METHOD FOR PRODUCING COMPRESSED DATA AVAILABLE BY AN ELECTRIC COUNTER

Description

[0001] The present invention relates to a compression of measurement data by an electric meter, said data being typically intended to be transmitted by the electric meter to a data concentrator by powerline communications or by other means.

[0002] In an AMM (Automated Meter Management) type power supply network, communications are established between electricity meters, known as smart meters, and a data concentrator, sometimes called nœud de base (The "base node" in English) is responsible for collecting measurement data from electricity meters, typically the energy consumed by electrical installations monitored by those meters. Communication between the electricity meters and the data concentrator relies on power line communication.

[0003] Modern electricity meters also typically include a customer tele-information interface (TIC) for retrieving measurement data. Data exchange via the TIC interface is conducted using an ASK (Amplitude Shift Keying) modulated signal. The Wikipedia article "smart meter," by an anonymous author and published on October 27, 2010, describes how an electricity meter makes measurement data available. This data originates from measurements taken by the meter on the energy consumption of an electrical installation monitored by the meter.

[0004] In each of the cases mentioned above, a large amount of data may need to be transferred by the electricity meter. Furthermore, in order to be transmitted to the data concentrator via power line communication or to a device connected to the ICT interface, this data must be stored.

[0005] It is therefore desirable to find a solution that can compress this data for storage and / or transmission, in order to conserve memory resources and / or accelerate transfer speeds. The review article entitled "Data compression: the complete reference, 3rd edition" by David Salomon, published on January 1, 2004, describes examples of data compression algorithms, particularly prefix codes.

[0006] It is also desirable to provide a solution that is simple to implement.

[0007] The invention relates to a method of making available by an electric meter measurement data, said data being derived from energy quantity measurements carried out by said electric meter on an energy consumption of an electrical installation supervised by said electric meter, the electric meter retrieving the measurement data to be made available.The process is such that the electric meter further performs the following steps: classify each retrieved data, according to a minimum number of bits to represent said data; determine, for each class of a predefined plurality of classes, a number of bits allocated to represent each data classified in said class, according to the minimum numbers of bits to respectively represent the data classified in said class; compress each retrieved measurement data, representing said retrieved data on a number of bits equal to the number of bits allocated to the class of said measurement data and inserting before the compressed data a predefined prefix associated with said class; and, include in a header associated with the compressed data, for each class, information representing the number of bits allocated to represent each measurement data classified in said class.The electricity meter performs the following steps: it determines the most frequent value among the retrieved measurement data; and replaces the measurement data, with a value equal to the determined most frequent value, with a single bit. This significantly reduces the resources required to make the measurement data available.

[0008] In one particular embodiment, where the measurement data is represented in absolute form, the electricity meter performs a preliminary step of converting the measurement data from absolute form to relative form. This significantly reduces the resources required to make the measurement data available.

[0009] In one particular embodiment, the electricity meter uses a truncated binary code to identify measurement data with the most frequent value and prefixes. Thus, it is possible to omit prefixes for measurement data whose value is equal to the determined most frequent value.

[0010] In one particular embodiment, where the measurement data is associated with at least one timestamp representing an absolute instant, the electricity meter performs a step of converting this timestamp into information relative to a predefined date. This significantly reduces the resources required to provide such timestamp information.

[0011] In one particular embodiment, where each measurement data point is associated with a timestamp, and the measurements taken by the electricity meter are periodic, the electricity meter only stores a timestamp and includes in its header information representing the periodicity of said measurements. Thus, the resources required to provide such timestamp information are significantly reduced.

[0012] According to a particular embodiment, to indicate a change in the time reference among the measurement data, the electricity meter performs the following steps: determine an index for each measurement data item for which the change in the time reference has occurred; determine a time offset due to the change in the time reference for said measurement data item whose index has been thus determined; and include in the header information representing said index and said time offset. Thus, the resources required to make such timestamp information available are significantly reduced, despite the change in the time reference ( e . g (daylight saving time / winter time change).

[0013] In one particular embodiment, the electricity meter, considering the measurement data as unsigned, performs the following steps: determine a minimum number of bits to represent the highest value among the measurement data; assign this minimum number of bits to a first class; test, for each other class in the predefined plurality, a value of bits assigned to represent the data of that class, and determine a resulting size of the measurement data after compression; and assign to each other class in the predefined plurality a number of bits that results in the smallest resulting size of the measurement data after compression. Thus, the dynamic determination of the bits assigned to the classes is simple to implement.

