Optimization of latency for a mobile communications network

Abstract

A computer-implemented method for optimizing transmission latency in a deployed self-organizing cellular mobile communications network (100) is provided. The cellular mobile communications network comprises network cells (110) each covering a respective geographic area. The method comprises: setting modifiable cell parameters of the cells (110) to initial configuration values (CD); receiving (402) for each cell of a set of cells performance indicator parameters (PI) providing information regarding an average behavior of the communications network (100) in the cell and radio parameters (RP) providing information regarding a behavior of the mobile communications network at a territorial pixel being a portion of the geographic area wherein the cell is deployed; iterating at least once a routine comprising: a) selecting (455) updated configuration values (CD) for the cell parameters; b) for each cell of the set, estimating (410), through a network simulator, the effect of the updated configuration values (CD) on the received performance indicator parameters (PI) and on the received radio parameters (RP) by calculating corresponding estimated performance indicator parameters (PI') and estimated radio parameters (RP') corresponding to a simulated condition in which the modifiable cell parameters are at the updated configuration values (CD); c) for each cell of the set, correlating (415) the xxxxxx estimated performance indicator parameters (PI') and estimated radio parameters (RP') to calculate: first quality of service parameters (QPC) indicative of a quality of service of the communications network (100) at the cell; second quality of service parameters (QPP) indicative of a quality of service of the communications network (100) at territorial pixels of the cell; d) for each cell of the set, calculating (420) through a computational model a set of first estimated transmission latency values (Lg) based on said first quality of service parameters (QPC) and on second quality of service parameters (QPP), each first estimated transmission latency value (Lg) being indicative of transmission latency of the communications network at a corresponding territorial pixel of said cell; e) for each cell of the set, calculating (430) a second estimated transmission latency value (La) based on the first estimated transmission latency values (Lg) calculated for the cell, the second estimated transmission latency value (La) being indicative of an average transmission latency of the communications network at said cell. After each iteration of the routine, calculating (440), based on the first and second estimated transmission latency values, a cost value indicative of a cost, in terms of transmission latency, of a communication network configuration corresponding to the updated configuration values (CD). An iteration of the routine is selected based on its calculated cost value. The modifiable cell parameters of the cells are set to the updated configuration values (CD) of the selected iteration of the routine.

Description

[0001] OPTIMIZATION OF LATENCY FOR A MOBILE COMMUNICATIONS NETWORK

[0002] DESCRIPTION

[0003] Technical field

[0004] The present disclosure relates to the field of mobile communications networks. In particular, the present disclosure relates to the optimization of mobile communications networks, particularly but not limitatively 5G or future generations networks. A method, and a system for implementing the method, for optimizing mobile communications networks is disclosed.

[0005] Technical background

[0006] In the field of cellular mobile communications networks (in the following also referred to as “cellular networks” or “mobile networks” or “mobile communications networks”, for the sake of conciseness), like fourth generation (“4G”) and fifth generation (“5G”) networks, “Coverage and Capacity Optimization” (“CCO” in short) has the aim of providing the best configuration for network cells able of maximizing network coverage, network capacity and network performances to mobile network users on the field. This can be achieved with modifications to adjustable, tunable parameters of the network cells. Examples of cells’ tunable parameters are transmission power, electrical tilt and azimuth of the antennas, and any other available parameter affecting radiation diagram and power spatial distribution, including but not limiting parameters controlling radiation patterns for active antennas, for beamforming techniques (typical of 5G networks).

[0007] “Self Organizing Networks” (“SON” in short) is an automation technology paradigm able to perform the actions of planning and optimization (but also configuration and management) of cellular mobile communications networks, particularly 4G networks, 5G networks and networks of the next generations, using a more flexible approach for continuous improvement of performance even in closed- loop configuration.

[0008] The SON paradigm aims to make the planning and optimization phases of a cellular mobile communications network (from the fourth generation onwards) easier and faster. Thanks to SON algorithms such as CCO - operating on cells’ tunable parameters such as cell’s antenna(s) electrical tilt, cell’s transmission power and cell’s antenna(s) azimuth, other parameters affecting radiation diagram and power spatial distribution, including but not limiting parameters controlling radiation patterns for active antennas, for beamforming techniques, or parameters related to procedures of cell selection and handover, just to mention a few - it is possible to improve the performance, for example in terms of user throughput and / or carried traffic, also through continuous closed-loop optimization.

[0009] A very important aspect of the planning and optimization phases of a cellular mobile communications network is the minimization of latency.

[0010] The minimization of latency is particularly important for a wide range of modern real time applications (including for example video conferencing, voice over IP, online gaming, autonomous driving, etc.) for ensuring efficient interactivity and smooth experiences, increasing the energy efficiency of 5G networks, and improving reliability in mission-critical applications. Moreover, reduction of latency is critical to foster a range of new applications, such as virtual and augmented reality, smart cities, and connected cars.

[0011] Paper “Data-driven Predictive Latency for 5G: A Theoretical and Experimental Analysis Using Network Measurements’" by Marco Skocaj, Francesca Conserva, Nicol Sarcone Grande, Andrea Orsi, Davide Micheli, Giorgio Ghinamo, Simone Bizzarri and Roberto Verdone, 2023 IEEE 34th Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC), July 2023, provides a thorough analysis of predictive latency within 5G networks by utilizing real-world network data that is accessible to mobile network operators.

[0012] WO2022144207A1 relates to a method for optimizing network configuration of a cellular communication system featuring SON functionalities by using both KPI parameters and tracing / reporting data.

[0013] W02023036440A1 relates to techniques for handling Quality of Service (QoS) prediction parameters, in particular a network device for determining QoS prediction parameters in a mobile communication network.

[0014] Summary

[0015] Applicant has found that the abovementioned solutions known in the art do not allow to efficiently plan and optimize cellular mobile communications networks in order to reduce latency.

[0016] In general terms, the present invention is based on the idea of estimating latency affecting a mobile communications network by exploiting quality of service parameters indicative of a quality of service of the mobile communications network at network cells thereof and at territorial pixels of said network cells.

[0017] An aspect of the present invention relates to a computer-implemented method for optimizing transmission latency in a deployed self-organizing cellular mobile communications network comprising network cells each one covering a respective geographic area.

[0018] The method comprises setting modifiable cell parameters of the network cells to initial configuration values.

[0019] The method comprises receiving for each network cell of a set of said network cells performance indicator parameters providing information regarding an average behavior of the communications network in said network cell and radio parameters providing information regarding a behavior of the mobile communications network at territorial pixels of said network cell.

[0020] Each territorial pixel corresponds to a portion of the geographic area wherein aid network cell is deployed.

