Method for estimating a load of a beam formed in a cell of a wireless cellular access network

Abstract

The present disclosure relates in particular to a method (20), carried out by a control device of a wireless cellular access network, for estimating a load of a beam of interest from among a plurality of beams able to be formed so as to serve user terminals in a cell of said wireless cellular access network, said estimating method (20) comprising: - determining (S20) a load indicator M E (T) in relation to the cell for a time window T, based on information regarding the use, during the time window, of communication resources throughout the cell, - determining (S21) a load indicator M b (T) in relation to the beam of interest for the time window T, based on information regarding the use, during the time window, of communication resources in the beam of interest, - estimating (S22) the load of the beam of interest based on the product M E (T) × M b (T).

Description

Description Title: Method for estimating a load of a beam formed in a cell of a wireless cellular access network Technical field

[0001] The present invention belongs to the field of communication systems, and more particularly relates to a method for estimating a load of a beam of interest among a plurality of beams formed in a cell of a wireless cellular access network, as well as methods for configuring communications between the wireless cellular access network and a user terminal, using estimated beam loads. Prior art

[0002] In current and future wireless cellular communication systems, it is planned to form a plurality of beams within each cell. For this purpose, a base station is typically equipped with a plurality of antennas and, by applying respective complex coefficients to the different antennas of the base station, it is possible to form different beams, i.e. to form different radiation patterns allowing, for example, spatial multiplexing of different user terminals within the same cell.

[0003] When the number of antennas becomes large, we speak of a massive multi-antenna system or massive MIMO ("massive MIMO" in the English literature, where the acronym MIMO stands for "Multiple Input Multiple Output"). Such massive MIMO systems are particularly considered in 5G and later communication systems.

[0004] In such massive MIMO systems, it is for example possible to use a grid of beams (GoB). Such GoBs are widely used in industry for control channels and / or data channels. The beams of a GoB are not adaptive, i.e. they are not specifically optimized to exchange data with a specific user terminal, but rather to serve determined geographical areas within the cell. Adaptive beamforming is also possible, especially for data channels (using for example techniques known as "eigen-based beamforming" in the English literature).

[0005] By using beams, performance and quality of service (QoS) indicators can be defined with greater resolution, at the beam level. In particular, load, which is a central indicator for wireless access networks, can be defined per beam.

[0006] Different definitions have been proposed to estimate the load of a beam of interest. For example, the 3GPP TS 38.423 V17.1.0 technical specification defines a load estimation function of a beam of interest, which corresponds to a ratio between the number of physical resource blocks (PRB in the 3GPP specifications) used by the beam b of interest and the number of PRBs used by all the beams included in the cell: [Math. 1] expression in which: B corresponds to the number of beams (for example of a GoB), L corresponds to the number of successive time intervals (“slots” or “mini-slots” in the 3GPP specifications) considered to estimate the load of the beam b of interest, - ij,, corresponds to an indicator function which is equal to 1 if the beam b' (1 < b' < B) is activated on the time interval i (1 < i < L), and which is equal to 0 otherwise, ■ N PRB,br corresponds to the number of PRBs used in beam b' (1 < b' < B) during time interval i (1 < i < L) (note that in 3GPP specifications the number N P l RB b , is the same for all beams of the same cell activated simultaneously, i.e. N P l RB b , = N PRB Vb').

[0007] However, such an estimation function for the load of a beam of interest is not robust. Indeed, this estimation function is based on the use of communication resources (PRB) in the beam of interest, and the communication resources are actually used only when the beam of interest is activated (scheduled). Thus, if the beam of interest is rarely activated, then the estimated load of this beam of interest will be low, so that this beam of interest could be considered as capable of carrying more traffic. However, the usable communication resources (PRB) are allocated to the whole cell. Therefore, the beam of interest can very well be rarely activated while it has a lot of traffic to carry, because the other beams in the cell also have a lot of traffic to carry using the same communication resources.Indeed, the activations of the beam of interest are then more spaced out in time to allow other beams that also have traffic to carry to be activated more often. In such a case, the load of the beam of interest, estimated with the estimation function above, would be low even though this beam of interest cannot carry more traffic (because it already has a lot of traffic to carry and because it cannot be activated more frequently because of the other beams).