[0014] According to a particular embodiment, where the measurement data is signed, the electricity meter performs the following steps: determine a first minimum number of bits to represent the highest absolute value among the positive measurement data; determine a second minimum number of bits to represent the highest absolute value among the negative measurement data; assign said first minimum number of bits determined to a first class; assign said second minimum number of bits determined to a second class; test, for each other class of said predefined plurality, a value of bits assigned to represent the data of said class, and determine a resulting size of the measurement data after compression; assign to each other class of said predefined plurality a number of bits allowing to obtain the smallest resulting size of the measurement data, after compression;and include in the header information indicating which class(es) concern positive measurement data and which class(es) concern negative measurement data. Thus, the compression of measurement data is more efficient when the measurement data is signed.

[0015] In one particular embodiment, where the measurement data is signed, the electricity meter first performs the following steps: determining a calibration value equal to the highest absolute value among the negative measurement data; adding the calibration value to each measurement data point; and including information representing the calibration value in the header. In this way, the measurement data is transformed into positive measurement data (equivalent to unsigned data), which simplifies and increases the efficiency of compression when dealing with signed measurement data.

[0016] In one particular embodiment, to make the measurement data available, the electricity meter, expecting to receive a predefined password, performs the following steps: receiving a password; determining the inverse of the received password; making the measurement data available in compressed form when the inverse of the received password matches the predefined password; and making the measurement data available in uncompressed form when the received password matches the predefined password. Thus, only one password needs to be configured, and support for off-the-shelf devices is ensured.

[0017] The invention also relates to an electric meter adapted to make available measurement data, said data being derived from energy quantity measurements carried out by said electric meter on an energy consumption of an electrical installation supervised by said electric meter, the electric meter comprising means for retrieving the measurement data to be made available.The electric meter is such that it further comprises: means for classifying each retrieved measurement data, according to a minimum number of bits to represent said data; means for determining, for each class of a predefined plurality of classes, a number of bits allocated to represent each data classified in said class, according to the minimum numbers of bits to respectively represent the data classified in said class; means for compressing each retrieved measurement data, by representing said retrieved measurement data on a number of bits equal to the number of bits allocated to the class of said data and by inserting before the compressed data a predefined prefix associated with said class; and means for including in a header associated with the compressed data, for each class, information representing the number of bits allocated to represent each data classified in said class.The electric meter also includes: means for determining the most recurring value among the retrieved measurement data; and means for replacing the measurement data, with a value equal to the most recurring value determined, by a single bit.

[0018] The invention also relates to a computer program that can be stored on a medium and / or downloaded from a communication network for reading by a processor. This computer program includes instructions for implementing the aforementioned method when the program is executed by the processor. The invention further relates to storage means comprising such a computer program.

[0019] The features of the invention mentioned above, as well as others, will become clearer upon reading the following description of an exemplary embodiment, said description being made in relation to the accompanying drawings, among which: there Fig. 1 schematically illustrates an electrical system in which the invention can be implemented; the Fig. 2 schematically illustrates an example of all or part of the hardware architecture of an electrical system device, such as an electricity meter; the Fig. 3 This schematically illustrates an algorithm, implemented by an electricity meter, for processing a request to send measurement data; Fig. 4 schematically illustrates an algorithm, implemented by the electricity meter, for compressing measurement data; the Fig. 5 schematically illustrates an algorithm, implemented by the electric meter, to determine whether the electric meter should send the data in compressed or uncompressed form.

[0020] The invention is described in detail below in the context of transferring measurement data from an electricity meter to a data concentrator via power line communication. The principles of data compression for the electricity meter are equally applicable both when storing said data by the electricity meter and when transferring said data via a customer tele-information interface. The same principles of data compression for the electricity meter are also equally applicable when transferring measurement data from an electricity meter to a data concentrator via another medium, such as GPRS (General Packet Radio Service). In general, the compression principles described herein relate to the provision of measurement data by the electricity meter.

[0021] There Fig. 1 schematically illustrates an electrical system in which the invention can be implemented.

[0022] The electrical system of the Fig. 1 includes a 100 electrical supply network consisting of a plurality of wires, one wire being dedicated to neutral and at least one wire being dedicated to at least one phase. In the embodiment example represented by the Fig. 1 The electrical supply network 100 is a three-phase system, and therefore consists of four wires: a neutral wire 104, a wire of a first phase 101, a wire of a second phase 102 and a wire of a third phase 103. The separation of the phases is therefore 120 degrees between the two phases, and would be 180 degrees in the case of a two-phase system, since in polyphase electrical networks, the phases are generally at equal angular distance.