[0021] The method comprises iterating at least once a routine comprising the following sequence of operations a) - e): a) selecting updated configuration values for said cell parameters; b) for each network cell of the set, estimating, through a network simulator, the effect of said updated configuration values on the received performance indicator parameters and on the received radio parameters by calculating corresponding estimated performance indicator parameters and estimated radio parameters corresponding to a simulated condition in which the modifiable cell parameters are at the updated configuration values; c) for each network cell of the set, correlating said estimated performance indicator parameters and estimated radio parameters to calculate:

[0022] - first quality of service parameters indicative of a quality of service of the communications network at said network cell;

[0023] - second quality of service parameters indicative of a quality of service of the communications network at territorial pixels of said network cell; d) for each network cell of the set, calculating through a computational model a set of first estimated transmission latency values based on said first quality of service parameters and second quality of service parameters, each first estimated transmission latency value being indicative of transmission latency of the communications network at a corresponding territorial pixel of said network cell; e) for each network cell of the set, calculating a second estimated transmission latency value based on the first estimated transmission latency values calculated for the network cell, said second estimated transmission latency value being indicative of an average transmission latency of the communications network at said network cell;

[0024] - after each iteration of the routine, calculating, based on the first and second estimated transmission latency values, a cost value indicative of a cost, in terms of transmission latency, of a communication network configuration corresponding to the updated configuration values;

[0025] - selecting an iteration of the routine based on its calculated cost value;

[0026] - setting the modifiable cell parameters of the network cells to the updated configuration values of the selected iteration of the routine.

[0027] According to an embodiment of the present invention, said estimating the effect of said updated configuration values on the received performance indicator parameters and on the received radio parameters further comprises calculating said corresponding estimated performance indicator parameters and estimated radio parameters taking into account also the initial configuration values.

[0028] According to an embodiment of the present invention, said radio parameters are obtained using georeferenced measures collected by user equipment camped in the network cell when the modifiable cell parameters are at the initial configuration values.

[0029] According to an embodiment of the present invention, said performance indicator parameters are obtained using at least one of:

[0030] - Key Performance Indicators related to said network cell when the modifiable cell parameters are at the initial configuration values, and

[0031] - a combination of said georeferenced measures;

[0032] According to an embodiment of the present invention, operation d) is carried out through a first neural network trained using for each network cell of a group of network cells, corresponding averaged measures of said first quality of service parameters collected from said network cell.

[0033] According to an embodiment of the present invention, operation d) is carried out through a first neural network trained using for each network cell of a group of network cells, georeferenced measures of said second quality of service parameters collected from territorial pixels of the geographic area of network cells of the communications network.

[0034] According to an embodiment of the present invention, operation d) is carried out through a first neural network trained using for each network cell of a group of network cells, georeferenced measures of transmission latency collected from territorial pixels of the geographic area of network cells of the communications network.

[0035] According to an embodiment of the present invention, operation e) is carried out through a second neural network trained using georeferenced measures of transmission latency collected from territorial pixels of the geographic area of network cells of the communications network.

[0036] According to an embodiment of the present invention, operation e) is carried out through a second neural network trained using for each network cell of a group of network cells, corresponding averaged measures of transmission latency of the communications network at said network cell.

[0037] According to an embodiment of the present invention, operation c) comprises calculating said first quality of service parameters and said second quality of service parameters by processing weighted combinations of the estimated performance indicator parameters and estimated radio parameters.

[0038] According to an embodiment of the present invention, said calculating a cost value comprises quantifying an average transmission latency affecting one or more network cells of said communication network configuration corresponding to the updated configuration values.

[0039] According to an embodiment of the present invention, said first quality of service parameters corresponding to a network cell comprise at least one of

[0040] - average number of downlink active user equipment in the network cell; - Physical Resource Block downlink usage percentage in the network cell;

[0041] - average Channel Quality Indicator in the network cell;

[0042] - average uplink Received Signal Strength Indication in the network cell;

[0043] - average uplink Signal to Interference plus Noise Ratio in the network cell;

[0044] - average downlink Modulation and Coding Scheme in the network cell;

[0045] - average uplink Modulation and Coding Scheme used in the network cell;

[0046] - average data transmission volume in the network cell in terms of number of Protocol Data Units.

[0047] According to an embodiment of the present invention, said second quality of service parameters comprise, for each one of a group of territorial pixels, at least one of:

[0048] - downlink Signal to Interference plus Noise-Ratio in the territorial pixel;

[0049] - downlink Received Signal Strength Indication in the territorial pixel;

[0050] - downlink Modulation and Coding Scheme used in the territorial pixel;

[0051] - uplink Signal to Interference plus Noise-Ratio in the territorial pixel;

[0052] - uplink Received Signal Strength Indication in the territorial pixel;

[0053] - uplink Modulation and Coding Scheme used in the territorial pixel.

[0054] According to an embodiment of the present invention, said performance indicator parameters corresponding to a network cell comprise at least one of:

[0055] - number of active users in the network cell;

[0056] - traffic volume in the network cell;

[0057] - number of Hand Overs in the network cell;

[0058] - user throughput in the network cell.

[0059] According to an embodiment of the present invention, said radio parameters corresponding to a network cell comprise at least one of:

[0060] - Reference Signal Received Power measured in the network cell;

[0061] - Received Signal Received Quality in the network cell;

[0062] - Received signal code Power in the network cell;

[0063] - Timing Advance in the network cell; frequency bands employed in the network cell for data transmission / reception. According to an embodiment of the present invention, said modifiable cell parameters comprise for each network cell cell antenna parameters comprising at least one of:

[0064] - cell’s antenna transmitted power;

[0065] - cell’s antenna electric tilt;

[0066] - cell’s antenna azimuth;

[0067] - cell’s antenna gain;

[0068] - cell’s antenna radiation pattern.

[0069] According to an embodiment of the present invention, said modifiable cell parameters comprise for each network cell cell selection and handovers parameters comprising at least one of:

[0070] - layer priority of the network cell;

[0071] - minimum signal level for layer of the network cell;

[0072] - intra-layer cell offset parameters of the network cell.

[0073] Another aspect of the present invention relates to a SON system for a cellular mobile communications network comprising network cells each one covering a respective geographic area.

[0074] The SON system comprises a cell configuration selector configured to set and update configuration values of modifiable cell parameters of the network cells.

[0075] The SON system comprises a network simulator configured to receive, for each network cell of a set of said network cells having cell parameters configured with initial configuration values, performance indicator parameters providing information regarding an average behavior of the communications network in said network cell and radio parameters providing information regarding a behavior of the mobile communications network at territorial pixels of said network cell.

[0076] Each territorial pixel corresponds to a portion of the geographic area wherein said network cell is deployed.

[0077] The network simulator module is further configured to estimate the effect of updated configuration values on the received performance indicator parameters and on the received radio parameters by calculating corresponding estimated performance indicator parameters and estimated radio parameters corresponding to a simulated condition in which the modifiable cell parameters are at the updated configuration values.

[0078] The SON system comprises a parameter correlator configured to correlate, for each network cell of the set, said estimated performance indicator parameters and estimated radio parameters to calculate:

[0079] - first quality of service parameters indicative of a quality of service of the communications network at said network cell;

[0080] - second quality of service parameters indicative of a quality of service of the communications network at territorial pixels of said network cell.

