WO2017108106A1

WO2017108106A1 - Method and network node for identifiying specific area of wireless communication system

Info

Publication number: WO2017108106A1
Application number: PCT/EP2015/081024
Authority: WO
Inventors: M Josefa VERA NADALES; Alejandro AGUILAR; Raquel BARCO MORENO; Josko ZEC
Original assignee: Telefonaktiebolaget LM Ericsson AB
Current assignee: Telefonaktiebolaget LM Ericsson AB
Priority date: 2015-12-22
Filing date: 2015-12-22
Publication date: 2017-06-29
Anticipated expiration: 2018-06-22

Abstract

A method and a network node (110) are disclosed. The network node (110) identifies (220) at least a specific area among a set of areas of a wireless communication system (100). The identifying (220) is based on relevance values for the set of areas and on a set of weighted Key Performance Indicators "KPIs" indicating performance of the wireless communication system (100), wherein the set of weighted KPIs comprises a respective set of weighted KPIs for each area of the set of areas. Each weighted KPI is associated with a respective weight value. Moreover, a corresponding computer program and a carrier therefor are disclosed.

Description

METHOD AND NETWORK NODE FOR IDENTIFYING SPECIFIC AREA OF

WIRELESS COMMUNICATION SYSTEM

TECHNICAL FIELD

Embodiments herein relate to wireless communication systems, such as cellular networks. In particular, a method and a network node for identification of a specific area among a set of areas of a wireless communication system are disclosed. A

corresponding computer program and a carrier therefor are also disclosed. BACKGROUND

For a wireless communication system, high performance in general terms is of interest to an operator of the wireless communication system. The performance may be measures by various Key Performance Indicators (KPIs), such as Drop Call Rate (DCR), Call Setup Failure Rate (CSFR), Bad Coverage percentage, Signal-to-noise ratio (SNR), Average CQI value and throughput.

When deploying a network, or an entity within an existing network, the operator typically optimizes the performance in view of one or more KPIs. This is a difficult and computational intensive task and involves analysis by engineers, typically highly skilled network engineers. As an outcome of the analysis, a particular design and/or optimization strategy may be proposed.

In order to reduce intervention by the engineers during deployment, a paradigm of Self-Organizing Networks (SON) has evolved. SON mechanisms help the engineers to save time and effort when planning, optimizing and troubleshooting the network. The SON mechanisms shall ideally detect problems, diagnose the cause of the problems and change configuration parameters in order to compensate faults and improve network performance.

The operators define several KPIs to assess status of the network. The KPIs usually measure metrics per cell or whole network. Based on the KPIs, operators build SON mechanisms to be triggered once a threshold is reached or a rule satisfied. With these mechanisms, operators improve network performance to ensure the end-user Quality of Experience (QoE).

Additionally, operators are interested in finding distinct areas of low performance, i.e. beyond a bad individual cell or global network issues, to identify the most problematic area(s) within the network. This allows efficient focusing of optimization efforts and resources to the(se) problematic area(s). Detecting such areas is based on the KPI calculation within a given area defining a geolocation. It has been proposed to detect and analyze KPIs in problematic areas over visualized in maps based on KPIs traced per area, i.e. so called geolocated trace events. Traditionally, this task is carried out by drive testing, a well-established technique to measure and assess capacity, coverage and Quality of Experience (QoE) of the network. Due to new localization techniques applied on Radio Frequency (RF) signals for user equipments in a deployed network, drive testing is becoming less important since the tasks of drive testing can be performed using reports from regular user equipments. Additionally, integration of GPS receivers into user equipments provides an accurate measurement source for determining the user equipments location.

In this context, EP1334417 discloses how to combine several observable elements, such as KPIs, to detect anomalies of a telecommunication network. Each anomaly may be associated with a location or situation in the telecommunication network. The method comprises assembling indicators indicating the behavior of the elements and arranging the assembled indicators such that each observable element's indicators are assigned to the same input data component. The learning mechanism is taught so that the input data of the learning mechanism comprises the input data components which are based on the assembled indicators. Points which approximate the input data are placed in the input space. A presentation of time is incorporated into at least one input data component wherein the presentation of time is periodic, continuous and unambiguous within the period of the at least one element with periodic time-dependent behavior. The method of EP1334417 is very complex and advanced.

In another known example, so called problematic areas of a telecommunication system are identified by analyzing a single Key Performance Indicator (KPI). The KPI are obtained as geo-located events, typically reported by user equipments in the telecommunication system. Analysis of the KPI is a computationally expensive task and results in that many area are identifies as problematic in view of the analyzed KPI. Several analyses, one of a respective KPI, may be performed, but super-positioning of the areas is not straight forward and relies on human expertise.