[0023] The electrical system of the Fig. 1 It also includes an electricity meter 120, that is, a device used to measure the amount of electrical energy consumed by an electrical installation that the electricity meter 120 is responsible for monitoring. The electricity meter 120 is connected to one of the phase wires of the electrical supply network 100, as well as to the neutral wire, by respective links 122 and 121. In this case, we are dealing with a single-phase electricity meter. The electrical system of the Fig. 1 could be based on the implementation of a polyphase electricity meter.

[0024] The electrical system of the Fig. 1 It also includes a data concentrator 110 responsible for collecting measurement data, these measurements being carried out by the electricity meters connected to the power supply network 100. The data concentrator 110 is connected to one of the phase wires of the power supply network 100 as well as to the neutral wire, by respective links 112 and 111. This can be seen in the example of the Fig. 1 The data concentrator 110 and the electricity meter 120 are connected to the power supply network 100 via separate phases. The data concentrator 110 and the electricity meter 120 could be connected to the power supply network 100 via the same phase.

[0025] The exchanges between the 120 electricity meter and the 110 data concentrator are carried out via powerline communications. For example, these exchanges are carried out according to the PRIME specifications (" PoweRline Intelligent Metering Evolution (in English). These exchanges allow the data concentrator 110 to request, from the electric meter 120, a transmission of measurement data corresponding to measurements, made by the electric meter 120, of energy consumption of the electrical installation supervised by the electric meter 120.

[0026] There Fig. 2 schematically illustrates an example of hardware architecture of at least part of the 120 electric meter and / or the 110 data concentrator.

[0027] Let's consider the Fig. 2 within the framework of the electric meter 120. The electric meter 120 then comprises, connected by a communication bus 220: a processor or CPU (“Central Processing Unit”) 210; a RAM (“Random Access Memory”) 211; a ROM (“Read Only Memory”) 212; a storage unit or a storage media reader, such as an SD card reader (“Secure Digital”) 213; a set 214 of connection interfaces allowing the electric meter 120 to be connected to the power supply network 100.

[0028] The processor 210 is capable of executing instructions loaded into RAM 211 from ROM 212, external memory (not shown), storage media such as an SD card, or a communication network. When the electrical meter 120 is powered on, the processor 210 can read instructions from RAM 211 and execute them. These instructions form a computer program that causes the processor 210 to implement all or part of the algorithms and steps described below.

[0029] All or part of the algorithms and steps described below can be implemented in software form by executing a set of instructions by a programmable machine, for example a DSP (Digital Signal Processor) or a microcontroller, or be implemented in hardware form by a dedicated machine or component, for example an FPGA (Field-Programmable Gate Array) or an ASIC (Application-Specific Integrated Circuit).

[0030] There Fig. 3 schematically illustrates an algorithm, implemented by the electric meter 120, for processing a request to send measurement data from the data concentrator 110.

[0031] In step S301, the electricity meter 120 receives a request to send measurement data from the data concentrator 110. A data type can be associated with each request to send measurement data when the electricity meter 120 is adapted to transmit different types of measurement data. For example, one type of request might involve sending measurement data corresponding to a weekly or monthly energy consumption summary for the electrical installation monitored by the electricity meter 120. Another type of request might involve sending measurement data corresponding to discrete points on an accumulated load curve for the electrical installation monitored by the electricity meter 120.

[0032] In a subsequent step S302, the electricity meter 120 retrieves the measurement data to be transmitted to the data concentrator 110. This measurement data corresponds to instantaneous measurement results and / or accumulated measurement results and / or a set of regular measurements. These measurements are performed by the electricity meter 110 at the request of the data concentrator 110 ( e.g. instantaneous measurements) or autonomously via the 110V electricity meter ( e.g. (Regular or cumulative measurements). These measurements typically concern the amount of electrical energy and / or electrical power consumed by the electrical installation monitored by the electricity meter 120. When the electricity meter performs such measurements, the electricity meter 120 stores the results of these measurements in non-volatile memory. In step S302, the electricity meter 120 reads this non-volatile memory to retrieve at least some of these stored results.