[0081] The SON system comprises a latency estimator configured to calculate through a computational model, for each network cell of the set:

[0082] - a set of first estimated transmission latency values based on said first quality of service parameters and second quality of service parameters, each first estimated transmission latency value being indicative of transmission latency of the communications network at a corresponding territorial pixel of said network cell;

[0083] - a second estimated transmission latency value based on the first estimated transmission latency values calculated for the network cell, said second estimated transmission latency value being indicative of an average transmission latency of the communications network at said network cell.

[0084] The SON system comprises a network optimizer configured to calculate, based on the first and second estimated transmission latency values, a cost value indicative of a cost, in terms of transmission latency, of a communication network configuration corresponding to the updated configuration values used for calculating said first and second estimated transmission latency values.

[0085] The network optimizer is configured to control the cell configuration selector to set the modifiable cell parameters of the network cells to said updated configuration values conditioned to the calculated cost value.

[0086] Brief description of the drawings

[0087] Features and advantages of the solution here disclosed, including those mentioned in the foregoing, will appear more clearly by reading the following detailed description of exemplary and non-limitative embodiments. For a better intelligibility, the following description should be read making reference to the annexed drawings, wherein:

[0088] Figure 1 schematically illustrates a portion of a cellular mobile network according to an embodiment of the present invention;

[0089] Figure 2 illustrates in terms of functional blocks a SON system of the cellular mobile network according to an embodiment of the present invention;

[0090] Figure 3 illustrates in terms of logical elements a parameter correlator module of the SON system of Figure 2 according to an embodiment of the present invention;

[0091] Figure 4 is a flow chart of the main operations carried out by the SON system during a CCO procedure according to an embodiment of the present invention.

[0092] Detailed description of exemplary embodiments

[0093] With reference to the drawings, a (portion of a) cellular mobile network system 100 (in short, “mobile network”) according to an embodiment of the present invention is schematically illustrated in Figure 1.

[0094] According to an embodiment of the present invention, the mobile network 100 comprises a plurality of cellular communication equipment 105 (e.g., eNodeB or gNodeB) providing radio coverage over a geographic area.

[0095] According to an embodiment of the present invention, each cellular communication equipment 105 is configured to provide radio coverage over (or, equivalently, is associated with) one or more portions of the geographic area, or network cells, 110.

[0096] In the exemplary, simplified scenario herein considered, each cellular communication equipment 105 is associated with a respective network cell 110. In a practical scenario, each cellular communication equipment 105 may be associated with a plurality of network cells, such as three network cells.

[0097] According to an embodiment of the present invention, as exemplary illustrated, each network cell 110 is hexagonal in shape. In practice, though, a cell shape may differ significantly from an ideal hexagonal shape, e.g. due to geographical and / or propagation characteristics or constraints of the area where the cell is located. According to an embodiment of the present invention, each cellular communication equipment 105 comprises one or more electronic apparatuses (not shown). Example of electronic apparatuses include, but are not limited to, transceivers and digital signal processors.

[0098] According to an embodiment of the present invention, each cellular communication equipment 105 comprises one or more antennas.

[0099] According to an embodiment of the present invention, the cellular communication equipment 105 allow user equipment UE within the respective network cells 110 (and connecting / connected to the mobile network 100) to exchange data traffic (e.g., web browsing, e-mailing, voice, or multimedia data traffic).

[0100] The user equipment UE may for example comprise personal devices owned by users of the mobile network 100 (the users being for example subscribers of services offered by the mobile network 100). Examples of user equipment UE comprise, but are not limited to, mobile phones, smartphones, tablets, personal digital assistants and computers.

[0101] According to an embodiment of the present invention, the network cells 110 and their corresponding cellular communication equipment 105 are part of a radio access network of the mobile network 100.

[0102] The radio access network may be based on any suitable radio access technology. Examples of radio access technologies include, but are not limited to, UTRA ^‘UMTS Terrestrial Radio Access"), WCDMA (^‘Wideband Code Division Multiple Access"), CDMA2000, LTE Toiig Term Evolution"), LTE-A ^‘LTE- Advanced"), and NR ^‘New Radio").

[0103] According to an embodiment of the present invention, the radio access network is communicably coupled with one or more core networks, such as the core network 115. The core network 115 may be any type of network configured to provide aggregation, authentication, call control / switching, charging, service invocation, gateway and subscriber database functionalities, or at least a subset (i.e., one or more) thereof.

[0104] According to an embodiment of the present invention, the core network 115 comprises a 4G / LTE core network or a 5G core network. According to an embodiment of the present invention, the core network 115 is communicably coupled with other networks, such as the Internet and / or public switched telephone networks (not shown).

[0105] According to an embodiment of the present invention, the mobile network 100 is provided with SON functionalities configured to perform a CCO of the mobile network 100, i.e. functionalities that allow setting (z.e., tuning or adjusting) one or more operative parameters of the network cells 110 (hereinafter, cell parameters) for maximizing network performances to mobile network users on the field.

[0106] According to an embodiment of the present invention, the cell parameters of each network cell define a configuration of that network cell (hereinafter, cell configuration), and the cell configurations of the network cells of the mobile network define a configuration of the mobile network (hereinafter, network configuration).

[0107] According to an embodiment of the present invention, the cell parameters comprise, but are not limited to, one or more antenna parameters or cell selection / handover parameters. Examples of antenna parameters include, but are not limited to, transmitted power, antenna electrical tilt, azimuth, gain and antenna radiation pattern (e.g. pointing direction, directivity and width of one or more lobes of the pattern of lobes exhibited by the antenna radiation pattern). Example of cell selection and handover parameters include, but are not limited to, layer priority, minimum signal level for layer, intra-layer cell offset.

[0108] According to an embodiment of the present invention, the mobile network 100 comprises a SON system 120, i.e. a processing system that allows implementing SON functionalities to perform the CCO of the mobile network 100.

[0109] According to an embodiment of the present invention, the cell parameters (or, more generally, the cell configurations, and hence the network configurations) are set through proper commands, hereinafter referred to as SON commands, from the SON system 120 to the cellular network.

[0110] According to an embodiment of the present invention, the SON system 120 is located external to both the cellular network and the core network 115. According to alternative embodiments of the invention, the SON system 120 is located in the core network 115 (e.g., in one or more modules thereof) or in any other entity of the cellular network or of the mobile network 100. According to an embodiment of the present invention, the physical location of the SON system 120 depends on the implemented SON network architecture (e.g., distributed SON network, centralized SON network or hybrid SON network).

[0111] According to an embodiment of the present invention, the SON system 120 is configured to set (z.e., tune or adjust) the cell parameters of only a subset of the network cells 110 of the mobile network 100 (hereinafter, referred to as “tunable network cells”). According to another embodiment of the present invention, the SON system 120 is instead configured to set (z.e., tune or adjust) the cell parameters of all the network cells 110 of the mobile network 100 (in this case, all the network cells 110 of the mobile network 100 are “tunable network cells”).