[0008] Figure 1 schematically represents scenarios illustrating the lack of robustness of the estimation function given by the expression [Math. 1]. Figure 1 schematically represents a cell 12 served by a base station 11 which can form, in this cell 12, seven (7) different beams 13-1 to 13-7, for example via a GoB. In Figure 1, the beam of interest is beam 13-4. In part a) of Figure 1, the traffic to be carried by beam 13-4 is significant, and the traffic to be carried in the other beams 13-1 to 13-3, 13-5 to 13-7 is much less significant than in beam 13-4. In such a case, the load estimated by means of the expression [Math. 1] for beam 13-4 is high. In part (b) of Figure 1, the traffic to be carried by bundle 13-4 is unchanged, but the traffic to be carried in the other bundles has increased. With the expression [Math.1], the estimated load for beam 13-4 for part b) of Figure 1 is lower than that estimated for part a) of Figure 1, because beam 13-4 is activated less frequently (due to the increase in traffic to be carried in the other beams 13-1 to 13-3, 13-5 to 13-7), even though the traffic to be carried by beam 13-4 has not decreased and the increase in traffic to be carried by the other beams 13-1 to 13-3, 13-5 to 13-7 prevents beam 13-4 from being activated more frequently. Summary

[0009] The present disclosure aims to remedy all or part of the limitations of the solutions of the prior art, in particular those set out above, by proposing a solution which makes it possible to improve the estimation of the load of a beam of interest among a plurality of beams formed in a cell of a wireless cellular access network.

[0010] To this end, the present disclosure relates to a method for estimating, by a control device of a wireless cellular access network, a load of a beam of interest among a plurality of beams that can be formed to serve user terminals in a cell of said wireless cellular access network, said estimation method comprising: a determination of a load indicator M E (T) of the cell for a time window T, based on information about a use, during the time window, of communication resources in the entire cell, a determination of a load indicator M b (T) of the beam of interest for the time window T, based on information about a use, during the time window, of communication resources in the beam of interest, an estimate of the load of the beam of interest based on the product M E (T) x M b (T).

[0011] It should be noted that the charge indicator M E T) of the cell and the charge indicator M b (T) of the beam of interest, both referred to as "charge indicator", are therefore consistent in the convention adopted to represent the charge of the cell and the beam respectively. In other words, if a high value of the charge indicator M E T) of the cell corresponds to a high charge level in the cell, then a high value of the charge indicator M b (T) of the beam of interest also corresponds to a high charge level in the beam of interest.

[0012] Thus the estimated load of the beam of interest takes into account not only a load indicator M b (T) of said beam of interest, but also of a charge indicator M ET) of the cell, that is to say of all the beams of said cell. Considering for example that a high value of these charge indicators corresponds to a high charge level then, even if the charge indicator M b (T) of the beam of interest decreases due to the fact that said beam of interest is activated less often, this decrease is compensated by the increase in the charge indicator M E T) of the cell. Thus, the product M E (T) x M b ( ) may increase if the other beams have to carry very heavy traffic, so that the estimated load of the beam of interest is then considered high, which is the expected behavior for the estimation function since the beam of interest in this case does not have the capacity to carry more traffic in this cell of the wireless cellular access network, over the time window.

[0013] Thus, the solution proposed for estimating the load of a beam of interest, among a plurality of beams formed in a cell, is more robust than the solutions of the prior art. The load of the beam of interest thus estimated can be used to improve the performance of certain existing functions, or even to allow the emergence of new functions. For example, the load of the beam of interest can be used by radio resource management procedures (RRM in the English literature) or by self-organizing network functions (SON in the English literature) such as: beam-based mobility robustness optimization (bMRO in the English literature), beam-level mobility load balancing (bMLB).

[0014] The bMRO and bMLB functions are the beam-wide extension of the cell-wide SON functions defined in the 3GPP LTE (4G) specifications, referred to as MRO and MLB, respectively. The loads used for bMRO mobility optimization or bMLB load balancing are therefore estimated at the beam level and not at the cell level.

[0015] In particular modes of implementation, the estimation method may also include, optionally, one or more of the following characteristics, taken individually or in all technically possible combinations.

[0016] In particular modes of implementation, the load indicator M E (T) of the cell corresponds to the value of a first monotonic function with the use of communication resources in the whole cell, and the load indicator M b(T) of the beam of interest corresponds to the value of a second function, said second function being, at constant use of communication resources in the other beams among the plurality of beams, monotonic with the use of communication resources in the beam of interest, of the same monotony as the first function.

[0017] In particular modes of implementation, the determination of the load indicator M b (T) of the beam of interest for the time window T comprises determining a rate of use, in the beam of interest, of communication resources usable in the beam of interest during said time window.