[0033] In a subsequent optional step S303, when the measurement data is associated with at least one timestamp, the electricity meter 120 compresses said timestamp(s). A single timestamp may be present for all the measurement data to be transmitted to the data concentrator 110. This is the case, for example, when the data concentrator 110 requires the transmission of a weekly or monthly energy consumption report for the electrical installation monitored by the electricity meter 120. Multiple timestamps may be present for all the measurement data to be transmitted to the data concentrator 110, typically one timestamp per measurement data item to be transmitted.This is the case, for example, when the data concentrator 110 requires the sending of discrete points of the accumulated load curve of the electrical installation supervised by the electric meter 120.

[0034] Thus, in step S303, the 120 electric meter transforms each timestamp information expressed in absolute form into a timestamp information expressed in relative form with respect to a predefined date known to both the 120 electric meter and the data concentrator. The 120 electric meter thereby compresses each timestamp information. The unit used for the relative form can be the second, but is preferably the minute for collecting electric meter measurement data.

[0035] In current smart electricity meters, such timestamp information is an absolute date value represented on 14 bytes. It is then possible to encode this timestamp information in relative form using 28 or 32 bits, with the minute as the unit, which allows for coverage of a time range well beyond the lifespan of electrical installations.

[0036] In a particular embodiment, when the data to be transmitted to the data concentrator 110 corresponds to periodic measurements and each data point is associated with a timestamp representing a specific moment at which the measurement was taken by the electricity meter 120, the electricity meter 120 retains only the timestamp associated with a measurement and deletes the other timestamps. Either the data concentrator 110 is aware of the periodicity of the measurements taken by the electricity meter 111, or the electricity meter 120 provides information representing this periodicity to the data concentrator 110 in association with the measurement data, preferably in a header associated with the measurement data.In a more specific embodiment, when at least one time discontinuity exists between the timestamp of one data point and the timestamp of the next (a break in the measurement periodicity relative to the time reference, as occurs during a switch to or from daylight saving time), the electricity meter 120 identifies at which index this time discontinuity occurs within the measurement data and determines the (signed) time offset at this discontinuity. The electricity meter 120 then transmits to the data concentrator 110, for each identified time discontinuity, preferably in a header associated with the measurement data after compression, information representing this index and this time offset.

[0037] In a subsequent step S304, the electricity meter 120 checks whether the measurement data to be transmitted to the data concentrator 110 is expressed in absolute form, i.e., independently of each other, or in relative form, i.e., the first measurement is typically provided in absolute form and subsequent measurements are provided as a difference from the previous respective measurements. If the measurement data is expressed in absolute form, a step S305 is performed; otherwise, a step S306 is performed.

[0038] In step S305, the electricity meter 120 transforms the measurement data expressed in absolute form into measurement data expressed in relative form. Preferably, the first measurement is stored in absolute form, and for subsequent measurements, the unsigned differences between said measurements and the respective previous measurements are determined and stored as replacements for said measurements. Step S306 is then performed.

[0039] When it is predefined that the measurement data is expressed in relative form, the 120 electric meter can go directly from step S303 to step S306. When it is predefined that the measurement data is expressed in absolute form, the 120 electric meter can go directly from step S303 to step S305.

[0040] In step S306, the electric meter 120 performs a compression of the measurement data to be transmitted to the data concentrator 110.

[0041] When measurement data is unsigned, there is a high probability that a large number of these measurements will contain several consecutive most significant bits set to "0". Rather than representing all unsigned measurement data with the same number of bits, typically 32 bits, it is possible to represent this measurement data using different numbers of bits. Similarly, when measurement data is negative, there is a high probability that a large number of these measurements will contain several consecutive most significant bits set to "1". Rather than representing all negative measurement data with the same number of bits, typically 32 bits, it is possible to represent this measurement data using different numbers of bits. This limits the number of non-significant bits transmitted.

[0042] To do this, the 120 electric meter classifies each measurement data point based on a minimum number of bits required to represent that data point; insignificant most significant bits are thus disregarded. Then, for each class within a predefined plurality of classes, the 120 electric meter determines the number of bits allocated to represent each data point classified within that class, based on the minimum number of bits required to represent the data classified within that class. In other words, the number of bits needed to represent all the data classified within a class is equal to the maximum of the minimum number of bits determined for the data classified within that class. The number of available classes is therefore predefined, but not the number of bits required to represent the data within each class, which needs to be dynamically determined.Next, the 120 electric meter compresses each measurement data point, representing said measurement data on a number of bits equal to the number of bits assigned to the class of said measurement data, and inserting a predefined prefix associated with said class before said compressed data. In other words, the 120 electric meter generates a bitstream that is a succession of prefixes and compressed measurement data points, each prefix representing a class into which the measurement data point following said prefix has been classified, each compressed data point being represented on a number of bits assigned by the 120 electric meter to said class. A particular embodiment of step S306 is detailed below in relation to the [missing information]. Fig. 4 .

