Controlling the energy consumption of a communication network

Abstract

The invention relates to a method for controlling the energy consumption of a cellular radiocommunication site in the event of a power outage, the site being provided with an energy storage device and operating in accordance with at least one frequency band and / or at least one cellular radiocommunication network generation, the method comprising: - activating (S1) the storage device, - calculating (S2) a time of deactivation of the at least one frequency band or of the at least one network generation, this calculating being carried out in accordance with a criterion for optimizing the energy stored in the storage device during the outage, - deactivating (S3) the at least one frequency band or the at least one network generation at the calculated time of deactivation.

Description

[0001]DESCRIPTION Title: Control of the energy consumption of a communication network Field of the invention The field of the invention is that of cellular radiocommunication networks and more particularly of the control of the energy consumption of these networks, in particular for the management of the autonomy of one or more energy storage devices present in these networks, for example one or more batteries, when a power outage occurs, whether this outage is involuntary or voluntary, which can happen for example in the case of load shedding or a switch to standby mode planned by an operator of such a network. Prior art Current communication infrastructures, in particular cellular radiocommunication sites on which radio stations are installed, are equipped with one or more energy storage devices. In the event of a power outage,an energy storage device is put into operation to take over and guarantee the continuity of the service offered to users of such infrastructures. This backup measure is crucial to avoid any interruption of communication or broadcasting. However, it is essential to keep in mind that such an energy storage device has limits in terms of autonomy. If the power outage persists for a prolonged period, the energy storage device may quickly become depleted, which would lead to an inevitable interruption of services. A simple solution is to let the energy storage device take over after a power outage, without prior planning. However, this approach has the disadvantage of quickly exhausting the capacity of the energy storage device. This results in a rapid deterioration of the quality of service (QoS) perceived by users of the network(s) of such infrastructures,which can have a negative impact on their user experience, since the communication services relying on these infrastructures can no longer function properly. In the case of a cellular radiocommunication infrastructure, a "reactive" solution can be considered, which means that the moments when the frequency bands (e.g.: 900 MHz for 2G to 4G, 2.1 GHz for 3G to 4G, 3.5 GHz for 5G, etc.) or network generations (2G, 3G, 4G, 5G, etc.) managed in this infrastructure must be switched off are not planned in advance. Instead, when a power outage occurs, a system in the infrastructure continuously monitors the autonomy of the energy storage device. When this autonomy reaches a critical threshold, a first frequency band is deactivated, which increases the duration during which the energy storage device can take over. Then,the system continues to monitor autonomy and gradually deactivates other frequency bands, each time the threshold is reached again, and so on. However, this last solution has notable drawbacks. First of all, it is reactive, which means that measurements of the autonomy of the energy storage device are continuously fed back. If there is a delay in the feeding back of these measurements, this can lead to a delay in the process of switching off frequency bands or network generations,which can reduce the overall performance of the system. Purpose and summary of the invention One of the aims of the invention is to remedy at least one of the drawbacks of the aforementioned state of the art by proposing a new technique for managing the autonomy of an energy storage device of a cellular radiocommunication site which makes it possible to find a balance between the reactivity necessary to effectively manage the energy storage device in the event of a power outage and the prior planning to avoid deterioration of the QoS, with a view to optimizing the management of the autonomy of the energy storage device for the benefit of a guarantee of stable performance for users of the communication services managed by a cellular radiocommunication site. To this end,an object of the present invention relates to a method for controlling the energy consumption of a cellular radiocommunication site during a power outage according to claim 1. Claims 2 to 9 describe preferred embodiments of said method. The site is provided with an energy storage device and operating according to at least one frequency band and / or at least one generation of cellular radiocommunication network. The method may comprise, at the time of the outage: - an activation of a storage device, - a calculation of a time of deactivation of said at least one frequency band or of said at least one generation of network, said calculation being implemented according to a criterion for optimizing the energy stored in the storage device during the outage, - a deactivation, at said calculated time of deactivation,of said at least one frequency band or said at least one network generation. Such a method thus makes it possible to optimally preserve the storage resources of the energy storage device, for the duration of the power outage, by scheduling a time for deactivating said at least one frequency band or said at least one network generation of the cellular radiocommunication site. Thus, thanks to the invention, the energy storage device avoids discharging too quickly, for the benefit, during the power outage, of maintaining the quality of service or QoS perceived by users, whose communications received or transmitted from their communication terminals pass through the cellular radiocommunication site. The site can then be quickly restored to full capacity once the electricity has returned to normal. According to a particular embodiment,when the site operates according to at least two frequency bands and / or at least two generations of cellular radiocommunication network, said method comprises, at the time of the cut-off, once the energy storage device is activated: - a calculation, according to said optimization criterion, of at least two successive instants of deactivation of respectively the at least two frequency bands or of respectively the at least two generations of network, - a deactivation, at said at least two calculated instants of deactivation, of respectively the at least two frequency bands or of respectively the at least two generations of network. Such an embodiment is based on an effective strategy implementing a progressive reduction of the frequency bands or of the generations of cellular radiocommunication network used on the site,thus reducing energy consumption and increasing the duration during which the energy storage device can take over, while waiting for the restoration of normal power supply. Thus, for example, in the case where a radio site uses four cells, such as for example LTE-2600, LTE-800, 3G-2100 and 2G-900, when a power outage occurs, rather than keeping these four cells active throughout the period of the outage, the site implements a gradual reduction strategy. At a first moment, it is for example the 2100 MHz frequency band of the 3G-2100 cell which is deactivated, which reduces the consumption of the site and extends the duration of the energy storage device. If the power outage persists, other frequency bands can be gradually deactivated. By following this gradual approach,the radio site extends the duration during which the energy storage device can support the services, while optimizing the use of the energy stored at the energy storage device. According to another particular embodiment, the at least two frequency bands or the at least two network generations are deactivated according to a priority order which is calculated, the calculation of the at least two successive instants of deactivation according to the calculated priority order being further implemented according