to claim 3 or 4, wherein the argument (Arg) of the long wait is coded over 14 bits.

6. The phase generator according to any of claims 1 to 5, wherein an instruction (Inst) enables to place the elementary machine (14) in a low power consumption mode, where the execution unit (16) inhibits the operation of a plurality of components of the elementary machine (14).

7. The phase generator according to any of claims 1 to 6, wherein the execution module (16) executes an instruction (Inst) while the loading unit (15) loads the next instruction (Inst) to be executed by the execution unit (16).

8. The phase generator according to any of claims 1 to 7, wherein the phase generator (10) comprises at least two elementary machines (14).

9. The phase generator according to any of claims 1 to 7, wherein the phase generator (10) comprises a programmable multiplexer (25) arranged at the output of said at least one elementary machine (14).