[0043] When measurement data is signed, and therefore can be positive, zero, or negative, several alternative options can be implemented.

[0044] According to a first embodiment, when the measurement data is signed, the 120 electric meter performs data classification and compression as if the measurement data were unsigned. This embodiment is particularly advantageous when there are more positive measurement values ​​than negative measurement values. In the case where there are more negative measurement values ​​than positive measurement values, the 120 electric meter can invert (one's complement) the measurement data and indicate in a header that such an inversion has been applied.

[0045] According to a second embodiment, when the measurement data is signed, the 120 electricity meter dedicates one or more classes to compressing negative measurement data and one or more classes to compressing positive measurement data. This class allocation can be predefined or dynamically determined. If this class allocation is dynamically determined, the 120 electricity meter provides information in a header indicating which allocation is applied.

[0046] According to a third embodiment, when the measurement data is signed, the 120V electricity meter can convert all measurement data to positive values. To do this, the 120V electricity meter determines which negative measurement data has the largest absolute value. This absolute value defines a calibration value for the measurement data. The 120V electricity meter then adds the calibration value to each measurement data. The measurement data values ​​then all become positive, and therefore unsigned. The 120V electricity meter then provides information representing this calibration value in a header.

[0047] In a subsequent step S307, the electricity meter 120 generates a response to the request received in step S301. The generated response includes the compressed measurement data and optionally at least one compressed timestamp. The generated response includes a header describing the data compression applied in step S306, and specifically indicating, for each class, the number of bits allocated to represent the compressed measurement data of that class. Since the number of usable classes is predefined, a field can be pre-established for each class in the header to describe the number of bits allocated to represent the compressed measurement data of that class. As already mentioned, the header can also include information indicating for each class whether that class concerns positive or negative measurement data.As previously mentioned, the header can also include registration information to convert all measurement data into positive measurement data. As previously mentioned, the header can also include information on the periodicity of the measurements taken by the 120 electric meter. As previously mentioned, the header can also include index and time offset information representing at least one time discontinuity within the measurement data. The header also includes information representing the number of measurement data points included in the response. The 120 electric meter can add padding bits to the compressed measurement data to allow for 8-, 16-, or 32-bit alignment of the generated response.The 120 electricity meter can also add Cyclic Redundancy Checksum (CRC) information, allowing the 110 data concentrator to verify the integrity of the received response. The 120 electricity meter can also add its serial number information, enabling the 110 data concentrator to verify the authenticity of the received data.

[0048] There Fig. 4 This schematically illustrates an algorithm, implemented by the 120 electric meter, for compressing measurement data. Within the framework of the algorithm... Fig. 4 It is considered that the 120 electric meter can classify measurement data into at least two distinct classes: a first class is suitable for representing the highest value, and at least a second class allows data to be represented using a number of bits lower than the minimum number of bits required to represent the highest value. Within the framework of the algorithm of the Fig. 4 , the 120 electric meter considers the measurement data to be unsigned, whether this measurement data is actually unsigned or not.

[0049] In step S401, the electricity meter 120 determines the most frequent value among the measurement data to be transmitted to the data concentrator 110. In a particular embodiment, the electricity meter 120 has predefined knowledge of the value a priori the most recurring among the measurement data. In the data currently transmitted by electricity meters, the value a priori The most recurring value among the measurement data is "0".

[0050] In a subsequent S402 step, the electric meter 120 identifies the measurement data indexes equal to said most recurring value, so that each of these measurement data can be replaced by a single bit in a later compression step.

[0051] In a subsequent step S403, the electricity meter 120 determines the highest value among the measurement data to be transmitted to the data concentrator 110. If the highest value among the measurement data is equal to the most recurring value, the electricity meter 120 directly performs step S406 (not shown in the diagram). Fig. 4 ).

[0052] In a subsequent S404 step, the electric meter 120 determines a minimum number of bits to represent the highest value among the measurement data.

[0053] In a subsequent step S405, the electricity meter 120 assigns the highest value among the measurement data to the first measurement data class. This defines the maximum number of bits to represent each measurement data item to be transmitted to the data concentrator 110 in response to the request received in step S301.

[0054] In a subsequent step S406, the electric meter 120 determines, for each remaining measurement data point ( i.e. apart from the most recurring value and the highest value) a minimum number of bits to represent said measurement data.