[0112] According to an embodiment of the present invention, the SON system 120 is configured to perform a method (hereinafter, CCO method) for configuring the cellular network, and particularly for setting the cell parameters (or equivalently, the cell configurations, and hence the network configuration) for maximizing network performances, and particularly to reduce the latency affecting the mobile network.

[0113] As it is well known to those skilled in the art, according to the 3rd Generation Partnership Project (3 GPP), latency affecting a mobile network can be subdivided into Control-plane (C-plane) latency and User-plane (U-plane latency). By making reference to the mobile network 100 illustrated in Figure 1:

[0114] - C-plane latency is a measure of the time elapsed between a user equipment UE Random Access Channel (RACH) preamble transmission and the successful reception at a cellular communication equipment 105 of a Radio Resource Control (RRC) Connection Complete message;

[0115] - U-plane latency is a measure of the transit time between a packet being available at a user equipment UE (or at a cellular communication equipment 105) IP layer, and the availability of this packet at the IP layer of a cellular communication equipment 105 (or of a user equipment UE).

[0116] According to an embodiment of the present invention, the CCO method is based on measurements of one or more parameters relating to the operation of the cellular network (hereinafter, operating parameters). According to an embodiment, the measurements of the operating parameters may be performed by any suitable entity of the mobile network 100. According to an embodiment, the measurements of the operating parameters are collected by the SON system 120. According to an embodiment, the measurements of the operating parameters are collected by the SON system 120 based on proper signaling exchange with the cellular network (and / or with the user equipment UE connected thereto, as better discussed in the following) and / or with the core network 115.

[0117] According to an embodiment of the present invention, the operating parameters comprise one or more performance indicator parameters PI comprising the so-called Key Performance Indicators.

[0118] According to an embodiment of the present invention, performance indicator parameters PI comprise, for each network cell 110 of at least a subset of the network cells 110 of the mobile network 100, at least one of:

[0119] - Number of active users in the network cell 110, for example in terms of average number of active user equipment UE per Transmission time Interval (TTI).

[0120] - Traffic volume in the network cell 110, for example in terms of average traffic load managed by the cellular communication equipment 105 of the network cell 110 in a predefined time interval.

[0121] - Number of handovers in the network cell 110, for example in terms of average handovers carried out by the user equipment UE in the network cell 100 in a predefined time interval.

[0122] - User throughput in the network cell 110, for example in terms of average number of correctly received bits by individual user equipment UE delivered over a certain period of time.

[0123] According to an embodiment of the present invention, the measurements of the performance indicator parameters PI are performed by proper performance counters of the cellular network. According to an embodiment of the present invention, the performance counters are implemented in the cellular communication equipment 105.

[0124] According to an embodiment of the present invention, the operating parameters further comprise radio parameters RP measured and reported by the user equipment UE connected to the cellular network. According to an embodiment, measurement for obtaining the radio parameters RP is performed by the user equipment UE through the Minimization of Drive Test (MDT) functionality.

[0125] According to an embodiment of the present invention, the radio parameters RP comprise at least one of

[0126] - Received Signal Received Power (RSRP);

[0127] - Received Signal Received Quality (RSRQ);

[0128] - Received Signal Code Power (RSCP);

[0129] - timing advance;

[0130] - frequency layers (or frequency bands, such as 800 MHz, 1800 MHz, 2600 MHz) employed for data transmission / reception.

[0131] By using the MDT functionality, the radio parameters RP measured and reported by the user equipment UE are advantageously combined with positioning information. Positioning information may for example be provided by the user equipment UE (e.g., by exploiting GPS and / or GNSS / A-GNSS functionalities thereof) and / or computed by the mobile network 100 e.g., by the core network 115) based on the radio parameters. Examples of positioning information computed by the mobile network 100 include, but are not limited to, ranging measurements based on localization signals emitted by any properly configured cellular communication equipment, and / or triangulations on signals of the cellular network.

[0132] An important difference between the performance indicator parameters PI and the radio parameters RP is that:

[0133] - The performance indicator parameters PI provide information about the mobile network 100 at a network cell level, i.e., information regarding the average behavior of the mobile network 100 in the various network cells 110 (obtained for each network cell 100 by collecting averaged information based on the operation of a plurality of different user equipment UE distributed within the network cell 110).

[0134] - The radio parameters RP provide information about the mobile network 100 at a geographic position level, i.e., georeferenced information regarding the behavior of user equipment UE individually collected by the latter at specific territorial locations, and aggregated per territorial portions (also known as “territorial pixels”), i.e., elementary portions of the geographical area wherein the network cells 110 are deployed.

[0135] Figure 2 illustrates in terms of functional blocks the SON system 120 according to an embodiment of the present invention.

[0136] It should be noted that the terms ‘system’, ‘module’, ‘unit’ are herein intended to comprise, but not limited to, hardware, firmware, a combination of hardware and software, or software.

[0137] For example, a system, a module or a unit may be, but is not limited to being, a process running on a processor, an object, an executable, a thread of execution, a program, and / or a computing device.

[0138] In other words, a system, a module or a unit may comprise an application being executed on a computing device and / or the computing device itself.

[0139] One or more systems, modules or units may be localized on one computing device and / or distributed between two or more computing devices.

[0140] Systems, modules or units may comprise and / or interact with computer readable media having storing data according to various data structures.

[0141] The systems, modules or units may communicate by exploiting local and / or remote processes, preferably by means of electrical, electromagnetic and / or optical signals providing one or more data packets, such as data packets from one node, system or module interacting with another node, system or module in a local system, in a distributed system, and / or across a radio network and / or a wired network.

[0142] According to an embodiment of the present invention, the SON system 120 comprises a network simulator module 206.

[0143] According to an embodiment of the present invention, the network simulator module 206 features electromagnetic simulation functionalities aimed at providing, over the geographic area, a priori estimate of the effects that network configuration changes i.e., changes in the cell configuration of one or more network cells 110 of the cellular network) have on the area coverages of the network cells 110 within the geographic area.

[0144] For this purpose, according to an embodiment of the present invention, the network simulator module 206 is configured to receive configuration data CD comprising data indicative of cell configurations for a set of tunable network cells 110. According to an embodiment of the present invention, the cell configuration for a tunable network cell 110 comprises setting values for a set of operative cell parameters of the tunable network cell 110 comprising at least one of:

[0145] - transmission power of the cellular communication equipment 105; gain of the antenna(s) of the cellular communication equipment 105; electrical tilt of the antenna(s) of the cellular communication equipment 105; azimuth of the antenna(s) of the cellular communication equipment 105; radiation pattern (e.g. pointing direction, directivity and width of one or more lobes of the pattern) of the antenna(s) of the cellular communication equipment 105.

[0146] - Minimum signal level for each layer and layer priority for cell selection and inter-frequency handover procedures

[0147] Cell Offset for intra-frequency handover procedures

[0148] The radiation pattern cell parameter is available only if the cellular communication equipment 105 is equipped with active antenna(s).