to a criterion for maximizing the volume of data traffic during the outage. Such an embodiment further makes it possible to optimize the performance of the control method by maximizing the volume of data traffic during the power outage. According to another particular embodiment, for each of the at least two frequency bands or the at least two network generations,the order of priority is calculated based on a ratio between the energy consumption and the volume of data traffic determined before the outage for each of the at least two frequency bands or the at least two network generations. Such an embodiment makes it possible to define an order of priority according to which at the two successive deactivation times, the frequency band or the network generation which has the lowest energy efficiency and the frequency band or the network generation which has the highest energy efficiency are deactivated respectively. According to another particular embodiment, the at least two frequency bands or the at least two network generations are deactivated according to a priority order which is predefined. Such an embodiment allows an operator or a manager of the cellular radiocommunication site, prior to the power outage,to configure, as desired, the order in which the at least two frequency bands or the at least two network generations will be deactivated, at said at least two successive instants. Thus, for example, it could be decided to deactivate, at said at least two successive instants calculated: - a 3G-2100MHz cell and a 2G-900MHz cell, respectively, - a 2100MHz frequency band and a 900MHz frequency band, respectively. According to another particular embodiment, the at least two deactivation instants, from which the at least two frequency bands or the at least two network generations are deactivated respectively according to said predefined priority order, are calculated according to an autonomy threshold of the energy storage device, which is crossed for each of the at least two frequency bands or the at least two network generations. Such an embodiment allows,by monitoring the autonomy of the energy storage device at the time of the power outage and for the entire duration of this outage, to: - deactivate, at a first instant of deactivation, a first frequency band or a first network generation as soon as they cross an autonomy threshold of the energy storage device, such an operation making it possible to increase the autonomy of the energy storage device, - deactivate, at a second instant of deactivation, a second frequency band or a second network generation, as soon as they cross an autonomy threshold of the energy storage device, such an operation making it possible to increase the autonomy of the energy storage device, - and so on in the case where there are more than two frequency bands or network generations on the site. This deactivation strategy is advantageously based on a single-pass calculation,at the time of the power outage, different deactivation times. According to another particular embodiment, the deactivation time or said at least two successive deactivation times are calculated according to a set of data relating to the site, said set of data comprising the energy stored in the energy storage device at the time of the outage, the duration of the outage, at least one data item relating to the operation of said cellular radiocommunication network at the time of the outage, and the energy consumed by the site at the time of the outage. Such an embodiment makes it possible, thanks to the specification of a certain number and a certain type of data relating to said site, to determine: - either the optimal deactivation time of at least one frequency band or at least one network generation, when the site operates according to this at least one frequency band or according to this at least one network generation,- either the optimal chain of the at least two instants of deactivation of respectively the at least two frequency bands or of respectively the at least two network generations, when the site operates according to at least two frequency bands or according to at least two network generations. According to another particular embodiment, the energy consumed by the site at the time of the outage is determined using an artificial intelligence model trained from indicators representative of the operation of the site, collected at different time intervals. Such an embodiment makes it possible to precisely determine the energy consumed by the site before the outage in order to thus avoid unwanted interruptions of communication service during the power outage. Furthermore, the more precise the determination of the energy consumed, the more efficient the strategy for deactivating the frequency bands or the network generations will be,starting with those which are the least critical for QoS. According to another particular embodiment, the modeling of the energy consumed at the time of the outage is implemented using a Lasso linear regression method. The advantage of such a linear regression method is that it is particularly well suited to the processing of indicators representative of the operation of the site, such as for example key performance indicators or KPIs (in English, "Key Performance Indicator"), raw indicators (number of resources occupied per cell, quantity of data sent via the network, etc.). The various aforementioned embodiments or characteristics can be added independently or in combination with each other, to the method for controlling the energy consumption of a cellular radiocommunication site during a power outage,as defined above. The invention also relates to a device for controlling the energy consumption of a cellular radiocommunication site during a power outage according to claim 10. The site is provided with an energy storage device and operating according to at least one frequency band and / or at least one generation of cellular radiocommunication network. The device can be configured to, at the time of the outage, implement: - an activation of the storage device, - a calculation of a time of deactivation of said at least one frequency band or of said at least one generation of network, said calculation being implemented according to a criterion for optimizing the energy stored in the energy storage device during the outage, - a deactivation, at said calculated time of deactivation,of said at least one frequency band or of said at least one network generation. Such a device is in particular configured to implement the aforementioned control method, according to one or other of its embodiments. The invention also relates to a computer program comprising instructions for implementing the control method according to the invention, according to any one of the particular embodiments described above, when said program is executed by a processor. Such instructions may be stored permanently in a non-transitory memory medium of the content reception device implementing the control method according to the invention. This program may use any programming language, and be in the form of source code, object code, or intermediate code between source code and object code, such as in a partially compiled form,or in any other desirable form. The invention also relates to a recording medium or information medium readable by a computer, and comprising instructions of a computer program as mentioned above. The recording medium can be any entity or device capable of storing the program. For example, the medium can comprise a storage means, such as a ROM, for example a CD ROM or a microelectronic circuit ROM, or a magnetic recording means, for example a mobile medium, a hard disk or an SSD. On the other hand, the recording medium can be a transmissible medium such as an electrical or optical signal, which can be conveyed via an electrical or optical cable, by radio or by other means, so that the computer program it contains is remotely executable. The program according to the invention can in particular be downloaded over a network,for example an Internet-type network. Alternatively, the recording medium may be an integrated circuit in which the program is incorporated, the circuit being adapted to execute or to be used in the execution of the aforementioned control method. According to an exemplary embodiment, the present technique is implemented by means of software