[0055] In a subsequent step S407, the electricity meter 120 classifies each remaining measurement data item into the first class, the second class, or one of the second classes, based on the minimum number of bits required to represent that measurement data item. If the same minimum number of bits is required to represent the measurement data item as to represent the highest value, then the measurement data item is assigned to the first class.

[0056] To determine the number of bits allocated to represent the measurement data for each class, the 120 electric meter, for example, can proceed exhaustively. Let's consider that i represents the number of bits allocated to encoding the measurement data of the first class, that j represents the number of bits allocated to encoding the measurement data of a second class, and that krepresents the number of bits allocated to encoding measurement data of another second class. The 120 electric meter tests all possible combinations of the triplet (i,j,k) and determines, for each of these combinations, the resulting size of the measurement data after compression. It should be noted that i is fixed at step S404. The 120 electric meter then retains the triplet (i,j,k) resulting in the smallest possible size. Once the triplet (i,j,k) Once fixed, the 120 electric meter still needs to assign each measurement data to one of the predefined classes, depending on the minimum number of bits required to represent said data: said class is the one whose number of assigned bits is immediately greater than, or equal to, the value of said data among the triplet (i,j,k).

[0057] In a subsequent S408 step, the electric meter 120 performs an encoding of each measurement data, according to the number of bits allocated to each data item of the class to which said measurement data is assigned.

[0058] A prefix can be added before each compressed data item to identify which class the measurement data has been assigned to. For example, if three classes are defined and a prefix is ​​added for each compressed measurement data item, a two-bit prefix identifies which class that measurement data item is assigned to.

[0059] A prefix can be added before each compressed data item to identify which class the measurement data has been assigned to when that measurement data does not have the most frequent value; and when that measurement data has the most frequent value, no prefix is ​​necessary. For example, if four classes are defined, the 120 electric meter uses: a bit with the value "0" to represent the most frequent value (without a prefix), the binary value "10" as a prefix for a compressed value of another class, the binary value "110" as a prefix for a compressed value of yet another class, and the binary value "111" as a prefix for a compressed value of the last class.Thus, the most frequent value is represented by one bit, and the other compressed measurement data is preceded by a prefix of at least two bits, the first bit of which has a value distinct from that used to represent the most frequent value. More generally, a truncated binary code can be used to identify the data with the most frequent value and the prefixes.

[0060] The remaining task for the 120V electric meter is to fill in the header associated with the measurement data, as already described in relation to the Fig. 3 , in particular by indicating, for each class, the number of bits dynamically allocated to represent the measurement data of said class.

[0061] Thus, when data concentrator 110 receives the response generated and transmitted by the electricity meter 120, it analyzes the header to determine, for each predefined class, the number of bits used to represent the measurement data of that class. The data concentrator then analyzes the bitstream of the measurement data. If the first bit is "0", then data concentrator 110 has received a measurement value equal to the most frequent value. Otherwise, data concentrator 110 has received a prefix associated with one of the classes. The data concentrator takes the bits one by one to determine which prefix is ​​represented, deduces the corresponding class, retrieves the number N of bits assigned to that class from the header, and retrieves the next N bits from the bitstream to extract the corresponding measurement data.The data concentrator 110 can then convert the extracted data back into its original format.

[0062] When timestamp information is associated with measurement data, the data concentrator 110 can perform the reverse operation of step S303 to convert the timestamp information back to its original format. The header can also provide measurement periodicity information to allow the data concentrator 110 to reassociate timestamp information with each unsigned measurement data point. The header can also provide time offset and index information to allow the data concentrator 110 to detect changes in the time reference.

[0063] In a particular embodiment, the 120 electric meter takes into account the signed nature of the measurement data. The 120 electric meter separates positive measurement data from negative measurement data. The 120 electric meter then determines the highest absolute value among the positive measurement data and the highest absolute value among the negative measurement data. The 120 electric meter determines a first minimum number of bits to represent the highest absolute value among the positive measurement data and a second minimum number of bits to represent the highest absolute value among the negative measurement data. The 120 electric meter assigns this first minimum number of bits to a first class and this second minimum number of bits to a second class.The 120 electric meter then tests, for each other class of the predefined plurality, a value of bits assigned to represent the data of that class, and determines a resulting size of the measurement data after compression. Next, the 120 electric meter assigns to each other class of the predefined plurality a number of bits to obtain the smallest resulting size of the measurement data after compression. Finally, the 120 electric meter indicates in a header which class(es) concern positive measurement data (from which successive most significant bits equal to "0" are removed by compression) and which class(es) concern negative measurement data (from which successive most significant bits equal to "1" are removed by compression).