[0149] According to an embodiment of the present invention, electromagnetic simulation provided by the network simulator module 206 may be based on morphological information of the geographic area. An example of such electromagnetic simulation is disclosed in Hata, M. “Empirical Formula for Propagation Loss in Land Mobile Radio Services”, IEEE Transactions on Vehicular Technology, August 1980, VT-29 (3): 317-325.

[0150] According to an embodiment of the present invention, the network simulator module 206 is configured to receive from the cellular network the performance indicator parameters PI and the radio parameters RP corresponding to measures collected from a group of network cells 110 of the mobile network 100 in a specific network configuration (z.e., with the network cells 110 that are configured according to specific, on field, cell configurations).

[0151] According to an embodiment of the present invention, the network simulator module 206 is configured to estimate the effects that a new network configuration - identified by (new) configuration data CD - have on the received performance indicator parameters PI and radio parameters RP by calculating (estimating) updated versions - corresponding to the configuration data CD - of said received performance indicator parameters PI and radio parameters RP. These updated versions of the performance indicator parameters PI and radio parameters RP corresponding to the configuration data CD are identified as estimated performance indicator parameters PI’ and estimated radio parameters RP’, respectively.

[0152] According to an embodiment of the present invention, the network simulator module 206 is configured to estimate the effects that a new network configuration identified by new configuration data CD have on the received performance indicator parameters PI and radio parameters RP by taking into account also previous configuration data CD (for example, configuration data CD corresponding to a (previous) on field cell configuration), so as to calculate the estimated performance indicator parameters PI’ and estimated radio parameters RP’ in form of variations with respect to the performance indicator parameters PI and radio parameters RP corresponding to the previous configuration data CD. For this purpose, according to an embodiment of the present invention, the network simulator module 206 may be configured to memorize previous configuration data CD (e.g., the ones corresponding to a (previous) on field cell configuration) to be exploited for the calculation of the estimates corresponding to the new configuration data CD.

[0153] According to an embodiment of the present invention, the SON system 120 comprises a parameter correlator module 220 configured to correlate the estimated performance indicator parameters PI’ and the estimated radio parameters RP’ for calculating corresponding Quality of Service (QoS) parameters indicative of a quality of service of the mobile network 100 corresponding to the estimated performance indicator parameters PI’ and to the estimated radio parameters RP’. In other words, the parameter correlator module 220 is configured to estimate the quality of service that would offer the mobile network 100 if the tunable network cells 110 thereof were configured with the configuration data CD used by the radio simulator module 220 to calculate the estimated performance indicator parameters PI’ and to the estimated radio parameters RP’.

[0154] According to an embodiment of the present invention, the QoS parameters calculated by the radio simulator module 220 comprise QoS parameters QPC that are aggregated on a per-network cell basis. According to an embodiment of the present invention, said QoS parameters QPC comprise for each one of a group of network cells 110 of the cellular network at least one of:

[0155] - average number of downlink active user equipment UE in the network cell 110;

[0156] - Physical Resource Block (PRB) downlink usage percentage in the network cell 110;

[0157] - average Channel Quality Indicator (CQI) in the network cell 110;

[0158] - average uplink Received Signal Strength Indicator (RSSI) in the network cell 110;

[0159] - average uplink Signal to Interference plus Noise Ratio (SINR) in the network cell 110;

[0160] - average downlink Modulation and Coding Scheme (MCS) used in the network cell 110;

[0161] - average uplink MCS used in the network cell 110;

[0162] - average data transmission volume in the network cell 110 in terms of number of Protocol Data Units (PDU).

[0163] It is known to those skilled in the art that the abovementioned QoS parameters QPC have been demonstrated to influence transmission latency in a mobile network.

[0164] According to an embodiment of the present invention, the QoS parameters calculated by the radio simulator module 220 further comprise QoS parameters QPP on a per-territorial pixel basis. According to an embodiment of the present invention, said QoS parameters QPP comprise for each one of a group of territorial pixels of the geographic area wherein the network cells 110 are deployed at least one of:

[0165] - downlink SINR in the territorial pixel;

[0166] - downlink RSSI in the territorial pixel;

[0167] - downlink MCS used in the territorial pixel;

[0168] - uplink SINR in the territorial pixel;

[0169] - uplink RSSI in the territorial pixel; - uplink MCS used in the territorial pixel.

[0170] Figure 3 illustrates in terms of logical elements the parameter correlator module 220 according to an embodiment of the present invention.

[0171] According to an embodiment of the present invention, the parameter correlator module 220 comprises a correlator unit 300 implementing a correlation model (for example, a mathematical and / or a neural network model) configured to receive the estimated performance indicator parameters PI’ and the estimated radio parameters RP’, and to obtain the QoS parameters QPC and QPP by processing weighted combinations of the estimated performance indicator parameters PI’ and the estimated radio parameters RP’. For example, the correlator unit 300 may be based on a correlation module corresponding to the one described in the already mentioned paper “Data-driven Predictive Latency for 5G: A Theoretical and Experimental Analysis Using Network Measurements’", suitably modified to take into account as input also data regarding per-territorial pixel basis information, i.e., the radio parameters RP’, and to produce as output also parameters on a per-territorial pixel basis, i.e., the QoS parameters QPP.

[0172] According to an embodiment of the present invention, the QoS parameters QPC may be advantageous calculated by the correlator unit 300 by correlations among estimated performance indicator parameters PI’ and / or correlations among aggregations of estimated radio parameters RP’, as schematically shown in Figure 3.

[0173] According to an embodiment of the present invention, the SON system 120 further comprises a latency estimator module 230 configured to process the QoS parameters QPC and QPP outputted by the parameter correlator module 220 for calculating latency estimation parameters indicative of latency of the mobile network 100 corresponding to the QoS parameters QPC and QPP In other words, the latency estimator module 230 is configured to estimate the latency that would affect the mobile network 100 if the tunable network cells 110 thereof were configured with the configuration data CD.

[0174] According to an embodiment of the present invention, the latency estimator module 230 comprises a first latency processing module 240(1) configured to calculate georeferenced latency estimates Lg each one indicative of an estimate of the latency experienced by the mobile network 100 in transmissions carried out at a corresponding territorial pixel if the tunable network cells 110 thereof were configured with the configuration data CD.

[0175] According to an embodiment of the present invention, the first latency processing module 240(1) is configured to calculate the georeferenced latency estimates Lg by processing the QoS parameters QPC and QPP generated by the parameter correlator module 220 through a first computational model, such as a neural network or an empirical model obtained from the observation of the operation of mobile networks.

[0176] According to an embodiment of the present invention, the first computational model comprises a first neural network NN1.