and / or hardware components. With this in mind, the term "device" or "module" may correspond in this document to a software component, a hardware component or a set of hardware and software components. Brief description of the drawings Other characteristics and advantages will appear on reading particular embodiments of the invention, given as illustrative and non-limiting examples, and the appended drawings,among which: - Figure 1 represents an architecture in which the method for controlling the consumption of a cellular radiocommunication network is implemented, according to a particular embodiment of the invention, - Figure 2 represents a device for controlling the consumption of a cellular radiocommunication network during a power outage, according to a particular embodiment of the invention, as implemented in the architecture of Figure 1, - Figure 3 represents the main actions implemented in the method for controlling the consumption of a cellular radiocommunication network during a power outage, according to a particular embodiment of the invention, as implemented in the architecture of Figure 1, - Figure 4 represents a curve of the evolution, in a real situation, of the energy consumption of an element of the cellular radiocommunication network, as a function of time,- Figure 5A represents the main actions implemented during one of the steps of the method for controlling the consumption of a cellular radiocommunication network during a power outage, as illustrated in Figure 3, according to a first embodiment, - Figure 5B represents the main actions implemented during one of the steps of the method for controlling the consumption of a cellular radiocommunication network during a power outage, as illustrated in Figure 3, according to a second embodiment, - Figure 6 represents a schematic curve illustrating the strategy for deactivating RRUs of the cellular radiocommunication network,in accordance with the second embodiment of Figure 5B. Detailed description of an embodiment of the invention Figure 1 represents an architecture in which a method for controlling the consumption of a cellular radiocommunication network RC during a power outage PO of duration Δ^^ is implemented, according to an embodiment of the invention. Such a network RC is for example of the 3G, 4G, 5G, etc. type. Such an architecture comprises: - a calculation device DME configured to estimate the total energy consumed by a site S,installed at a given geographical location and comprising the equipment necessary for the transmission and reception of communication data via the RC network; - an energy storage device STO which equips the site S; - a DCC device configured to control the energy consumption of the site S during a power outage PO. The energy storage device STO is capable of storing energy in a given form, and of restoring electrical energy, on command, to the site S. The storage device STO may for example be an electric battery. Alternatively, the storage device STO may be a generator, an inertial flywheel, or any other storage device capable of storing energy in a given form and of restoring electrical energy on command. According to the invention,such a DCC control device is in particular configured to activate the energy storage device STO at the time of the power outage PO and to calculate, optimally from a quality of service point of view and in a single pass, K instants of deactivation of respectively K elements EL1, EL2, …, Ek, …, ELK of the same nature which make up the site S, such that 1≤k≤K. The RC network being a cellular radiocommunication network, the K elements EL1, EL2, …, ELK may be respectively K different frequency bands, for example 1800 MHz and 2100 MHz or at least two different generations of cellular network, for example 3G and 4G, in the case for example where K=2. The elements EL1 to ELK may also be K RRUs (in English “Remote radio Unit”) each operating on a different frequency band. Conventionally,an RRU corresponds to a group of cells that operate on the same frequency band with different network generations (2G and 3G on the 900 MHz band for example). The DCC device is for example a server, a platform, for example of the CSON type ("Centralized - self-organizing networks" in English), RIC intelligent controller ("Radio access network Intelligent Controller" in English) or OSS ("Operations Support System" in English), etc. It can be installed on the site S or remotely. According to the invention, the DME calculation device is an artificial intelligence module that generates a mathematical modeling of the energy ^^, ^^ consumed on site S, from the collection of DR data relating to the operation of the RC network. Such generation is implemented prior to the power outage. The DME calculation device is configured to transmit the energy ^^ to the DCC calculation device ^^estimated according to the aforementioned modeling. According to the invention, the DCC calculation device is configured to:- calculate K deactivation instants Δ^^1, Δ^^2, … , Δ^^^^ … , ^^^^ of respectively the K elements EL1, EL2, …, ELK, as a function of the energy ^^ ^^ , of the value of energy ^^ ^^stored in the energy storage device STO at the time of the PO outage, the duration ΔT of the power outage, as well as DR data relating to the operation of the RC network, collected in a time interval preceding the PO outage, - control the deactivation of the K elements EL1, EL2, …, ELK, respectively at said K calculated deactivation times. We will now describe, with reference to Figure 2, the simplified structure of the DCC control device.The DCC device comprises: - a communication interface COM configured to communicate, via the RC network of figure 1 or another network, with in particular the calculation device DME and the site S, - a module COL for collecting data DR relating to the operation of said RC network, - a module COB for controlling activation / deactivation of the energy storage device STO, - a module CAL for calculating K deactivation instants Δ^^1, Δ^^2, … , Δ^^^^ or ^^1, ^^2, … , of respectively K elements EL1, EL2, …, ELK of the same nature which compose the site S, - a module COE for controlling the deactivation of the K elements EL1, EL2, …, ELK, at said K calculated deactivation instants. At initialization, the code instructions of the computer program PG are, for example, loaded into a RAM memory (not shown) before being executed by the processor PROC.The processor PROC of the processing unit UTR notably implements the following actions, within the framework of the method for controlling the energy consumption of the site S during a power outage PO, which will be described below, according to the instructions of the computer program PG: - receive, from a communication terminal associated with the site S, information IPO according to which a power outage PO has occurred, - command the activation of the energy storage device STO, - receive, from the calculation device DME, an estimated value of the energy ^^. ^^ consumed by site S before the power outage PO, - receive, from a communication terminal associated with site S or from the energy storage device STO if the latter is equipped with a dedicated communication module, the value of the energy ^^ ^^stored in the energy storage device STO at the time of the PO outage, - receive, from a communication terminal associated with the site S or an electricity supplier FE with which the site S is referenced, a value of the duration Δ^^ of the power outage PO, - collect data DR relating to the operation of said network RC, - calculate K deactivation times Δ^^1, Δ^^2, … , ^^^^ of respectivelyK elements EL1, EL2, …, ELK of the same nature which compose the site S. Optionally, the control device DCC is configured to: - control the deactivation of the energy storage device STO after restoration of the current, - control the activation of the K elements EL1, EL2, …, ELK, after restoration of the current. We now describe, in relation to figure 3, together figures 1 and 2, the progress of a method for controlling the consumption of the cellular radiocommunication network RC during the power outage PO, according to a particular embodiment of the invention.Such a power outage may for example be caused by the electricity supplier of the radio site S or result from load shedding or putting into sleep mode decided by the network manager RC, when the latter wishes