[0064] There Fig. 5 This schematically illustrates an algorithm, implemented by the 120 electricity meter, to determine whether the 120 electricity meter should send data in compressed or uncompressed form. This allows the 120 electricity meter to operate with a data concentrator that is not designed to interpret compressed measurement data. Furthermore, the algorithm's session opening principle... Fig. 5 allows the 110 data concentrator to work with electricity meters that are not suitable for compressing measurement data.

[0065] In an S501 step, the electric meter 120 receives a logon request from a data hub, such as data hub 110. The logon request is accompanied by a password.

[0066] In a subsequent S502 step, the 120 electric meter determines a reverse value of the password. For example, if the password is the string "ABCDEF", the reverse value determined by the 120 electric meter is "FEDCBA". Other types of reverse conversion can be used, such as representing the password in binary and reading the bits in reverse, or using hexadecimal and reading the hexits in reverse. Generally, the 120 electric meter applies a predefined transformation to the received password to obtain a transformed password.

[0067] In a subsequent step S503, the power meter 120 checks whether the reversed (or, more generally, transformed) password from step S502 matches the password expected from the data hub 110 to open the session. In step S504, the power meter 120 tests whether the check is successful. If so, the session is opened and step S505 is performed; otherwise, step S507 is performed.

[0068] In step S505, the 120 electricity meter processes a request to send measurement data. This request can be received by the 120 electricity meter from a data concentrator after the session has been opened. This request can also be implicit in the session opening. The request is processed according to the algorithm of the Fig. 3 That is, with compression of the measurement data. Then, step S506 is performed, in which the 120 electric meter closes the session. The algorithm is then terminated.

[0069] In step S507, the power meter 120 checks if the password received in step S501 matches the password expected from the data hub 110 to open the session. In step S508, the power meter 120 tests whether the check is successful. If so, the session is opened and step S510 is performed; otherwise, step S509 is performed, in which the power meter 120 rejects the session opening request. The algorithm then terminates.

[0070] In step S510, the electricity meter 120 processes a request to send measurement data. This request can be received by the electricity meter 120 from a data concentrator after the session has been opened. This request can also be implicit in the session opening. The electricity meter 120 retrieves the relevant data and transmits this data, uncompressed, to the data concentrator that issued the request. Then, step S506 is executed, and the algorithm terminates.

[0071] Thus, by applying the algorithm of the Fig. 5By using a single password to log in to the electricity meter 120, the electricity meter 120 can determine whether the measurement data should be sent to the data concentrator in compressed or uncompressed form. Furthermore, from the perspective of the data concentrator 110, it is possible to collect data from both electricity meters capable of compressing said data and those incapable of doing so. Indeed, if the data concentrator 110 transmits a reversed password (or, more generally, a transformed password according to the function applied by the electricity meter 120 in step S502) to an electricity meter that does not implement measurement data compression, then that electricity meter will refuse to log in.Data hub 110 can then renew its login request with the non-reversed (or, more generally, untransformed) password. Data hub 110 can also keep track of the success or failure of a login attempt with the reversed (or, more generally, transformed) password, in order to directly send the correct password the next time the same electricity meter is logged in.

Claims

1. Method for making measurement data available by an electrical meter (120), said data coming from electrical energy quantity measurements carried out by said electrical meter on an energy consumption of an electrical installation monitored by said electrical meter, the electrical meter retrieving (S302) the measurement data to be made available, characterized in that the electrical meter further carries out the following steps: - classifying (S405, S407) each retrieved datum according to a minimum number of bits to represent said datum; - determining (S404, S406), for each class of a predefined plurality of classes, a number of bits allocated to represent each datum classified in said class equal to the maximum of the minimum numbers of bits determined for the data classified in said class; - compressing (S306; S408) each retrieved measurement datum, by representing said retrieved measurement datum on a number of bits equal to the number of bits allocated to the class of said measurement datum and by inserting, in front of the compressed datum, a predefined prefix associated with said class; and - including (S307), in a header associated with the compressed data, for each class, information representative of the number of bits allocated to represent each datum classified in said class, and characterized in that the electrical meter carries out the following steps: - determining (S401) the most recurring value among the retrieved measurement data; and - replacing (S408) the measurement data, having a value equal to the determined most recurring value, with a single bit.

2. Method according to Claim 1, characterized in that, with the measurement data being represented in absolute form, the electrical meter carries out a prior step of converting (S305) the measurement data in absolute form into measurement data in relative form.