[0177] According to an embodiment of the present invention the first neural network NN1 is a neural network that has been trained using a significantly large set of samples comprising:

[0178] - measures of QoS parameters aggregated on a per-network cell basis collected from network cells 110 of the mobile network;

[0179] - measures of QoS parameters on a per-territorial pixel basis collected from territorial pixels of the geographic area wherein the network cells 110 of the mobile network 100 are deployed;

[0180] - measures of georeferenced latency on a per-territorial pixel basis collected from territorial pixels of the geographic area wherein the network cells 110 of the mobile network 100 are deployed.

[0181] According to an embodiment of the present invention, the measured QoS parameters aggregated on a per-network cell basis used for training the first neural network NN1 correspond to the QoS parameters QPC already described above, i.e., comprises at least one of: average number of downlink active user equipment UE, PRB downlink usage, average CQI, average uplink RS SI, average uplink RS SI, average uplink SINR, average downlink MCS, average uplink MCS, average volume in PDU.

[0182] According to an embodiment of the present invention, the QoS parameters aggregated on a per-network cell basis used for training the first neural network NN1 are obtained from measures of performance indicator parameters performed by performance counters of the cellular network, such as for example performance counters implemented in the cellular communication equipment 105 of the cellular network.

[0183] According to an embodiment of the present invention, the measured QoS parameters on a per-territorial pixel basis used for training the first neural network NN1 correspond to the QoS parameters QPP already described above, i.e., comprises at least one of: downlink SINR per territorial pixel, downlink RS SI per territorial pixel, downlink MCS per territorial pixel.

[0184] According to an embodiment of the present invention, the measured QoS parameters on a per-territorial pixel basis used for training the first neural network NN1 are obtained from MDT campaign measures.

[0185] According to an embodiment of the present invention, the measures of georeferenced latency used for training the first neural network NN1 are obtained from MDT campaign measures and / or individually carried out by user equipment UE or by proper agents installed thereon.

[0186] According to an embodiment of the present invention, the large set of samples used to train the first neural network NN1 comprises measures advantageously collected with the mobile network 100 that is in a large number of different conditions influencing the correlations among parameters used to calculate the latency.

[0187] Particularly, according to an embodiment of the invention, the samples are collected by taking into account variations in at least one of:

[0188] - network configuration (in terms of configuration of the network cells 110);

[0189] - traffic typology (e.g., data traffic having different QoS or voice traffic);

[0190] - use of Network slicing;

[0191] - propagation environment (e.g., urban, rural, . . .);

[0192] - radio environment conditions (e.g., signal to interference ratio, useful signal level, distance between user equipment UE and cellular communication equipment 105, modulation and code schemes, Block Error Rate (BER), data call setup time, voice call setup time);

[0193] - network cells load (e.g., in terms of traffic and radio resource usage);

[0194] - reference time period (time, day, month, year). According to an embodiment of the present invention, the latency estimator module 230 comprises a second latency processing module 240(2) configured to calculate, for each network cell 110 of one or more of the network cells 110 of the mobile network 100, a corresponding average latency estimate La indicative of an estimate of the average latency experienced by the mobile network 100 in transmissions carried out at said network cell 110 if the tunable network cells 110 of the mobile network 100 were configured with the configuration data CD.

[0195] According to an embodiment of the present invention, the second latency processing module 240(2) is configured to calculate the average latency estimate La of a network cell 110 by processing through a second computational model, such as a neural network or an empirical model obtained from the observation of the operation of mobile networks, the georeferenced latency estimates Lg (generated by the first latency processing module 240(1)) for that network cell 110.

[0196] According to an embodiment of the present invention, the second computational model comprises a second neural network NN2.

[0197] According to an embodiment of the present invention the second neural network NN2 is a neural network that has been trained using a significantly large set of samples comprising, for each one of at least one network cell 110:

[0198] - measures of georeferenced latencies aggregated on a per-territorial pixel cell collected from territorial pixels of the geographic area wherein the network cell 110 is deployed;

[0199] - measures of average latency on a per-cell basis collected from the network cell 110.

[0200] According to an embodiment of the present invention, the measures of georeferenced latency used for training the second neural network NN2 are obtained from MDT campaign measures and / or individually carried out by user equipment UE or by proper agents installed thereon.

[0201] According to an embodiment of the present invention, the average latencies used for training the second neural network NN2 are obtained from measures of performance indicator parameters performed by performance counters of the cellular network, such as for example performance counters implemented in the cellular communication equipment 105 of the cellular network. Similarly to the case of the first neural network NN1, according to an embodiment of the present invention, the large set of samples used to train the second neural network NN2 comprises measures advantageously collected with the mobile network 100 that is in a large number of different conditions influencing the correlations among parameters used to calculate the latency.

[0202] Particularly, according to an embodiment of the invention, in this case as well the samples are collected by taking into account variations in at least one of:

[0203] - network configuration (in terms of configuration of the network cells 110);

[0204] - traffic typology (e.g., data traffic having different QoS or voice traffic);

[0205] - use of Network slicing;

[0206] - propagation environment (e.g., urban, rural, . . .);

[0207] - radio environment conditions (e.g., signal to interference ratio, useful signal level, distance between user equipment UE and cellular communication equipment 105, modulation and code schemes, Block Error Rate (BER), data call setup time, voice call setup time);

[0208] - network cells load (e.g., in terms of traffic and radio resource usage);

[0209] - reference time period (time, day, month, year).

[0210] According to an embodiment of the present invention, the SON system 120 further comprises a network optimization module 260 configured to evaluate the performance of the mobile network 100 from the latency point of view as a function of the configuration data CD.

[0211] For this reason, according to an embodiment of the present invention, the network optimization module 260 is configured to determine a value of a respective cost function CF indicative of a cost, in terms of network latency, such as a quantification of an average latency affecting one or more network cells 110, of the network configuration corresponding to the configuration data CD.

[0212] According to an embodiment of the present invention, the cost function CF depends on an average of the average latencies experienced by the mobile network 100 in transmissions carried out in a set of network cells 110. For this purpose, according to an embodiment of the present invention, the network optimization module 260 is configured to calculate a value of the cost function CF by using the average latency estimates La and / or the georeferenced latency estimates Lg generated by the latency estimator module 230.

[0213] According to an embodiment of the present invention, the cost function CF may also depend on at least one of: handled traffic in normal and / or in over-target conditions in the set of network cells 110; an average of the cell average throughput over the set of network cells 110; an average of the user average throughput over the set of network cells 110.

[0214] According to an embodiment of the invention the network optimization module 260 may be further configured to calculate a value of the cost function CF by using - in addition to the average latency estimates La and / or the georeferenced latency estimates Lg - also one or more of the QoS parameters QPC and QPP generated by the parameter correlator module 220.

[0215] According to an embodiment of the present invention, the network optimization module 260 is configured to span (at least a portion of) the space (tree) defined by all the possible network configurations determined by the possible values of the configuration data CD of the tunable network cells 110 in order to minimize the value of the cost function CF. For example, said space of the network configurations may be spanned using one of the already known exploration algorithms employed in the field of mobile communication network optimization, such as the one descripted in W02022 / 207402 filed by the same applicant.