for example to deactivate, for a duration ΔT, one or more RRUs, one or more frequency bands or one or more generations of radiocommunication network. As already mentioned above, such a method comprises a preliminary phase S0 of energy estimation ^^. ^^consumed on site S, from the collection of DR data relating to the operation of the RC network. Such a phase is implemented by the DME computing device. Such DR data may include raw indicators relating to the operation of the network, such as for example the number of resources occupied per cell covered by site S, the quantity of data sent via the RC network, etc. Alternatively, such DR data may include performance indicators or KPIs relating to the RC network, such as for example, depending on the generation of cellular network considered 2G, 3G, 4G, 5G, etc.: the volume of traffic, the throughput for each cell, the average number of users for each cell, the cell load, etc. The energy consumption KPI is also collected by the DME computing device.Typically, such an energy consumption KPI is available for the radio site S, for each of the frequency bands or network generations of the radio site S. The KPIs are reported with a granularity of ^^ which is variable depending on the modeling context, for example a time interval of 15 min, 30 min, 1 hour, etc. The goal of the S0 phase is to model the energy consumption ^^ accurately and reliably. ^^using an artificial intelligence technique. The accuracy of the modeling is crucial to avoid unwanted service interruptions. By having an accurate estimate of the power consumption, the strategies for deactivating the K elements EL1, EL2, …, ELK which will be detailed later in the description will be particularly suitable for selectively deactivating certain network generations or certain frequency bands, starting with those which are the least critical for QoS. According to the invention, the DME calculation device uses for the modeling of the energy consumption ^^ ^^a linear regression method, for example the LASSO method (in English, "Least Absolute Shrinkage and Selection Operator"), Ridge, etc. In the embodiment described below, it is the LASSO method which is considered, this type of method being particularly well adapted to the processing of indicators representative of the operation of the site, whether KPIs, or raw indicators (number of resources occupied per cell, quantity of data sent via the network, etc.). The LASSO regression which links the variable of interest, which is here the energy consumed ^^ ^^ , with the explanatory variables, which are here the KPIs or the raw indicators mentioned above, is expressed as follows: where: - ^^ ^^ is the total energy consumed by site S, - ^^ ^^corresponds to the energy consumed when there is no user on the radio site and therefore no traffic, - ^^ corresponds to a generation of cellular network used on site S, - ^^ ( ^^ ) corresponds to a set of different frequency bands used on site S and corresponding to the network generation ^^, - ^^(^^) is the number of KPIs considered for the network generation ^^, - ^^ ^^,^^ is a linear regression coefficient that is driven in relation to an ith frequency band considered for a network generation ^^, such that 1≤i≤ ^^ ( ^^ ), and for a jth KPI considered, such that 1≤j≤ ^^(^^). In the example shown, three generations of network 2G, 3G, 4G are considered, such that ^^ ∈ {2,3,4}. It goes without saying that this number can be less than 3, for example 3G or 3G and 4G, or greater than 3, for example 2G, 3G, 4G, 5G. The expression of the energy consumed per generation of network ^^ is represented in the following form: where ^^ ^^^^ is the energy consumed when data traffic is zero or very low, as is the case for example at night. The expression of the energy consumed by a considered RRU or a considered frequency band is represented in the following form: where ^^ ^^ represents the intercept which is a fixed value of the energy consumed for the RRU or the frequency band ^^. In a manner known per se, such a linear regression is subjected to a Lasso regularization which is a regularization technique to penalize the coefficients ^^^^,^^,^^ (expression (1)), ^^ ^^,^^ (expression (2)), ^^ ^^,^^ (expression (3)) of less important features, by bringing these coefficients to zero. Thus, Lasso regularization adds a penalty proportional to the absolute value of the coefficients to the loss function. Such regularization is applied during the processing of the model and when it processes new samples by going from i to i+1. Modeling the energy consumed ^^ ^^ , ^^ ^^^^ , ^^ ^^ , according to the above Lasso regression is implemented based on a set of real data (KPIs or raw indicators mentioned above) collected according to a time interval , where, by way of non-exhaustive example, = 60 min, according to this modeling. The performance indicator RE of the modeling of the consumed energy is expressed according to the relation (4) below, in the form of a ratio between the difference in absolute value of the real value of ^^ ^^ , ^^ ^^^^ or ^^ ^^ , and the estimated value of ^^ ^^ , ^^ ^^^^ or ^^ ^^ , and the actual value of ^^ ^^ , ^^ ^^^^ or ^^ ^^ , that is to say the relative error RE between the real and estimated values ​​of ^^ ^^ , ^^ ^^^^ or ^^ ^^ : of the estimation of the energy consumed ^^ ^^ , ^^ ^^^^ , ^^ ^^was performed for different metrics: mean RE(%) which represents the average of the RE values ​​over all test samples, RE<15% which represents the probability (or rate) that the RE value is less than 15% over all test samples, RE<10% which represents the probability (or rate) that the RE value is less than 10% over all test samples, considering the estimation of ^^ ^^ consumed by site S, the estimate of ^^ ^^^^ consumed for each of the 2G, 3G, 4G network generations and the estimate of ^^ ^^for an RRU operating at a low frequency, for example 800 MHz or 900 MHz, or a high frequency, for example 2100 MHz or 2600 MHz. The results of this training, as shown in table TAB1 below, show that the error in the energy estimation is low, whether this energy is estimated per site, per network generation or per frequency band. TAB1: Relative error on the estimation of the consumed energy Site S 2G 3G 4G RRU Mean RE (%) 5.98 2.61 7.98 6.07 6.53 RE<15% (%) 97.83 100 90.83 99.57 96.92 RE<10% (%) 92.3 97.45 79.19 97.09 84.17 The results shown in Table TAB1 show that the error on the energy estimation is low. Thus, the modeling of the energy estimation based on the aforementioned Lasso regression is robust to accurately estimate the energy consumption based on the aforementioned KPIs or raw indicators.Lasso regression, which incorporates a regularization term as defined above, is particularly effective in dealing with multicollinearity of predictive values ​​of raw KPIs or indicators, i.e., the high correlation between these predictive values. In the context of the aforementioned modeling, based on linear regression with several raw KPIs or indicators, including traffic, with some raw KPIs or indicators being predominant compared to other considered raw KPIs or indicators, Lasso regression introduces a penalty term according to which some regression coefficients, trained for non-dominant raw KPIs or indicators, are pushed to reach exactly zero. Such a penalization is notably described in the paper PK Matthew, FA Chama, and NS Agog, “Penalization Techniques for Remedying Multicollinearity in Multiple Regression Model” KASU Journal of Mathematical Science, vol.3.1, pp.41-49, 2022.Thus, when Lasso regression is applied to a set of raw KPIs or indicators, it is likely to identify and retain one or more of the most relevant raw KPIs or indicators, for example the "traffic" KPI or the "amount of data sent" raw indicator, while forcing the regression coefficients associated with the least relevant or highly correlated raw KPIs or indicators to be zero. This not only improves the interpretability of the model, but also improves its performance by focusing on the most informative explanatory variables. We now describe, with reference to Figure 4, a curve representing the evolution, in a real situation, of energy consumption, for example per RRU, as a function of time and over several days "Day 1", "Day 2", "Day 3", "Day 4".It is found that the fixed value of energy, which does not depend on traffic, is quite high compared to the consumption