In view of the above, a problem may be how to improve the method of detecting anomalies, aka problematic areas.

SUMMARY

An object may thus be to improve detection of at least one specific area of a telecommunication system, in which specific area a degradation of performance is detected.

According to an aspect, the object is achieved by a method performed by a network node. The network node identifies at least a specific area among a set of areas of a wireless communication system. The identifying is based on relevance values for the set of areas and on a set of weighted Key Performance Indicators (KPIs) indicating performance of the wireless communication system. The set of weighted KPIs comprises a respective set of weighted KPIs for each area of the set of areas. Each weighted KPI is associated with a respective weight value.

According to another aspect, the object is achieved by a network node configured for identifying at least a specific area among a set of areas of a wireless communication system. The network node is configured for performing the identifying based on relevance values for the set of areas and on a set of weighted KPIs indicating performance of the wireless communication system, wherein the set of weighted KPIs comprises a respective set of weighted KPIs for each area of the set of areas. Each weighted KPI is associated with a respective weight value.

According to further aspects, the object is achieved by a computer program and a carrier therefor corresponding to the aspects above. Thanks to that the network node identifies, or detects, the specific area based on the relevance values and on the set of weighted KPIs, improved identification of the specific area is achieved. The improvement may be in terms of flexibility, since an operator, or rather an employee thereof, may choose the set of weighted KPIs, i.e. the values thereof, according to its own preferences. For example, the operator may wish to reduce number of dropped calls. Then, a weight of one or more corresponding KPI(s) can be tuned; for example, a weight value assigned to a certain -first- KPI can be set higher than another weight value assigned to another -second- KPI. Moreover, flexibility in terms of which area(s) are more important than other areas is achieved thanks to the relevance value. As an example, the relevance value may be set based on traffic volume in each area of the set of areas. Therefore, by taking the relevance values and the set of weighted KPIs into account, the network node identifies the specific area in a more flexible manner. An advantage is that the network node may consistently and repeatedly perform the method above without, or with little, intervention by a human. That is to say, human intervention may only be required initially, e.g. in order to set the weighted KPIs and/or to set the relevance value or a basis on which the relevance value may be determined. As mentioned above, the basis may be traffic volume in each area.

BRIEF DESCRIPTION OF THE DRAWINGS

The various aspects of embodiments disclosed herein, including particular features and advantages thereof, will be readily understood from the following detailed description and the accompanying drawings, in which:

Figure 1 is a schematic overview of an exemplifying wireless communication system in which embodiments herein may be implemented,

Figure 2 is a flowchart illustrating embodiments of the method in the network node,

Figure 3 is another flowchart illustrating another embodiment of the method in more detail,

Figure 4 is a further flowchart illustrating a further embodiment of the method in more detail,

Figure 5 is yet another flowchart illustrating yet another embodiment of the method in more detail,

Figures 6a-6d are diagrams illustrating maps in which performance indicated by

KPIs is plotted,

Figure 6e is a further diagram illustrating a map in which traffic volume is plotted, Figure 7 is a still further diagram illustrating a map in which a specific area has been identified,

Figures 8a-8c are diagram illustrating maps in which causes is plotted,

Figure 9 is a diagram illustrating influence per each cause in a specific area, and

Figure 10 is a block diagram illustrating embodiments of the network node.

DETAILED DESCRIPTION

Throughout the following description similar reference numerals have been used to denote similar features, such as nodes, actions, steps, modules, circuits, parts, items elements, units or the like, when applicable. In the Figures, features that appear in some embodiments are indicated by dashed lines.

Figure 1 depicts an exemplifying wireless communication system 100 in which embodiments herein may be implemented. In this example, the wireless communication system 100 is a Long Term Evolution (LTE) system.

In other examples, the wireless communication system 100 may be any cellular or wireless system, such as a Global System for Mobile communications (GSM), Universal Mobile Telecommunication System (UMTS), LTE and Worldwide

Interoperability for Microwave Access (WiMAX) or evolutions thereof. The wireless communication system 100 comprises a network node 1 10. The network node 1 10 may be part of an Operation and Maintenance (OAM) system of the wireless communication system 100, or may be part of an Operations Support System (OSS) of said communication system 100. The network node 110 may be hosted by a so called cloud service or the like. In some examples, the network node 1 10 may be external to a network domain of the wireless communication system 100.