[0216] For this purpose, according to an embodiment of the present invention, the network optimization module 260 controls a cell configuration selector module 270 to progressively change the cell configuration data CD to be fed to the network simulator module 206 (causing a corresponding variation in the average latency estimates La and / or georeferenced latency estimates Lg) until a minimum in the cost function CF is identified.

[0217] According to an embodiment of the present invention, the network optimization module 260 is further configured to control the cell configuration selector module 270 to provide the cell configuration data CD corresponding to the identified cost function CF minimum to a cell configuration enforcer module 280.

[0218] According to an embodiment of the present invention, the cell configuration enforcer module 280 interacts with the network cells 110 of the mobile network in order to implement the network configuration indicated by the cell configuration data CD corresponding to the identified cost function CF minimum.

[0219] For this purpose, according to an embodiment of the present invention, the cell configuration enforcer module 280 sends corresponding SON commands to selected network cells 110 among the tunable network cells 110 of the cellular network for adjusting values of cell parameters (transmission power, antenna gain, tilt, azimuth, radiation pattern, ...) of said tunable network cells 110 according to said cell configuration data CD.

[0220] Figure 4 is a flow chart of the main operations carried out by modules of the SON system 120 during a CCO procedure according to an embodiment of the present invention.

[0221] At the beginning of the CCO procedure according to an embodiment of the present invention, the network simulator module 206 receives from the cellular network performance indicator parameters PI and radio parameters RP corresponding to measures collected from a group of network cells 110 of the mobile network 100 (block 402). The measures are collected from network cells 110 that are configured according to a currently on field network configuration, identified by an initial configuration data CD. According to an embodiment of the present invention, the network simulator module 206 also collects and memorizes said initial configuration data CD.

[0222] Then, according to an embodiment of the present invention, the network simulator module 206 receives from the cell configuration selector module 270 new configuration data CD identifying a new network configuration and accordingly estimates updated versions of the received performance indicator parameters PI and radio parameters RP by generating corresponding estimated performance indicator parameters PI’ and estimated radio parameters RP’ (block 410). According to an embodiment of the present invention, the network simulator module 206 may generate the estimated performance indicator parameters PI’ and estimated radio parameters RP’ by taking into account also the memorized initial configuration data CD corresponding to the currently on field network configuration.

[0223] According to an embodiment of the present invention, the parameter correlator module 220 correlates the estimated performance indicator parameters PI’ and the estimated radio parameters RP’ generated by the network simulator module 206 to obtain QoS parameters QPC aggregated on a per-network cell basis and QoS parameters QPP on a per-territorial pixel basis (block 415).

[0224] At this point, according to an embodiment of the present invention, the first latency processing module 240(1) calculates georeferenced latency estimates Lg by processing the QoS parameters QPC and QPP through the first neural network NN1 (block 420).

[0225] Then, according to an embodiment of the present invention, the second latency processing module 240(2) calculates average latency estimates La by processing the georeferenced latency estimates Lg through the second neural network NN2 (block 430)

[0226] According to an embodiment of the present invention, the network optimization module 260 processes the average latency estimates La and / or the georeferenced latency estimates Lg (and optionally, also one or more of the QoS parameters QPC and QPP) to determine a value of the cost function CF indicative of a cost, in terms of network latency of the network configuration corresponding to the configuration data CD (block 440).

[0227] At this point, according to an embodiment of the present invention, the network optimization module 260 verifies if this determined value of the cost function CF corresponds to a minimum of the cost function CF (block 450).

[0228] According to an embodiment of the present invention, if the network optimization module 260 assesses that the determined value of the cost function CF does not correspond to a minimum (exit branch N of block 450), the network optimization module 260 assesses that the network configuration corresponding the analyzed cell configuration data CD is not an optimum configuration from the latency point of view. Therefore, according to an embodiment of the present invention, the network optimization module 260 controls the cell configuration selector module 270 to change the cell configuration data CD (block 455), so that new estimated performance indicator parameters PI’ and estimated radio parameters RP’ corresponding to said new cell configuration data CD are generated for a new evaluation of the cost function CF corresponding to said new cell configuration data CD (return to block 410, iterate / reiterate operations of blocks 415 - 450). According to an embodiment of the present invention, if the network optimization module 260 assesses that the determined value of the cost function CF corresponds to a minimum (exit branch Y of block 450), the network optimization module 260 assesses that the network configuration corresponding the analyzed cell configuration data CD is an optimum configuration from the latency point of view. Therefore, according to an embodiment of the present invention, the network optimization module 260 controls the cell configuration selector module 270 to provide the cell configuration data CD corresponding to the identified cost function CF minimum to the cell configuration enforcer module 280, which interacts through SON commands with the network cells 110 of the mobile network in order to implement the network configuration indicated by the cell configuration data CD corresponding to the identified cost function CF minimum (block 460).

[0229] Naturally, in order to satisfy local and specific requirements, a person skilled in the art may apply to the invention described above many logical and / or physical modifications and alterations. More specifically, although the present invention has been described with a certain degree of particularity with reference to preferred embodiments thereof, it should be understood that various omissions, substitutions and changes in the form and details as well as other embodiments are possible. In particular, different embodiments of the invention may even be practiced without the specific details set forth in the preceding description for providing a more thorough understanding thereof; on the contrary, well-known features may have been omitted or simplified in order not to encumber the description with unnecessary details. Moreover, it is expressly intended that specific elements and / or method steps described in connection with any disclosed embodiment of the invention may be incorporated in any other embodiment.

Claims

CLAIMS1. A computer-implemented method for optimizing transmission latency in a deployed self-organizing cellular mobile communications network (100) comprising network cells (110) each one covering a respective geographic area, the method comprising:- setting modifiable cell parameters of the network cells (110) to initial configuration values (CD);- receiving (402) for each network cell of a set of said network cells performance indicator parameters (PI) providing information regarding an average behavior of the communications network (100) in said network cell and radio parameters (RP) providing information regarding a behavior of the mobile communications network at territorial pixels of said network cell, each territorial pixel corresponding to a portion of the geographic area wherein said network cell is deployed;- iterating at least once a routine comprising the following sequence of operations a) - e): a) selecting (455) updated configuration values (CD) for said cell parameters; b) for each network cell of the set, estimating (410), through a network simulator, the effect of said updated configuration values (CD) on the received performance indicator parameters (PI) and on the received radio parameters (RP) by calculating corresponding estimated performance indicator parameters (PI’) and estimated radio parameters (RP’) corresponding to a simulated condition in which the modifiable cell parameters are at the updated configuration values (CD); c) for each network cell of the set, correlating (415) said estimated performance indicator parameters (PI’) and estimated radio parameters (RP’) to calculate:- first quality of service parameters (QPC) indicative of a quality of service of the communications network (100) at said network cell;- second quality of service parameters (QPP) indicative of a quality of service of the communications network (100) at territorial pixels of said network cell; d) for each network cell of the set, calculating (420) through a computational model a set of first estimated transmission latency values (Lg) based on said firstquality of service parameters (QPC) and on second quality of service parameters (QPP), each first estimated transmission latency value (Lg) being indicative of transmission latency of the communications network at a corresponding territorial pixel of said network cell; e) for each network cell of the set, calculating (430) a second estimated transmission latency value (La) based on the first estimated transmission latency values (Lg) calculated for the network cell, said second estimated transmission latency value (La) being indicative of an average transmission latency of the communications network at said network cell;- after each iteration of the routine, calculating (440), based on the first and second estimated transmission latency values, a cost value indicative of a cost, in terms of transmission latency, of a communication network configuration corresponding to the updated configuration values (CD);- selecting an iteration of the routine based on its calculated cost value;- setting the modifiable cell parameters of the network cells to the updated configuration values (CD) of the selected iteration of the routine.