linked to traffic. The accuracy of the intercept value ^^ is also found. ^^ in equation (3), since late at night, users do not transmit data via their mobile terminals and the energy consumed corresponds to ^^ ^^ . On the other hand, we notice that ^^ ^^ exceeds half of the energy consumption. Of course, correspondingly, the evolution, in real situations, of the energy consumption, per site S or generation of cellular radiocommunication network, as a function of time and over several days, can also be obtained to verify the accuracy of the values ​​^^ ^^ and ^^ ^^^^The representative curve in Figure 4 gives an indication of the deactivation strategy to be implemented, although the curves representing the evolution, in real situations, of the energy consumption, by site S or generation of cellular radiocommunication network, as a function of time and over several days, have not been shown. Indeed, this curve shows that it is not recommended to proceed with a cell-by-cell extinction of the RC network. Indeed, if one cell, for example 3G-2100 MHz, is deactivated, then another cell, for example 3G-1800 MHz, then yet another cell, for example 4G-2100 MHz, the energy management of the STO energy storage device is not optimal. The RRU operating at the 2100 MHz frequency remains functional even when the 3G-2100 MHz cell is deactivated, because the other 4G-2100 MHz cell is still switched on.This means that the fixed part of the RRU (intercept) still consumes energy even after the 3G-2100 MHz cell has been deactivated. And the intercept consumes a lot of energy. The optimal strategy to favor is to deactivate RRU by RRU or frequency band by frequency band, to save the energy of the fixed part, once the RRU is deactivated. Referring again to Figure 3, once the preliminary phase S0 of energy estimation ^^. ^^consumed on the site S has been implemented, the energy storage device STO is activated in S1 when a power outage PO is detected on the site S. The activation S1 is implemented by the aforementioned COB module of the control device DCC. The detection of the power outage PO may for example comprise a reception by the control device DCC, via its communication interface COM, of information IPO according to which a power outage PO has occurred, coming from a communication terminal associated with the site S or from communication equipment of an electricity supplier which supplies the site S with electricity. The method continues by implementing, in S2, a calculation of K deactivation instants … , ^^^^ of respectively said K elements EL1, EL2, …, ELK of the same nature which compose the site S. The calculation S2 is implemented by the aforementioned CAL module. In S3, said K elements EL1, EL2, …, ELK are then deactivated respectively at said K deactivation times Δ^^1, Δ^^2, … , calculated, using the aforementioned COE control module. The invention can be implemented for a single element (K=1), in the case where the site S comprises only one frequency band or one network generation, as well as for several elements, i.e. at least two elements of the same nature. Thus, the method which has just been described above can of course be applied to two, three, four, five etc. different frequency bands, different RRUs or different cellular radiocommunication network generations, which implies the calculation in S2 of respectively two, three, four, five, etc. deactivation times Δ^^2, Δ^^3, Δ^^4, Δ^^5, etc. or ^^2, ^^3, ^^4, ^^5, etc. We now describe, with reference to Figure 5A, a first embodiment of the calculation step S2 of said K deactivation instants Δ^^1, Δ^^2, … , Δ^^^^ .According to this first embodiment, the calculation step S2 is implemented so as to maximize one of the aforementioned KPIs or raw indicators, for example according to the data volume, the amount of data sent via the RC network, etc. during the power outage PO. The first embodiment will be described by considering the maximization of the data volume passed by RRU of the radio site S. It is considered that the radio site S operates with ^^ RRUs, where K≥1 and (Δ^^1, Δ^^2, … , Δ^^^^) are the switching-off times of the ^^RRUs. According to the embodiment shown, the DCC control device is configured to determine the optimal switching-off times ∆^^1. ∗ , ∆^^2 ∗ , …, ∆^^ ^ ∗^ allowing to maximize the volume of traffic carried during the duration Δ^^ of the power outage PO. A set of data relating to the operation of the radio site S, as used in the calculation S2 comprises the following data: - ^^0 the energy stored in the energy storage device at the time of the power outage PO, - Δ^^ the duration of the power outage PO, included for example in the information IPO, - ^^^^ = ^^^^,2^^ + ^^^^,3^^ + ^^^^,4^^ the volume of traffic carried at the time of the power outage PO, for a kth RRU among K, during a time interval ^^ , where ^^ ^^,^^^^ is the KPI corresponding to the volume of traffic of the network generation nG collected at the time of the outage, where ^^ is for example such that 2 ≤ ^^ ≤ 4.- ^^ ^^the total energy consumed by the site S, as estimated at S0. For this purpose, the method comprises, at the time of the power cut PO, a reception step S20, during which the control device DCC receives, via its communication interface COM: - an estimated value of the energy ^^ ^^ consumed by site S before the power cut PO, from the DME calculation device, - an energy value ^^ ^^ stored in the energy storage device STO at the time of the PO outage, coming from a communication terminal associated with the site S or from the energy storage device STO if it is equipped with a dedicated communication module, - a value of the duration Δ^^ of the power outage PO, coming from a communication terminal associated with the site S or from an electricity supplier FE with which the site S is referenced. Of course, as an alternative, the estimated value of the energy ^^ ^^, the value of energy ^^ ^^ and value of the duration Δ^^ of the power outage PO can be received simultaneously or each at different times. The method further comprises, at the time of the power outage PO, a step S21 of collecting the data DR relating to the operation of said RC network, such as the value of the traffic volume ^^^^ = ^^^^,2^^ + ^^^^,3^^ + ^^^^,4^^, in the example shown, indicators of energy consumption associated with the kth RRU, etc. During a step S22, the calculation module CAL solves the following optimization problem (5): s oumis à ∑ ^^ ′ ^^=1 ^^^^ ≤ ^^0,(5) ∆^^^^ ≥ 0, for 1≤k≤K,^^ ^^ ≤^^ ^^,^^ , for 1≤k≤K, where ^^^^ = ^^^^,2^^ + ^^^^,3^^ + ^^^^,4^^ to be optimized here corresponds to the amount of traffic transmitted by the kth RRU during the reference time interval and ^^ ^^,^^^^corresponds to the KPI of the traffic quantity of the kth RRU of the network generation ^^^^ , in equation (3). Thus, ^^ is the total number of bits passed by the kth RRU during the PO power outage. Note that if one or more RRUs is / are deactivated, its users will switch to the other RRUs still active. This results in a new distribution of traffic and users on the remaining bands after the deactivation of one or more RRUs. Thus, ^^ ^^ may vary due to traffic transfer from a powered-off RRU to another powered-on RRU. Furthermore, in optimization problem (5) above: - ^^ ^^ is the amount of traffic of the kth RRU that is collected at the time of PO power outage, - ^^ ^^,^^ is the maximum amount of traffic that the kth RRU can support, the maximum amount of traffic being obtained from the data collected in S21, - ^^ ^^is the corresponding energy consumed during the reference time interval ^^, - is the energy consumed by the kth RRU during Δ^^ ^^ , from the occurrence of the PO power outage, when this kth RRU transmits ^^ ^^ bits instead of ^^ ^^ , where ^^ ^^,^^ is the energy consumed when the kth RRU transmits ^^ ^^ bits during the reference time interval ^^. ^^^^,^^ is calculated by considering the above expression (3) and using ^^ ^^,^^^^ instead of ^^ ^^,^^^^ . Remember that ^^ ^^,^^^^ and ^^ ^^,^^^^are used in (3) as the traffic KPIs of the technology ^^^^. For the other KPI values ​​to be used in expression (3), it is sufficient to assume that they change proportionally in the same way as the traffic KPI or to create a learning model that links each KPI to the traffic. In order to solve the optimization problem (5), an energy efficiency metric, measured in bits / Joule, is defined for each