The wireless communication system 100 comprises a set of areas 131 , 132, 133. Each area 131 , 132, 133 may comprises one or more service areas, such as cells 140 (only one cell denoted with 140 for reasons of simplicity) or sectors of a cell. The set of areas may thus be geographical areas, which may be predetermined or selected by an operator of the wireless communication system 100. In other cases, the areas may be selected independently of the service areas.

The areas may be overlapping (not shown). This means that a service area, of a portion thereof, may be associated with two areas that are overlapping. As an example, each area of the set of areas may enclose one or more service areas of the wireless communication system 100. Moreover, each area of the set of area may overlap with at least one other area. Overlap of the areas may ensure that a potential problematic area is not identified due to unfortunate splitting of the potential problematic area into two different neighboring areas. Such splitting of the potential problematic area would cause any degraded performance to been seen as a lesser degradation in the two different neighboring areas.

Moreover, shapes of the set of areas 131 , 132, 133 may be the same or may be different for the set of areas. Examples of shapes are square, triangle, rectangle, pentagon, hexagon, polygon, titled versions thereof, and the like.

Furthermore, a radio network node 150 is shown in Figure 1. The radio network node 110 may be associated with one or more service areas, such as cells 140. As used herein, the term "radio network node" may refer to a Base Station System (BSS), a Radio Network Controller (RNC), a Radio Base Station (RBS), an evolved Node B (eNB), a control node controlling one or more Remote Radio Units (RRUs), an access point or the like. The radio network node 150 may communicate with the network node 1 10. This communication may include user transmissions and/or control transmissions in respect to one or more wireless devices. The user transmissions may include user data, payload data, content data etc. The control transmissions may include control information relating to e.g. scheduling, authentication, mobility etc.. The communication may include uplink transmission and/or downlink transmission.

Figure 2 illustrates an exemplifying method according to embodiments herein when implemented in the wireless communication system 100 of Figure 1.

One or more of the following actions may be performed in any suitable order.

Action 210

The network node 1 10 may initialize KPIs and/or weights and/or relevance values. In order to provide the network node 1 10 with information about weight values for the KPIs, the operator may provide, e.g. via user interface, a respective weight value for each KPI. The KPIs are thus associated with weight values, wherein each KPI is associated with the respective weight value for said each KPI. In preferred

embodiments, at least two weight values are different from each other. Expressed differently, the respective weight value for a first KPI is different from the respective weight value for a second KPI. In some cases, the weight values may be predefined.

As an example, the operator may set the weight values according to its own policies. For example, KPI(s) relating to DCR may be defined as twice as important as KPI(s) relating to CSFR.

Moreover, in order to provide the network node 1 10 with information about the relevance values, the operator may provide, e.g. via the user interface, the relevance values for the set of areas 131 , 132, 133. A first relevance value for a first area among the set of areas 131 , 132, 133 may be equal to a second relevance value for a second area among the set of areas 131 , 132, 133. In order to differentiate relevance for the areas, it may be preferred that at least one area among the set of areas 131 , 132, 133 has a specific relevance value that is different from at least one other area among the set of areas 131 , 132, 133. In some case, the relevance values may be predefined.

In addition or alternatively, the relevance values may be calculated from a metric, such as a data volume in said each area per time unit, a number of wireless

communication devices in said each area, a number of connections in said each area, and the like. See also action 220.

When the network node 1 10 initializes the KPIs, it may be that the network node 1 10 receives, reads, or measures, values of KPIs for one or more of the set of areas 131 , 132, 133.

As will be apparent from the present disclosure, one or more of weights, a selection of KPIs, a basis for relevance value and target KPIs may be configured by the operator.

The initialization is referred to as "STAGE 1 " and is further elaborated below with reference to Figure 3. Action 220

The network node 1 10 identifies at least a specific area among the set of areas of the wireless communication system 100.

The identifying is based on relevance values for the set of areas and on a set of weighted Key Performance Indicators (KPIs) indicating performance of the wireless communication system 100.

The respective relevance value may be indicated by one or more of:

a data volume in said each area per time unit,

a number of wireless communication devices in said each area,

a number of connections in said each area, and the like.

This means that a basis, on which the respective relevance value may be generated from, may be any one or more of these above listed items.

The data volume may be referred to as traffic volume herein.

The number of wireless communication devices may refer to in-active and/or active wireless device in said each area, wherein active may refer to attached and/or connected mode with respect to the wireless communication network 100.