[0018] In particular implementation modes, the determination of the utilization rate of the communication resources usable in the beam of interest during the time window takes into account time intervals during which said beam of interest is activated.

[0019] In particular embodiments, the wireless cellular access network uses orthogonal frequency division multiple access, in which the usable communication resources correspond to physical resource blocks, called PRBs, and the load indicator M b (T) of the beam of interest for the time window T is determined according to the following expression: expression in which: l bcorresponds to an indicator function which is equal to 1 if the beam b of interest is activated on a time interval i of the time window, the time window comprising L time intervals, and which is equal to 0 otherwise, N PRB,b corresponds to a number of PRBs used in the beam b of interest over the time interval i of the time window, ■ N pR l B,b corresponds to a maximum number of PRBs that can be used in the beam b of interest over the time interval i of the time window.

[0020] In particular modes of implementation, the determination of the load indicator M E (T) of the cell for the time window T comprises determining a rate of use, in the cell, of communication resources usable in said cell during the said time window.

[0021] In particular embodiments, the wireless cellular access network uses orthogonal frequency division multiple access, in which the usable communication resources correspond to physical resource blocks, called PRBs, and the load indicator M E (T) of the cell for the time window T is determined according to the following expression: expression in which: - Rf,i(T) corresponds to a number of PRBs multiplexed by each of the f spatial streams over a time interval i of the time window, the time window comprising L time intervals, P;(r) corresponds to a maximum number of PRBs that can be used over time interval i for a single spatial flow in the cell, LM(T) corresponds to a temporal average, over the time window, of a maximum number of spatial flows that can be used.

[0022] In particular modes of implementation, the load of the beam of interest is determined according to the following expression: K being a normalization constant, or according to the following expression:

[0023] According to a second aspect, there is provided a method of configuring, by a control device of a wireless cellular access network, communications with user terminals, said wireless cellular access network comprising base stations serving a plurality of cells, a plurality of beams being able to be formed in each cell, said configuration method comprising: an estimation, by implementing an estimation method according to any one of the embodiments of the present disclosure, of respective loads of beams of cells of the wireless cellular access network, - use of estimated beam loads to configure communications with user terminals.

[0024] According to a third aspect, there is provided a control device included in a wireless cellular access network, said control device comprising at least one memory and at least one processor configured to implement a method load estimation or a configuration method according to any of the embodiments of the present disclosure.

[0025] According to a fourth aspect, there is provided a wireless cellular access network comprising base stations serving a plurality of base station cells, a plurality of beams being formable in each cell, said wireless cellular access network comprising at least one control device according to any of the embodiments of the present disclosure.

[0026] According to a fifth aspect, there is provided a method of configuring, by a user terminal, a communication with a wireless cellular access network comprising base stations serving a plurality of base station cells, a plurality of beams being able to be formed in each cell, said configuration method comprising: receiving respective beam loads of one or more cells of the wireless cellular access network, said beam loads being estimated by implementing an estimation method according to any of the embodiments of the present disclosure, - use of the estimated beam loads to configure communication with the wireless cellular access network.

[0027] According to a sixth aspect, there is provided a user terminal for exchanging data with a wireless cellular access network, said user terminal comprising at least one memory and at least one processor configured to implement a configuration method according to any one of the embodiments of the present disclosure.

[0028] According to a seventh aspect, there is provided a computer program product, comprising a set of program code instructions which, when executed by at least one processor, configure said at least one processor to implement a method according to any of the embodiments of the present disclosure. Brief description of the drawings

[0029] The invention will be better understood by reading the following description, given as a non-limiting example, and made with reference to the figures which represent: - [Fig. 1] Figure 1: already described, a schematic representation of two scenarios of use of beams in a cell of a wireless cellular access network, [Fig. 2] Figure 2: a diagram illustrating the main steps of an example of implementation of a method for estimating the load of a beam of interest, [Fig. 3] Figure 3: a schematic representation of an example of the realization of a device for controlling a wireless cellular access network, for implementing the method for estimating the load of a beam of interest, - [Fig. 4] Figure 4: simulation results illustrating the advantages of the estimation method according to the present disclosure, [Fig. 5] Figure 5: A diagram illustrating the main steps of an example of implementation of a method for configuring communications by a wireless cellular access network, [Fig. 6] Figure 6: a diagram illustrating the main steps of an example of implementation of a method for configuring communications by a user terminal, [Fig. 7] Figure 7: a schematic representation of an example of the implementation of a user terminal.