3. Method according to Claim 1, characterized in that the electrical meter uses a truncated binary code to identify the measurement data having said most recurring value and the prefixes.

4. Method according to any one of Claims 1 to 3, characterized in that, with the measurement data being associated with at least one item of timestamp information representative of an absolute time, the electrical meter carries out a step of converting (S303) said timestamp information into relative information with respect to a predefined date.

5. Method according to any one of Claims 1 to 4, characterized in that, with each measurement datum being associated with an item of timestamp information, and with the measurements carried out by the electrical meter being periodic, the electrical meter retains only an item of timestamp information and includes, in the header, information representative of the periodicity of said measurements.

6. Method according to Claim 5, characterized in that, to indicate a change in time reference among the measurement data, the electrical meter carries out the following steps: - determining an index of each measurement datum for which the change in time reference occurred; - determining a time offset due to the change in time reference for said measurement datum the index of which has been determined in this way; and - including, in the header, information representative of said index and of said time offset.

7. Method according to any one of Claims 1 to 6, characterized in that the electrical meter, considering the measurement data to be unsigned data, carries out the following steps: - determining (S404) a minimum number of bits to represent the highest value among the measurement data; - allocating (S405) said determined minimum number of bits to a first class; - testing, for each other class of said predefined plurality, a value of bits allocated to represent the data of said class, and determining a resulting size of the measurement data after compression; and - allocating to each other class of said predefined plurality a number of bits making it possible to obtain the smallest resulting size of the measurement data after compression.

8. Method according to any one of Claims 1 to 6, characterized in that, with the measurement data being signed, the electrical meter carries out the following steps: - determining a first minimum number of bits to represent the highest absolute value among the positive measurement data; - determining a second minimum number of bits to represent the highest absolute value among the negative measurement data; - allocating said determined first minimum number of bits to a first class; - allocating said determined second minimum number of bits to a second class; - testing, for each other class of said predefined plurality, a value of bits allocated to represent the data of said class, and determining a resulting size of the measurement data after compression; - allocating to each other class of said predefined plurality a number of bits making it possible to obtain the smallest resulting size of the measurement data after compression; and - including, in the header, information indicating which class(es) concern(s) positive measurement data and which class(es) concern(s) negative measurement data.

9. Method according to any one of Claims 1 to 6, characterized in that, with the measurement data being signed, the electrical meter carries out the following steps beforehand: - determining a readjustment value equal to the highest absolute value among the negative measurement data; - adding the readjustment value to each of the measurement data; and - including, in the header, information representative of the readjustment value.

10. Method according to any one of Claims 1 to 9, characterized in that, to make said measurement data available, and with the electrical meter waiting to receive a predefined password, said electrical meter carries out the following steps: - receiving (S501) a password; - determining (S502) an inverted value of the received password; - making (S505) said measurement data available in compressed form when the inverted value of the received password corresponds to said predefined password; and - making (S510) said measurement data available in uncompressed form when the received password corresponds to said predefined password.

11. Computer program, characterized in that it comprises instructions for implementing, by an electrical meter, the method according to any one of Claims 1 to 10 when said program is executed by a processor of said electrical meter.

12. Storage means, characterized in that they store a computer program comprising instructions for implementing, by an electrical meter, the method according to any one of Claims 1 to 10 when said program is executed by a processor of said electrical meter.

13. Electrical meter (120) designed to make measurement data available, said data coming from electrical energy quantity measurements carried out by said electrical meter on an energy consumption of an electrical installation monitored by said electrical meter, the electrical meter comprising means for retrieving the measurement data to be made available, characterized in that the electrical meter further comprises: - means for classifying (S405, S407) each retrieved datum according to a minimum number of bits to represent said datum; - means for determining (S404, S406), for each class of a predefined plurality of classes, a number of bits allocated to represent each datum classified in said class equal to the maximum of the minimum numbers of bits determined for the data classified in said class; - means for compressing (S306; S408) each retrieved measurement datum, by representing said retrieved datum on a number of bits equal to the number of bits allocated to the class of said datum and by inserting, in front of the compressed datum, a predefined prefix associated with said class; and - means for including (S307), in a header associated with the compressed data, for each class, information representative of the number of bits allocated to represent each datum classified in said class, and characterized in that the electrical meter further comprises: - means for determining (S401) the most recurring value among the retrieved measurement data; and - means for replacing (S408) the measurement data, having a value equal to the determined most recurring value, with a single bit.