The set of weighted KPIs may comprise one or more KPIs relating to a service area and/or one or more KPIs relating geo-located events reported by a user equipment.

The set of weighted KPIs, relating to the geo-located events and/or the service area, may comprise one or more of:

Drop Call Rate (DCR),

Call Setup Failure Rate (CSFR),

Bad Coverage percentage,

Handover failure rate (HOFR),

Handover drop rate (HODR),

Low Signal-to-Noise Ratio (SNR) samples Rate,

Low CQI samples Rate,

High Uplink Interference samples Rate.

Low cell throughput samples Rate,

Low user throughput samples Rate,

QPSK transmissions Rate, RANK1 transmissions Rate, and the like.

The set of weighted KPIs comprises a respective set of weighted KPIs for each area of the set of areas. In this manner, each area of the set of areas may have its own weighting of the KPIs. For example, in some areas, such as along a highway, a low number of Handover failures may be more relevant for performance of the wireless communication system 100, e.g. in terms of QoE, than other KPIs. Thus, KPI(s) relating to Handover failures are assigned a weight, or weight value aka value of weight, higher than the one assigned to other KPIs. Also, for example, it may be more relevant that Call Setup Failure rate is low than other KPIs at certain areas, such as at an airport where phones often are turned on (and off).

As mentioned above, each weighted KPI is associated with a respective weight value. This means that the weighted KPIs may be associated with weight values. In particular, at least two weight values may be different from each other. Expressed differently, a first one of the weighted KPIs may be different from a second one of the weighted KPIs.

In some embodiments, the relevance values may comprise a respective relevance value for each area of the set of areas, wherein a set of penalty values comprises a respective penalty value for each area of the set of areas. In this example, a higher penalty value indicates a more problematic area.

In these embodiments, the network node 1 10 identifies the specific area by calculating, for each area of the set of areas, the respective penalty value representing weighted and normalized KPIs multiplied by the respective relevance value for said each area. The respective penalty value for the specific area exceeds a first threshold value for identification of at least the specific area. Thus, one or more specific areas may be identified as exceeding the first threshold value. This means that the first threshold value sets out limits for the set of penalty values. Thus, the expression "exceeds" means that the respective penalty value is above the first threshold and/or that the respective penalty value is below the first threshold value. The latter case may be applicable when lower penalty values indicates a more problematic area.

An advantage may be the method according to some embodiments is scalable, since e.g. the respective penalty value is calculated on a per said each area basis, which allows for parallel processing. It shall be understood that parallel processing may enable scalability. The first threshold value may be set as an absolute value or as a relative value. When the first threshold is set as an absolute value, it is ensured that the specific area is problematic in an absolute sense, not only that the specific area is problematic relatively other areas. However, a risk is that no area is identified as the specific area. When the first threshold is set as a relative value, it is instead ensured that the specific area is problematic in view of the other areas. As an example, the first threshold may be set as a percentage value, which may have an effect of identifying one or more specific areas.

In some embodiments of the method herein, the network node 1 10 performs a method for selecting a Self-Organizing-Network "SON" function to be applied to the specific area among the set of areas of the wireless communication system 100. In these embodiments, action 230 and 240 below may be performed. In these embodiments, it is described how to automatically, i.e. without or with little intervention by a human, select a SON function to be applied in the specific area.

The SON function may comprise one or more of a tool for adjusting network parameters to mitigate at least one cause, an activity adjusting network parameters to mitigate at least one cause, a system for adjusting network parameters to mitigate at least one cause, an automatic execution from any network entity adjusting network parameters to mitigate at least one cause, and the like.

This action will help operators, such as employees, engineers of the operator, to concentrate optimization efforts to one or more areas with a highest penalty. This will reduce time and required resources, which means that cost for operational maintenance will be less. Additionally, the identified area may be an input to action 230.

Reduction of resources may be that instead of applying a number of optimization techniques to an area, only the SON functions relating to the cause(s) of problems in the area are applied since the SON function will obtain a larger impact than more or less arbitrarily applying the number of optimization techniques.

Moreover, it may be that there is no need to do feasibility studies for

understanding what needs to be done in the area to achieve improvements of the selected KPIs.

Furthermore, instead of applying optimization techniques to the entire network, the SON function is applied only to one or more areas identified as problematic. It may also be that there is no need for engineers to analyze multiple KPIs in order to find out which SON function to apply.