[0030] In these figures, like references from one figure to another designate identical or similar elements. For reasons of clarity, the elements shown are not to scale, unless otherwise indicated.

[0031] Furthermore, the order of steps shown in these figures is given only as a non-limiting example of the present disclosure which can be applied with the same steps performed in a different order. Description of the embodiments

[0032] Figure 2 represents the main steps of a method 20 for estimating load, by a device 30 for controlling a wireless cellular access network, of a load of a beam of interest among a plurality of beams that can be formed to serve user terminals 70 in a cell of said wireless cellular access network.

[0033] As illustrated in Figure 3, the control device 30 comprises for example at least one processor 31 and at least one memory 32 (magnetic hard disk, electronic memory, optical disk, or any type of computer-readable recording medium) in which a computer program product is stored, in the form of a set of program code instructions to be executed to perform all or part of the operations to be performed by said control device 30. In certain cases, the control device 30 may optionally comprise one or more programmable logic circuits (FPGA, PLD, etc.), and / or one or more specialized integrated circuits (ASIC, etc.), and / or a set of discrete electronic components, etc., adapted to perform all or part of the operations to be performed by said control device 30.The control device 30 is for example included in one or more base stations 11 and / or is connected to one or more base stations 11 (for example integrated in whole or in part in a radio network controller).

[0034] As illustrated in Figure 2, the estimation method 20 comprises steps of: 520 determination of a load indicator M E T) of the cell for a time window T, 521 determination of a load indicator M b (T) of the beam of interest for the time window T, 522 estimation of the load of the beam of interest as a function of the load indicator M E (T) of the cell and the charge indicator M b (T) of the beam of interest.

[0035] The M charge indicator E(T) of cell 12 is determined, during step S20, based on information on a use, during the time window T, of communication resources in the entire cell 12. The load indicator M b (T) of the beam of interest is determined, during step S21, based on information on a use, during the time window T, of communication resources in the beam of interest. The information on the use of communication resources in the cell or in the beam of interest is for example provided by a scheduler of the wireless cellular access network.

[0036] As shown above, the charge indicator M E T) of the cell and the charge indicator M b(T) of the beam of interest, both referred to as "charge indicator", are consistent in the convention adopted to represent the charge of the cell and the beam respectively.

[0037] In particular modes of implementation, the load indicator M E T) of cell 12 corresponds to the value of a first determined function, and the load indicator M b (T) of the beam of interest corresponds to the value of a second determined function. If applicable, the first function is monotonic (i.e., increasing or decreasing) with the use of communication resources throughout the cell. For example, if the first function is increasing, then the load indicator M E(T) of cell 12 increases with the utilization of communication resources in the whole cell, i.e., if the utilization of communication resources in the whole cell is higher during time window T than in a previous time window, then the load indicator M E T) of cell 12 determined for time window T is greater than that determined for the previous time window. The second function is, at constant use of communication resources in the other beams among the plurality of beams, monotonic with the use of communication resources in the beam of interest. In other words, if the traffic to be carried in the other beams does not vary from one time window to another, then the load indicator M b(T) of the bundle of interest is monotonic with the utilization of communication resources in the bundle of interest. For example, if the second function is increasing (resp. decreasing) then, if the use of communication resources in the beam of interest is higher (resp. lower) during the time window T compared to a previous time window, and if in addition the use of resources by the other beams is the same during the time window T and during the previous time window, then the load indicator M b (T) of the beam of interest determined for the time window T is higher (resp. lower) than that determined for the previous time window.

[0038] Note that the second function is of the same monotonicity as the first function. In other words, if the first function is an increasing function, then the second function is also an increasing function. Alternatively, if the first function is a decreasing function, then the second function is also a decreasing function. The first function and the second function are of the same monotonicity to ensure that the load indicator M E (T) of cell 12 and the charge indicator M b (T) of the beam of interest are consistent in representing the load. The choice of increasing or decreasing monotonicity for the first and second functions depends on the convention adopted for the load indicator M E T) of cell 12 and for the charge indicator M b (T) of the beam of interest. Typically, with increasing monotony, a high value of the corresponding indicator (M E (T) or Mb (T)) corresponds to a high load level. With decreasing monotony, a high value of the corresponding indicator (M B (T) or M b (T)) corresponds to a low load level and is therefore rather representative of a capacity to be able to charge the cell or the beam of interest more. In the remainder of the description, it is considered in a non-limiting manner that the first function and the second function are both of increasing monotony (for the second function: with constant use of communication resources in the beams of the cell other than the beam of interest).