RRU. For a kth RRU, the energy efficiency metric, denoted ^^ ^^ , is expressed as follows: ^^ ^^ ^^ = ^^ ^^ ^^ (6)Without loss of generality, we assume for this approach that ^^1 ≥ ^^2 … ≥ ^^^^. Thus the calculation of ^^1 ≥ ^^2 … ≥ ^^^^ defines a priority order in which to successively deactivate the K elements by first deactivating the element ELK with the lowest energy efficiency and last deactivating the element EL1 with the highest energy efficiency. Note that ^^^^ is not constant and varies from measurement to measurement in reality. However, from several experiments based on a real data set, it is found that when ^̅^ ^^ (the average value of ^^ ^^ ), such that 1≤i≤K, is significantly greater than^̅^^^, such that 1≤i≤K, then ^^^^ > ^^^^ almost all the time. The case where the means of ^̅^^^ and^̅^ ^^ are close does not pose a problem since the optimal solution will not differ much. It is obvious that if the energy ^^0 stored in the energy storage device is sufficient to carry all the requested traffic, all the RRUs remain functional for the entire duration ΔT of the PO power outage. Ainsi, ∆^^ ∗ ^^ = ∆^^ for 1≤k≤K (7)^^ ∗ ^^ = ^^^^ for 1≤k≤K,where ^^ ^ ∗ ^ is the optimal volume of traffic carried for a kth RRU at the optimal deactivation time ∆^^ ^ ∗ ^.And if ^^0 < , it is no longer possible to keep all RRUs functional for the entire duration ΔT of the PO power outage. It is clear that more ^^ ^^ is high, the higher the volume of traffic transferred, because more bits are transmitted per unit of energy. Thus, the solution to optimization problem (5) is to keep the RRUs that have high energy efficiency for the duration ΔT of the PO power outage. These RRUs that have high energy efficiency will also transmit with their maximum capacity. Conversely, RRUs with low energy efficiency are deactivated because they consume more and transmit fewer bits. Thus, the solution (8) to optimization problem (5) is as follows: ∆^^ ∗ ^^ = ∆^^, ^^∗^^ = ^^^^,^^, if 1≤k≤N-1 ∆^^ ^ ∗ ^ = 0, ^^ ^ ∗ ^ =0, if k>N, where ^^ = arg is the energy consumed by the kth RRU during ^^ when it transfers ^^ ^^,^^. Energy ^^ ^^,^^ is calculated using expression (3), when the traffic volume is equal to ^^ ^^,^^. A second embodiment of the calculation step S2 of the K deactivation instants is now described with reference to FIG. 5B. In accordance with this second embodiment, the calculation step S2 is implemented so as to optimize the autonomy of the energy storage device STO during the duration ΔT of the power outage PO, the order in which the K RRUs are deactivated being predefined by the network manager RC. For this purpose, a preliminary step SP2 is implemented, during which the autonomy threshold ^^ℎ of the energy storage device STO is defined from which it is appropriate to switch off a first RRU, then a second and so on. The threshold ^^ℎ is defined in minutes in the embodiment shown. Of course, in other embodiments, the threshold ^^ℎ can be defined in hours.In the example shown, for reasons of simplification, it is defined that: - the first RRU, RRU1, is deactivated at t1, - the second RRU, RRU2, is deactivated at t2, - …, - the Kth RRU, RRUK, is deactivated at tK. It goes without saying that this example is not exhaustive. It could be defined for example that: - the third RRU, RRU3, is deactivated at t1, - the Kth RRU, RRUK, is deactivated at tK, - …, - the first RRU, RRU1, is deactivated at tK-1, - etc. In this second embodiment, it is considered that the radio site S operates with ^^ RRUs, where K≥1 and (^^1, ^^2, … , ^^^^) are the instants of deactivation of the ^^ RRUs. According to the embodiment shown, the DCC control device is configured to determine in a single pass the deactivation times (^^, ^^2, …,. auof the PO power outage, making it possible to maximize the autonomy of the STO energy storage device during the duration Δ^^ of the PO power outage. As in the first embodiment of FIG. 5A, a set of data relating to the operation of the radio site S, as used in the calculation S2, comprises the following data: - ^^0 the energy stored in the STO energy storage device at the time of the PO power outage, - Δ^^ the duration of the PO power outage, included for example in the IPO information, - ^^^^ = ^^^^,2^^ + ^^^^,3^^ + ^^^^,4^^ the volume of traffic carried at the time of the PO power outage, for a kth RRU among K, during a time interval ^^ , where^^ ^^,^^^^ is the KPI corresponding to the volume of traffic of the network generation nG collected at the time of the outage, where ^^ is for example such that 2 ≤ ^^ ≤ 4.- ^^ ^^the total energy consumed by the site S, as estimated at S0. For this purpose, the method comprises, at the time of the power cut PO, a reception step S200, during which the control device DCC receives, via its communication interface COM: - an estimated value of the energy ^^ ^^ consumed by site S before the power cut PO, from the DME calculation device, - an energy value ^^ ^^ stored in the energy storage device at the time of the PO outage, coming from a communication terminal associated with the site S or from the STO energy storage device if the latter is equipped with a dedicated communication module, - a value of the duration Δ^^ of the power outage PO, coming from a communication terminal associated with the site S or from an electricity supplier FE with which the site S is referenced. Of course, as an alternative, the estimated value of the energy ^^ ^^ , the value of energy ^^^^ and value of the duration Δ^^ of the power outage PO can be received simultaneously or each at different times. Also as in the first embodiment of FIG. 5A, the method according to the second embodiment further comprises, at the time of the power outage PO, a step S201 of collecting data DR relating to the operation of said RC network, such as the value of the traffic volume ^^^^ = ^^^^,2^^ +^^^^,3^^ + ^^^^,4^^, in the example shown, energy consumption indicators associated with the kth RRU, etc. During a step S202, the calculation module CAL then calculates the time ti+1 of deactivation of the i+th RRU, RRUi+1, as follows:^^^^+1 = ^^^^ − ^^ℎ + ^^^^ (9)- where ^^ ^^ is the previous instant of deactivation of the ith RRU, RRUi, - where ^^ ^^ is the autonomy of the STO energy storage device, calculated at ^^ ^^. During a step S203, the calculation module CAL calculates the remaining energy at the energy storage device STO, as follows: - where E(i) ^ ^ is the remaining energy at the STO energy storage device, which was calculated at ^^ ^^ . A ^^ ^^+1 , the i+1 th RRU, RRUi+1 being deactivated, the traffic or data volume associated with it is distributed between the RRUs not yet deactivated, i.e. RRUi+2, …, RRUK. During a step S204, the calculation module CAL calculates the energy consumed by each of the RRUs, RRUi+2 to RRUK, from the aforementioned expression (3): During a step S205, the calculation module CAL calculates the autonomy of the energy storage device STO at ^^ ^^+1 according to the following relationship: S202 S205 are iterated for remaining RRUs, i.e. RRUi+2 to RRUK. The last time tK of deactivation of RRUK is then calculated as follows in S202: At the initialization of the calculation according to steps S202 to S205, it is considered that the residual energy of the energy storage device STO is such that E (0)^^ = ^^0 to ^^0 = 0, where ^^0 corresponds to the instant of the power outage PO. In addition, at ^^0, a parameter ^^ for re-parameterizing the threshold ^^ℎ is initialized, such that 0 < ^^ < 1 and such that if ^^0 < ^^ℎ, then ^^ℎ = ^^^^0. We now describe, in relation to Figure 6, a schematic curve illustrating the strategy for deactivating K RRUs in accordance with the calculation S2 implemented according to the second embodiment, which has just been described above. In the example shown, K=4. Figure 6 shows from which autonomy threshold ^^ℎ of the energy storage device STO, it is necessary to deactivate a first RRU, RRU1, then a second RRU, RRU2, then a third RRU, RRU3, then possibly a fourth RRU, RRU4. Such a deactivation strategy implies that during a PO power failure, the autonomy of the STO energy storage device should be continuously monitored, as shown in Figure 6.Thus, when the autonomy of the STO energy storage device reaches the threshold ^^ℎ, RRU1 is deactivated. This increases the autonomy of the STO energy storage device. Then, the autonomy of the STO energy storage device