Action 230

The network node 1 10 may determine cause scores for at least the specific area based on the set of weighted KPIs, wherein the cause scores comprises a respective cause score for each of a plurality of causes, wherein a respective subset of weighted KPIs is related to the respective cause score, wherein the respective subset of weighted KPIs is selected from the set of weighted KPIs.

In some embodiments, this action is performed for any one of the areas of the set of area. In a particular embodiment, the network node 1 10 may determine respective cause scores for said each area. As an example, the cause scores may be a set of cause scores. Then, according to these embodiments, the set of cause scores comprises sub-sets of cause scores. Each sub-set is a respective sub-set of cause scores for said each area.

Moreover, the network node 110 may identify at least the specific area and determine the cause scores and/or the respective cause scores at least partially concurrently. Thus, allowing for efficient scalability of at least some embodiments herein.

Action 240

Subsequent to action 230, the network node 1 10 selects, based on the cause scores, a Self-Organizing-Network (SON) function to be applied to the specific area. The SON function is selected, by the network node 1 10, from among a set of SON functions. Each SON function of the set of SON functions is adapted to mitigate performance degradation due to at least one cause of the plurality of causes.

It shall be noted that each cause score indicates a respective cause, such as Radio Frequency (RF) problems, missing neighbours, or the like. The network node 1 10 may select the SON function for which the respective cause score exceeds a second threshold value for selection of SON function. As mentioned above in relation to the first threshold value, one or more SON functions may be selected due to that the respective cause score exceeds the second threshold value. This means that the second threshold value sets out limits for the cause scores. Thus, the expression "exceeds" means that the respective cause score is above the second threshold and/or that the respective cause score is below the second threshold value.

The second threshold value may be set as an absolute value or as a relative value.

An aim of this action is to propose an appropriate optimization strategy, as an example of a SON function, to be deployed in a polygon area in order to efficiently focus efforts in resolving faults.

Operators' resources are limited and must be allocated wisely to ensure maximum profit. Resource allocation includes distribution of optimization efforts. With embodiments including action 240, a SON function is selected. As further examples, the SON function may be an optimization activity, design activity and/or the most suitable software tool from an optimization tool portfolio that operators should deploy to improve performance of their cellular network. The selection of the SON function is at least based on chosen KPIs combined with flexible multiplicative weights. Based on KPI analysis, the proposed method automatically identifies problematic areas, i.e. the specific areas, where the network performance is degraded. Identified areas are prioritized for optimization and design actions. Such problematic area detection and optimization approach selection should significantly help operators and engineers in complex troubleshooting and optimization tasks.

According to the embodiments herein, a majority of calculations uses only one or more of multiplication, division, addition and subtraction. Thus, the embodiments herein may be realized at a low computational cost. According to the various embodiments herein, a requirement on input from the operator is low. This means that the embodiments herein are simple to apply for the operator (e.g. telecommunications network operator owning at least part of the wireless communication system 100), and also minimizes human user intervention. Typically, as indicated above, it may be enough that the operator inputs weight values for a number of KPIs to be monitored in order to e.g. find the specific area. Furthermore, in some embodiments, it may be that the operator may need to identify for which causes the cause scores may be determined. The operator may be able to tune the wireless communication system 100 by providing a small amount of input, e.g. at a high level, such as in terms of the target KPI values, weight values for weighting KPIs, metric for indicating relevance and/or the like.

In view of the above, a further advantage that some embodiments herein may be repeated without being impacted by a human's expertise, which may very well be the case according to some known manners of finding and selecting a SON function.

An additional advantage is that the embodiments herein may easily be repeated without any human intervention once configurable parameters, such as one or more of weight values, number of KPIs, target KPI etc, have been set.

As a consequence the directly above mentioned ability to be repeated, the operator may allow tuning, by means of the method herein, to be performed as often as deemed necessary, e.g. twice a day, one a month, etc. In a further example, a problematic area may be identified by use of known methods, such as described in aforementioned EP1334417. Thereafter, action 230 and 240 may be applied. Thus, in this further example, action 220 is not performed.

Figure 3 illustrates exemplifying sub-actions of action 210 in "STAGE 1 ". One or more of the following actions may be performed in any suitable order.

Action 310

The network node 1 10 may read KPIs, which may be computed from geolocated trace events and/or cell level counters that can e.g. be received in the network node 1 10.

This action may be separated from the initialization in STAGE 1. As an example, the reading, or gathering, of the KPIs, i.e. the values of the KPIs for each area, may preferably be performed before the values of the KPIs are used in action 220 and/or action 230. Accordingly, in some examples, action 310 is a sub-action of action 220 and/or action 230.