2. The method of claim 1, wherein said estimating (410) the effect of said updated configuration values (CD) on the received performance indicator parameters (PI) and on the received radio parameters (RP) further comprises calculating said corresponding estimated performance indicator parameters (PI’) and estimated radio parameters (RP’) taking into account also the initial configuration values (CD).

3. The method of claim 1 or 2, wherein:- said radio parameters (RP) are obtained using georeferenced measures collected by user equipment (UE) camped in the network cell (110) when the modifiable cell parameters are at the initial configuration values (CD);- said performance indicator parameters (PI) are obtained using at least one of:- Key Performance Indicators related to said network cell (110) when the modifiable cell parameters are at the initial configuration values (CD), and- a combination of said georeferenced measures.

4. The method of any of the preceding claims, wherein operation d) (420) is carried out through a first neural network (240(1)) trained using:- for each network cell (110) of a group of network cells, corresponding averaged measures of said first quality of service parameters (QPC) collected from said network cell (110);- georeferenced measures of said second quality of service parameters (QPP) collected from territorial pixels of the geographic area of network cells (110) of the communications network (100);- georeferenced measures of transmission latency collected from territorial pixels of the geographic area of network cells (110) of the communications network (100).

5. The method of any of the preceding claims, wherein operation e) (430) is carried out through a second neural network (240(2)) trained using:- georeferenced measures of transmission latency collected from territorial pixels of the geographic area of network cells (110) of the communications network (100);- for each network cell (110) of a group of network cells, corresponding averaged measures of transmission latency of the communications network (100) at said network cell.

6. The method of any of the preceding claims, wherein operation c) (415) comprises calculating said first quality of service parameters (QPC) and said second quality of service parameters (QPP) by processing weighted combinations of the estimated performance indicator parameters (PI’) and estimated radio parameters (RP )7. The method of any of the preceding claims, wherein said calculating (440) a cost value comprises quantifying an average transmission latency affecting one or more network cells (110) of said communication network configuration correspondingto the updated configuration values (CD).

8. The method of any of the preceding claims, wherein said first quality of service parameters (QPC) corresponding to a network cell (110) comprise at least one of- average number of downlink active user equipment (UE) in the network cell;- Physical Resource Block downlink usage percentage in the network cell;- average Channel Quality Indicator in the network cell;- average uplink Received Signal Strength Indication in the network cell;- average uplink Signal to Interference plus Noise Ratio in the network cell;- average downlink Modulation and Coding Scheme in the network cell;- average uplink Modulation and Coding Scheme used in the network cell;- average data transmission volume in the network cell in terms of number of Protocol Data Units.

9. The method of any of the preceding claims, wherein said second quality of service parameters (QPP) comprise, for each one of a group of territorial pixels, at least one of- downlink Signal to Interference plus Noise-Ratio in the territorial pixel;- downlink Received Signal Strength Indication in the territorial pixel;- downlink Modulation and Coding Scheme used in the territorial pixel;- uplink Signal to Interference plus Noise-Ratio in the territorial pixel;- uplink Received Signal Strength Indication in the territorial pixel;- uplink Modulation and Coding Scheme used in the territorial pixel.

10. The method of any of the preceding claims, wherein said performance indicator parameters (PI) corresponding to a network cell (110) comprise at least one of- number of active users in the network cell;- traffic volume in the network cell;- number of Hand Overs in the network cell;- user throughput in the network cell.

11. The method of any of the preceding claims, wherein said radio parameters (RP) corresponding to a network cell (110) comprise at least one of:- Reference Signal Received Power measured in the network cell;- Received Signal Received Quality in the network cell;- Received signal code Power in the network cell;- Timing Advance in the network cell; frequency bands employed in the network cell for data transmission / reception.

12. The method of any of the preceding claims, wherein said modifiable cell parameters comprise for each network cell (110) cell antenna parameters comprising at least one of:- cell’s antenna transmitted power;- cell’s antenna electric tilt;- cell’s antenna azimuth;- cell’s antenna gain;- cell’s antenna radiation pattern.

13. The method of any of the preceding claims, wherein said modifiable cell parameters comprise for each network cell (110) cell selection and handovers parameters comprising at least one of:- layer priority of the network cell;- minimum signal level for layer of the network cell;- intra-layer cell offset parameters of the network cell.

14. A SON system (200) for a cellular mobile communications network (100) comprising network cells (110) each one covering a respective geographic area, the SON system comprising:- a cell configuration selector (270) configured to set and update configuration values (CD) of modifiable cell parameters of the network cells (110);- a network simulator (206) configured to receive, for each network cell of a set of said network cells having cell parameters configured with initial configuration values (CD), performance indicator parameters (PI) providing information regarding an average behavior of the communications network (100) in said network cell and radio parameters (RP) providing information regarding a behavior of the mobile communications network at territorial pixels of said network cell, each territorial pixel corresponding to a portion of the geographic area wherein said network cell is deployed, the network simulator module (206) being further configured to estimate the effect of updated configuration values (CD) on the received performance indicator parameters (PI) and on the received radio parameters (RP) by calculating corresponding estimated performance indicator parameters (PI’) and estimated radio parameters (RP’) corresponding to a simulated condition in which the modifiable cell parameters are at the updated configuration values (CD);- a parameter correlator (220) configured to correlate, for each network cell of the set, said estimated performance indicator parameters (PI’) and estimated radio parameters (RP’) to calculate:- first quality of service parameters (QPC) indicative of a quality of service of the communications network (100) at said network cell;- second quality of service parameters (QPP) indicative of a quality of service of the communications network (100) at territorial pixels of said network cell;- a latency estimator (230) configured to calculate through a computational model, for each network cell of the set:- a set of first estimated transmission latency values (Lg) based on said first quality of service parameters (QPC) and second quality of service parameters (QPP), each first estimated transmission latency value (Lg) being indicative of transmission latency of the communications network at a corresponding territorial pixel of said network cell;- a second estimated transmission latency value (La) based on the first estimated transmission latency values (Lg) calculated for the network cell, said second estimated transmission latency value (La) being indicative of an average transmission latency of the communications

Images