is checked again and RRU2 is deactivated when the autonomy of the STO energy storage device reaches the threshold ^^ℎ, and so on as shown in Figure 6. Note that the fourth RRU, RRU4, may not be deactivated when the autonomy of the STO energy storage device reaches the threshold ^^ℎ, as symbolized by the dot-dash circle in Figure 6. The fourth RRU4 may indeed remain activated until the STO energy storage device is empty. As shown in Figure 6, at the time of the power outage PO, the RRUs RRU1 to RRU4 are all active. In S205, the autonomy of the energy storage device STO is calculated as follows from the aforementioned relation (11):. In S202, deactivation of RRU1 is then calculated as follows, from the aforementioned relation (9): ^^1 = ^^0 − ^^ℎA , the autonomy of the STO energy storage device reaches the value of ^^ℎ. The remaining energy (in joules) at the level of the STO energy storage device is then calculated in S203 as follows, from the aforementioned relation (10): As already explained above in relation to Figure 5B, when an RRU is deactivated, the assumption is made that its traffic or data volume is distributed equally over the remaining RRUs, provided that the maximum capacity, in terms of traffic, of each of these remaining RRUs is not exceeded. Such an assumption is well-founded since a positive error on the traffic estimation added on one RRU will be corrected by a negative error on the traffic estimation added on another RRU. In addition, according to Figure 4, the fixed part of the RRU is the one that consumes the most. This means that a slight error on the estimation of the new traffic distribution following a deactivation of an RRU does not have a great impact on the estimation of the energy consumed by the RRU. The indicators (KPIs or raw indicators) related to the traffic on an active RRU increase at the time of the deactivation of another RRU.For the other KPIs, it is assumed that they increase proportionally in the same way as the traffic-related indicators. It is also possible to use a learning model, based for example on linear regression, which expresses each indicator (KPI (apart from traffic) or raw indicator) as a function of the traffic transmitted on the active RRU. Using the new values ​​of the different indicators following a deactivation of an RRU, a recalculation of the energy consumed by the RRUs still active is implemented in accordance with equation (3). At ^^2, the autonomy of the STO energy storage device reaches the value of ^^ℎ. The remaining energy (in joules) at the STO energy storage device is then calculated in S203 as follows, from the aforementioned relation (10):. Once RRU2 is deactivated at ^^2, the new values ​​of the energy ^^^^, 3 ≤ ^^ ≤ ^^ are calculated in S204 from the aforementioned equation (3) and the distribution of the traffic from RRU2 to the other RRU3 to RRU4 still active. In S205, the autonomy of the energy storage device STO is calculated again as follows from the aforementioned equation (11): In S202, the instant ^^3 of RRU deactivation 3 is then calculated as follows, from the aforementioned relation (9): ^^3 = ^^2 − ^^ℎ + ^^2A ^^3, the autonomy of the STO energy storage device reaches the value of ^^ℎ. The remaining energy at the STO energy storage device is then calculated in S203 as follows, from the aforementioned relation (10): Once RRU3 is deactivated at ^^3, a new value of the energy ^^4 is calculated in S204 from the aforementioned equation (3) and the transfer of traffic from RRU3 to the still active RRU4. In S205, the autonomy of the STO energy storage device is calculated again as follows from the aforementioned equation (11): In S202, the instant ^^4 of deactivation of RRU4 is then calculated as follows, from the aforementioned relation (12): ^^4 = ^^3 + ^^3Although in relation to FIGS. 5A, 5B and 6, the step of calculating the K instants of deactivation has been described in the context of an RRU by RRU deactivation strategy, it goes without saying that such a step can also be implemented, correspondingly, in the context of a deactivation strategy, and without departing from the scope of the invention: - frequency band by frequency band, - generation of cellular radiocommunication network by generation of cellular radiocommunication network, - set of RRUs by set of RRUs, a set possibly containing one or more RRUs.The invention just described above not only ensures the continuity of the services provided by the radio site S, but also optimizes the use of energy, thus contributing to a more efficient and sustainable management of communication infrastructures. The CAL calculation module operates according to an ALGO algorithm, an example of which is shown above. In this example, the algorithm is written in a generic algorithmic language, adaptable to any programming language, such as for example Python, R, C++, Java, Scala, MATLAB, etc. ALGO algorithm: (1) Inputs: ^^0, Cutoff plan, Th, ^^ , ^^. ^^,^^ (1≤k≤K) (2) Inputs: KPIs and ^^ ^^ (1≤k≤K) at the event of power outage and ω (3) Initialization: E (0) ^^ = ^^0, ^^0 = 0(4) Calculate ^^0based on Eq. (11 ) (5) If Th > A0, then ^^ℎ = ^^^^0(6) For i from 0 to K – 2 (7) Use Eq. (9) to calculate ^^ ^^+1 (8) Use Eq. (10) to calculate (9) #Comment: Strategy of mobility (10) Calculate (11) Calculate the vector ^^^^ (12) Find the set S1=find (^^^^ <^^0) (13) Vk=^^ ^^,^^ for all k∈S1 (14) Find the set S2=find (^^^^ ≥^^0) (15) Vk=Vk + ^^0for all k∈S2 (16) u=sum[^^^^ (k)] for all k∈S1 (17) c=cardinal(S2) #Comment: Length S2 (19) While [(a>0) & (c≥1)] (20) Calculate the vector ^^^^ #Comment: of size c (21) Find the set S1=find (^^^^ ≤ ^^) (22) Vk=^^ ^^,^^ for all (23) Find the set S2=find (^^^^ > ^^) (24) Vk=Vk + ^^ for all k∈S2 (25) ^^ =sum[^^^^(k)] for all k∈S1 (26) ^^ =cardinal(S2)(27) ^^ =^^- ^^ - ^^ ^^; ^^ =^^ ^^ (28) End While (29) Update the other KPIs for RRU i+2 to K (30) #Comment: end of the strategy of mobility (31) Calculate the new expressions of ^^ ^^ based on Eq. (3) (32) Calculate ^^ ^^+1 based on Eq. (11 ) (33) End For (34) Calculate t K based on Eq. (12)A la ligne (10), is the value of the RRU traffic volume (^^ + 1) which is disabled at^^ ^^+1 , this value being averaged over all remaining RRUs. In line (11), ^^^^ is a vector that calculates the remaining capacity (in terms of traffic volume) of RRUs (^^ + 2) to ^^. Then, ^^^^ = ^^^^,^^ − ^^^^ for ^^ + 2 ≤ ^^ ≤ ^^. As shown in line (12), if ^^^^(^^) < ^^0, then RRUk belongs to a set ^^1. This means that the RRU is not able to accept the entire quantity ^^0 because it exceeds its capacity. In this case, as shown in line (13), ^^^^ = ^^^^,^^ . Conversely, in line (14), if the remaining capacity of RRUk is greater than (^^^^(^^) ≥ ^^0), then RRUk belongs to a set ^^2. In this case, as shown in line (15), ^^^^ = ^^^^ + ^^0. In lines (16) and (17), an update ^^ of the traffic KPIs of the RRUs belonging to ^^1 and an update ^^ of the traffic KPIs of the RRUs belonging to ^^2 are implemented. The remaining traffic of the RRUs (^^ + 1) not yet distributed is given in line (18): ^^ =^^^^+1 − ^^ − ^^^^0 , where ^^ and ^^ are given in lines (16) and (17) respectively. In line (18), ^^ is redefined, such that ^^ =^^ ^^ , as the remaining traffic of the RRU(^^ + 1) averaged over the remaining RRUs that still have the capacity to absorb the traffic (RRUs of the set ^^2). In line (19), as long as ^^ > 0, which means there is still undelivered traffic, and ^^ ≥1, which means there are still one or more RRUs that still have capacity, the While loop between lines (19) and (28) is activated to deliver the remaining traffic of the RRU (^^ + 1).As shown in line (24), the algorithm finds the new value of ^^. ^^ for an RRU ^^, such that (1≤k≤K). ^^ ^^ is composed of traffic from the 2G, 3G and 4G layers. In the same way as the algorithm just described, the distribution of ^^ ^^ on all layers is also performed iteratively without exceeding the maximum traffic of each layer. For simplicity, we do not add this step in this algorithm. In line (29), an update of the KPIs of the remaining RRUs, RRUs (^^ + 2) to K is implemented. In line (31), ^^ ^^ is again recalculated from the above equation (3). In line (32), the autonomy ^^ ^^+1 of the energy storage device STO is calculated from expression (11). At line (34), the algorithm ends with the calculation of the instant ^^ ^^ deactivation of the last RRU, RRUK, from equation (12).