[0039] The estimated load of the beam of interest is determined, during step S22, as a function of the product M E T) x M b (T). In the non-limiting example considered, in which the first function and the second function are both increasing, then the load indicator M E(T) of the cell increases when the amount of communication resources used in the other beams increases. Thus, even if the load indicator M b (T) of the beam of interest decreases due to the fact that said beam of interest is activated less often, this decrease is compensated by the increase in the charge indicator M E T) of the cell. Thus, the product M E (T) x M b (T) may increase if other beams have to carry more traffic, so that the estimated load of the beam of interest is then considered high, which is the expected behavior for the estimation function since the beam of interest in this case does not have the capacity to carry more traffic in this cell, over the time window.

[0040] It should be noted that the product M E T) x M b(T) is used when considering a linear scale, and such a product becomes a sum if a logarithmic scale is considered. In other words,

[0041] In the remainder of the description, we consider, in a non-limiting manner, the case where the wireless cellular access network uses orthogonal frequency division multiple access, known as OFDMA (Orthogonal Frequency Division Multiple Access in the English literature), in which the usable communication resources correspond to physical resource blocks, known as PRBs. 4G and 5G wireless cellular communication systems, in particular, are OFDMA-type systems. In 4G communication systems, PRBs correspond to time-frequency blocks, each PRB extending over a time interval of 0.5 ms (slot in the 3GPP 4G specifications, which comprises 7 OFDM symbols) and over 12 subcarriers (subcarriers in the 3GPP specifications).In 5G communication systems, PRBs correspond to frequency blocks of 12 subcarriers, each PRB being able to be allocated over a time interval ("slot" or "mini-slot" in the 3GPP 5G specifications) of variable duration which depends in particular on the spacing between the subcarriers. For example, if the time interval corresponds to a "slot", the duration of the time interval is between 0.125 ms (for a spacing between subcarriers of 120 kHz) and 1 ms (for a spacing between subcarriers of 15 kHz). It is also considered in a non-limiting manner that the beams in the same cell cannot all be activated simultaneously.

[0042] We now give non-limiting examples of expressions for the load indicator M E (T) of the cell and for the charge indicator M b (T) of the beam of interest.

[0043] In particular modes of implementation, the load indicator M ET) of the cell for the time window T is for example representative of a rate of use, in the cell, of communication resources usable in the cell during said time window T. For example, this rate of use can be determined according to the following expression: [Math. 2]

[0044] The expression [Math. 2] considers the same notations as those used in section 5.1 .1 .2.1 1 of the 3GPP TS 28.552 V17.7.1 specification for the indicator "PDSCH PRB usage per cell for MIMO". More specifically: - Rf,i(T) corresponds to a number of PRBs multiplexed by each of the f spatial streams (also known as "spatial streams" or "spatial layers" in the English literature, "MIMO layers" in the 3GPP TS 28.552 V17.7.1 specification) on a time interval i ("slot" or "mini-slot") of the time window, the time window T comprising L time intervals; if for example there are only two spatial streams (f = 2) on the time interval i of the time window T, then all the R^ T) are null except R2,i(T > ), P i (T') corresponds to a maximum number of PRBs that can be used over time interval i for a single spatial flow in the cell, LM(T) corresponds to a temporal average, over the time window, of a maximum number of spatial flows that can be used.

[0045] For example, it is possible to use the same expression as given in the 3GPP TS 28.552 V17.7.1 specification, namely: or a standardized value: expression in which |%J corresponds to the integer part of x.

[0046] More generally, the charge indicator M E (T) of the cell is for example given by the expression: expression in which K corresponds to a determined coefficient, preferably an integer, or is more simply given by or determined as a function of the expression:

[0047] However, other expressions are possible for the load indicator M E (T) of cell 12, and in particular any form which is based on a first monotonic function of the use of communication resources in the cell, and the choice of a particular type of first function only corresponds to an implementation variant.

[0048] In particular modes of implementation, the load indicator M b(T) of the beam of interest for the time window T is representative of a utilization rate, in the beam of interest, of communication resources usable in the beam of interest during said time window. Preferably, the utilization rate of the communication resources usable in the beam of interest during the time window takes taking into account the time intervals during which said beam of interest is activated. For example, this utilization rate can be determined according to the following expression: [Math. 3] expression in which: l b corresponds to an indicator function which is equal to 1 if the beam b of interest is activated on a time interval i (“slot” or “mini-slot”) of the time window, the time window comprising L time intervals, and which is equal to 0 otherwise, N PRB bcorresponds to a number of PRBs used in the beam b of interest over the time interval i of the time window, ■ N pR l B,b corresponds to a maximum number of PRBs that can be used in the beam b of interest over the time interval i of the time window.

[0049] It should be noted that in the 3GPP specifications the number N P l RB b is the same for all beams of the same cell activated simultaneously, i.e. N PRB b = N PRB Vb. Similarly, in the 3GPP specifications, the number PR a B x is the same for all beams of the same cell activated simultaneously, i.e. br"t p \ a D p x ,U R = N rp n K p a D x ' 1 vb.

[0050] It is therefore also possible, in certain modes of implementation, to have in the expression [Math. 3] the same number N PRB b and the same number N^ B x b l for all beams of the same cell activated simultaneously. In such a case, in the expression [Math. 3], the use of communication resources by the other beams of the cell (other than the beam b of interest) essentially influences the indicator functions 1L b of the beam b of interest.

[0051] For example, it is possible to determine the charge indicator M b (T) of the beam b of interest according to the following expression: or more simply according to the following expression:

[0052] However, other expressions are possible for the load indicator M b) of the bundle b of interest, and in particular any form which is based on a second monotonic function of the use of communication resources in the bundle of interest, with constant use of communication resources in the other bundles, and the choice of a particular type of second function corresponds only to an implementation variant.

[0053] The estimated load is determined based on the product M E T) x M b T), which can be determined, for example, by considering any combination of expressions of M E T) and M b (T) data above.

[0054] Simulation results are now given demonstrating the improvement in the robustness of the estimated load compared to prior art solutions, and more particularly compared to the prior art solution based on the expression [Math. 1] above. For the purposes of comparison, the estimated load of the beam b of interest, hereinafter referred to as p b T), was determined using, in a non-limiting manner, the following expression, which corresponds to a preferred mode of implementation: [Math. 4]

[0055] The simulations were carried out considering a uniform deployment of 7 tri-sector base stations 11 in a macro urban environment (inter-base station distance 1 1 of about 300 m). In these simulations, each cell is served by an antenna array of dimensions 8 x 8 and serves the user terminals 70 using a GoB forming 7 beams, similar to those shown in Figure 1. The multiplexing technique used is OFDMA. For these simulations, a traffic model called File Transfer Protocol (FTP) was also considered according to which user terminals 70 arrive in the wireless cellular access network following a Poisson process. These user terminals 70 must download a 10 MB file and leave the wireless cellular access network when the download is completed.The 70 user terminals attached to the same cell were scheduled using a Proportional Fair (PF) scheduler. In addition, the propagation conditions between the 70 user terminals and the 11 base stations were simulated using the 3GPP channel model.

[0056] In the following, we evaluate the performance of a cell 12 located in the center of the wireless cellular access network. For this cell 12, we consider different arrival rates of user terminals 70 in the coverage areas of the GoB beams. More precisely, we consider two arrival rate patterns: first arrival rate pattern: 0.2 user terminals arrive per second in the coverage area of ​​beams 13-1, 13-2, 13-4, 13-5 and 13-7, while 1.3 user terminals arrive per second in the coverage area of ​​beams 13-3 and 13-6, second arrival rate pattern: 0.5 user terminals 70 arrive per second in the coverage area of ​​beams 13-1, 13-2, 13-4, 13-5 and 13-7 while 1.3 user terminals arrive per second in the coverage area of ​​beams 13-3 and 13-6.

[0057] It should be noted that the first and second arrival rate patterns make it possible to simulate the scenarios represented respectively in parts a) and b) of Figure 1. Indeed, the traffic demand to which the beams 13-3 and 13-6 are subjected is high and does not increase between the first arrival rate pattern and the second arrival rate pattern, whereas the traffic demand to which the beams 13-1, 13-2, 13-4, 13-5 and 13-7 are subjected increases between the first arrival rate pattern and the second arrival rate pattern.

[0058] Figure 4 schematically represents the results obtained in terms of estimated load. More particularly, part a) of Figure 4 represents the estimated loads, using the expression [Math. 1] of the prior art and a time window of 20 seconds, for the first arrival rate pattern and the second arrival rate pattern. Part b) of Figure 4 represents the estimated loads, using the expression [Math. 4] above, for the first arrival rate pattern and the second arrival rate pattern.