Claims

CLAIMS

1. Method for controlling the energy consumption of a cellular radiocommunication site during a power outage, said site being provided with an energy storage device (STO) and operating according to at least one frequency band and / or at least one generation of cellular radiocommunication network, said method comprising: - an activation (S1) of the storage device, - a calculation (S2) of a time of deactivation of said at least one frequency band or of said at least one generation of network, said calculation being implemented according to a criterion for optimizing the energy stored in the storage device during the outage, - a deactivation (S3), at said calculated time of deactivation, of said at least one frequency band or of said at least one generation of network.

2. Method for controlling the energy consumption of a cellular radiocommunication site during a power outage according to claim 1, wherein when said site operates according to at least two frequency bands and / or at least two generations of cellular radiocommunication network, said method comprises, once the storage device is activated: - a calculation, according to said optimization criterion, of at least two successive instants of deactivation of respectively said at least two frequency bands or of respectively said at least two generations of network, - a deactivation, at said at least two calculated instants of deactivation, of respectively said at least two frequency bands or of respectively said at least two generations of network.

3. Method for controlling the energy consumption of a cellular radiocommunication site according to claim 2, in which the at least two frequency bands or the at least two network generations are deactivated according to a priority order which is calculated, said calculation of the at least two successive instants of deactivation according to the calculated priority order being further implemented according to a criterion of maximizing the volume of data traffic during the cut-off.

4. A method for controlling the energy consumption of a cellular radiocommunication site according to claim 3, wherein for each of the at least two frequency bands or the at least two network generations, the order of priority (^^ ^^) is calculated based on a ratio between the energy consumption and the data traffic volume determined before the cut-off for each of the at least two frequency bands or the at least two network generations.

5. A method for controlling the energy consumption of a cellular radiocommunication site according to claim 2, wherein the at least two frequency bands or the at least two network generations are deactivated according to a priority order which is predefined.

6. Method for controlling the energy consumption of a cellular radiocommunication site according to claim 5, in which said at least two deactivation instants, from which the at least two frequency bands or the at least two network generations are respectively deactivated according to said predefined priority order, are calculated as a function of a threshold ( ^^ℎ ) of autonomy of the energy storage device, which is crossed for each of the at least two frequency bands or the at least two network generations.

7. Method for controlling the energy consumption of a cellular radiocommunication site according to any one of claims 1 to 6, in which said instant of deactivation or said at least two successive instants of deactivation are calculated as a function of a set of data of said site, said set of data comprising the energy stored in the energy storage device at the time of the outage, the duration of the outage, at least one data item relating to the operation of said cellular radiocommunication network at the time of the outage, and the energy consumed by the site at the time of the outage.

8. A method for controlling the energy consumption of a cellular radiocommunication site according to claim 7, wherein the energy consumed by the site at the time of the outage is determined using an artificial intelligence model trained from indicators representative of the operation of the site, collected at different time intervals.

9. A method for controlling the energy consumption of a cellular radiocommunication site according to claim 8, wherein the modeling of. the energy consumed at the time of the outage is implemented using a Lasso linear regression method.

10. Device (DCC) for controlling the energy consumption of a cellular radiocommunication site during a power outage, said site being provided with an energy storage device and operating according to at least one frequency band and / or at least one generation of cellular radiocommunication network, the device being configured to implement: - an activation of the storage device, - a calculation of a time of deactivation of said at least one frequency band or of said at least one generation of network, said calculation being implemented according to a criterion for optimizing the energy stored in the storage device during the outage, - a deactivation, at said calculated time of deactivation, of said at least one frequency band or of said at least one generation of network.

11. A computer program comprising program code instructions for implementing the control method according to any one of claims 1 to 9, when executed on a computer.

12. A computer-readable information medium comprising instructions of a computer program according to claim 11.

Images