[0059] As illustrated in part a) of Figure 4, when moving from the first to the second arrival rate pattern, we observe an overall increase in the loads of beams 13-1, 13-2, 13-4, 13-5 and 13-7, while the estimated loads of beams 13-3 and 13-6 decrease. This illustrates the lack of robustness of the estimation function [Math. 1],

[0060] As illustrated in part b) of Figure 4, when moving from the first to the second arrival rate pattern, we observe a significant increase in the estimated loads of all beams, which shows the robustness of the proposed expression estimation function [Math. 4].

[0061] Figure 5 represents the main steps of a method 50 for configuring, by a device 30 for controlling a wireless cellular access network, communications with user terminals 70. As illustrated by Figure 5, said configuration method comprises steps of: S50 estimation, by implementing a beam load estimation method 20, of respective loads of cell beams of the wireless cellular access network, - S51 use of estimated beam loads to configure communications with user terminals 70.

[0062] As indicated previously, the configuration of communications by the wireless cellular access network corresponds for example to carrying out at least one of: management of the communication resources of the wireless cellular access network (for example within the framework of RRM procedures), - mobility management of user terminals (for example in the context of SON functions, of the bMRO type), load distribution between beams of the same cell and / or between beams of different cells (for example in the context of SON functions, of the bMLB type).

[0063] The loads estimated by means of the estimation method 20 can be implemented, on the wireless cellular access network side, to improve the performance of certain functions used in the wireless communication systems, or even to allow the emergence of new functions. However, it should be noted that such estimated loads, in particular because they are more robust than those estimated in the prior art, can also be used on the user terminals 70 side.

[0064] Figure 6 represents the main steps of a method 60 for configuring, by a user terminal 70, a communication with a wireless cellular access network.

[0065] As illustrated in Figure 7, the user terminal 70 comprises for example at least one processor 71 and at least one memory 72 (magnetic hard disk, electronic memory, optical disk, or any type of computer-readable recording medium) in which a computer program product is stored, in the form of a set of program code instructions to be executed to perform all or part of the operations to be performed by said user terminal 70. In certain cases, the user terminal 70 may optionally comprise one or more programmable logic circuits (FPGA, PLD, etc.), and / or one or more specialized integrated circuits (ASIC, etc.), and / or a set of discrete electronic components, etc., adapted to perform all or part of the operations to be performed by said user terminal.The user terminal 70 also comprises at least one communication module 73 for exchanging data with base stations 11 of the wireless cellular access network. The user terminal 70 is for example a mobile phone, a smartphone, a connected object, a laptop, a tablet, etc.

[0066] As illustrated by FIG. 6, the configuration method 60 implemented by the user terminal 70 comprises steps of: - S60 reception, by the communication module 73, of respective beam loads of one or more cells of the wireless cellular access network, for example beams in which the user terminal 70 is located and / or towards which it moves; said beam loads are estimated by the wireless cellular access network by implementing the estimation method 20, and transmitted to the user terminal 70 by one or more base stations 11, - S61 use of estimated beam loads to configure communication with the wireless cellular access network.

[0067] For example, the configuration of communications by the user terminal 70 corresponds to a use of the estimated beam loads for mobility management of said user terminal 70, in particular to select one beam rather than another to exchange data with the wireless cellular access network. For example, the user terminal 70 can use one or more estimated beam loads: as part of a cell selection mechanism (initial access procedure to the wireless cellular access network): when a user terminal 70 is switched on or finds a coverage area, it performs power measurements on beams transmitted by different cells to select the most suitable cell to serve it;in such a cell selection process, a user terminal 70 could also take into account the load of the beams in order to favor cells with lightly loaded beams, as part of a cell re-selection mechanism (mobility mechanism used when the user terminal 70 is in the inactive mode): when a user terminal 70 is in the inactive mode, it continues to perform periodic power measurements on beams transmitted by different cells to periodically update the attachment of the user terminal 70 (identity of the cell with which the user terminal 70 will exchange if it switches to connected mode);in such a cell re-selection process, a user terminal 70 could also take into account the beam load in order to favor cells with lightly loaded beams, as part of a cell change mechanism (mobility mechanism used when a user terminal is in connected mode ("handover" in the English literature)): in mobile networks, mobility mechanisms make it possible to change the attachment of the user terminals 70 when they detect better radio conditions on a cell other than the one on which they are attached; using the power measurements carried out periodically on the beams of different cells, the user terminals 70 can trigger events making it possible to initiate or interrupt mobility, and such events could also be triggered by taking into account the beam load, etc.;

Claims

Claims

1. Method (20) for estimating, by a device (30) for controlling a wireless cellular access network, a load of a beam (13-4) of interest among a plurality of beams that can be formed to serve user terminals (70) in a cell of said wireless cellular access network, said estimation method (20) comprising: - a determination (S20) of a load indicator M E ( ) of the cell for a time window T, based on information about a use, during the time window, of communication resources in the whole cell, - a determination (S21) of a load indicator M b (T) of the beam of interest for the time window T, based on information about a use, during the time window, of communication resources in the beam of interest, - an estimate (S22) of the load of the beam of interest as a function of the product M E (T) x M b (T).

2. Estimation method (20) according to claim 1, wherein the determination of the load indicator M b (T) of the beam of interest for the time window T comprises determining a rate of use, in the beam of interest, of communication resources usable in the beam of interest during said time window.

3. Method (20) of estimation according to claim 2, in which the determination of the rate of use of the communication resources usable in the beam of interest during the time window takes into account time intervals during which said beam of interest is activated.

4. Estimation method (20) according to any one of the preceding claims, in which the wireless cellular access network uses orthogonal frequency division multiple access, in which the usable communication resources correspond to physical resource blocks, called PRBs, and the load indicator M b (T) of the beam of interest for the time window T is determined according to the following expression: expression in which: - L b corresponds to an indicator function which is equal to 1 if the beam b of interest is activated on a time interval i of the time window, the time window comprising L time intervals, and which is equal to 0 otherwise, ■ N PRB,b corresponds to a number of PRBs used in the beam b of interest over the time interval i of the time window, ■ N pR lB,b corresponds to a maximum number of PRBs that can be used in the beam b of interest over the time interval i of the time window.

5. Method (20) of estimation according to any one of the preceding claims, in which the determination of the load indicator M E (T) of the cell for the time window T comprises determining a rate of use, in the cell, of communication resources usable in said cell during said time window.

6. Estimation method (20) according to any one of the preceding claims, in which the wireless cellular access network uses orthogonal frequency division multiple access, in which the usable communication resources correspond to physical resource blocks, called PRBs, and the load indicator M ET) of the cell for the time window T is determined according to the following expression: expression in which: - Rf't T) corresponds to a number of PRBs multiplexed by each of the f spatial streams over a time interval i of the time window, the time window comprising L time intervals, - P;(r) corresponds to a maximum number of PRBs that can be used over the time interval i for a single spatial flow in the cell, - LM(T) corresponds to a temporal average, over the time window, of a maximum number of spatial flows that can be used.

7. Estimation method (20) according to claim 4 and 6, wherein the load of the beam of interest is determined according to the following expression: or according to the following expression:

8. Method (50) of configuring, by a device (30) for controlling a wireless cellular access network, communications with user terminals, said wireless cellular access network comprising base stations serving a plurality of cells, a plurality of beams being able to be formed in each cell, said method (50) of configuration comprising: - an estimation (S50), by implementing an estimation method (20) according to any one of the preceding claims, of respective loads of cell beams of the wireless cellular access network, - use (S51) of the estimated beam loads to configure communications with user terminals.

9. A method (50) of configuration according to claim 8, wherein the estimated beam loads are used to perform at least one of: - management of communication resources of the wireless cellular access network, - mobility management of user terminals, - a load distribution between beams of the same cell and / or between beams of different cells.

10. A computer program product comprising a set of program code instructions which, when executed by at least one processor, configure said at least one processor to implement a method (20, 50) according to any one of the preceding claims.

11. Control device (30) comprising at least one memory and at least one processor configured to implement a method (20, 50) according to any one of claims 1 to 9.

12. A method (60) of configuring, by a user terminal (70), a communication with a wireless cellular access network comprising base stations serving a plurality of base station cells, a plurality of beams being formable in each cell, said method (60) of configuration comprising: - a reception (S60) of respective beam loads of one or more cells of the wireless cellular access network, said beam loads being estimated by implementing an estimation method (20) according to any one of claims 1 to 7, - a use (S61) of the estimated beam loads to configure communication with the wireless cellular access network.

13. A method (60) of configuration according to claim 12, wherein the estimated beam loads are used for mobility management of said user terminal.

14. A computer program product comprising a set of program code instructions which, when executed by at least one processor, configure said at least one processor to implement a method (60) according to any one of claims 12 to 13.

15. User terminal (70) for exchanging data with a wireless cellular access network, said user terminal comprising at least one memory and at least one at least one processor configured to implement a method (60) according to any one of claims 12 to 13.

Images