Action 320

The network node 1 10 may select KPIs and a relevance indicator. In particular, an operator can control the network node 1 10, e.g. via a user interface, to select the KPIs and the relevance indicator. This means that the operator may select a sub-set of KPIs to be monitored and used in order to identify the specific area, e.g. as in action 220.

The relevance indicator may be used to create a respective relevance value for each of the areas of the wireless communication system 100. As it may turn out, due to the relevance indicator, one or more areas may have the same relevance value. For example, if traffic volume is used as basis for the respective relevance value, any two areas will have the same relevance value when the traffic volume in these two areas is equal. Action 330

The network node 1 10 may define weights and target KPIs (Τ_ΚΡΙ,). Similarly, to the preceding action, the operator may control the network node 110, e.g. via the user interface, to perform this action. As an example, the operator may set dynamically weights for one or more target KPIs, such as as 0.5% for DCR related KPI(s) and 0.1 % for CSFR related KPI(s).

Action 340

The network node 1 10 may define the set of areas of the wireless communication system 100. Again, it may be that the operator controls the network node, via the user interface, to perform this action.

Figure 4 illustrates exemplifying sub-actions of action 220 in "STAGE 2". Action 410, 420, 430 and 440 may be performed for each area of a set of areas of the wireless communication system 100. One or more of the following actions may be performed in any suitable order.

Action 410

The network node 1 10 may select an area for which a respective penalty value is to be calculated. The calculation of the respective penalty value may be based on values of KPIs, e.g. computed from geolocated events and/or cell level counters, for the selected areas. The values of KPIs may be retrieved from a memory, gathered or otherwise obtained by the network node 110.

Notably, this selected area is not yet necessarily compared to other areas. At this stage, i.e. initially in "STAGE 2", the network node 1 10 has no knowledge about whether or not this selected area is, or will be considered to be, the specific area, e.g. a more problematic area compared to other areas.

After action 410, action 420 may be performed. Alternatively, action 430 is performed after action 410. Moreover, it may be that actions 420 and 430 are performed at least partially concurrently after action 410, i.e. any one of actions 420 and 430 may begin slightly before or after the other. When actions 420 and 430 are performed at least partially concurrently, use of parallel processing is possible. This means that at least some embodiments herein are scalable while not suffering serious performance degradation.

Action 420

In order to prepare for action 440 the network node 1 10 may determine a normalized KPI factor based on weight values, values of KPIs and target KPIs.

The target KPI may, similarly to the weight values, be predetermined or selected by the operator. The target KPI may define a maximum acceptable KPI value, while assuming a low KPI means "good" conditions.

Action 430

The network node 1 10 may calculate the respective relevance value for the selected area. As already mentioned above, the respective relevance value may be determined based on a metric, such as one or more of a data volume in said each area per time unit, a number of wireless communication devices in said each area, a number of connections in said each area, and the like.

Action 440

For the selected area, the network node 1 10 calculates a penalty value, e.g. according to a penalty function (or more generally ction):

P_a = relevance_a ^■∑f_{= 1} w_t ^■

o) where identifies the selected area, relevance_a is the relevance value for the selected area , N indicates the number of analyzed KPIs, Wj is the respective weight value associated with KPI i, KPI? is a value of the KPI i in the area and T_KPIi is a value of the "target KPI" for KPI i. As explained above, the factor may have been calculated

in action 420 and the relevance _a may have been calculated in action 430. Of course, actions 420, 430 and 440 may be seen as one action, in which the penalty value is calculated.

Action 450

The network node 1 10 may check whether there are further areas for which a respective penalty value is to be calculated according to action 440.

In case there are further areas, the method proceeds with action 410. Otherwise, the method proceeds to action 460 below.

Action 460

Based on the penalty values determined for each of the areas, the network node

1 10 may build, e.g. construct, generate or similar, information from which the specific area, i.e. one or the most problematic areas, may be extracted. The information may be organized a map illustrating the areas, where a pattern, or color of the area indicates the penalty value. Maps for different KPIs are illustrated in Figures 6a to 6d below and a map for the relevance values is illustrated in Figure 6e. Finally, Figure 7 illustrates a further map in which one or more specific areas are marked, i.e. one or more areas that are considered to be more problematic than others in view of the penalty values determined in the preceding actions. As an example, the most problematic area may be given by argmax( ⁾ _a).

Figure 5 illustrates exemplifying sub-actions of action 230 in "STAGE 2". Action 510, 520, 530, 531 ...539, 540, 541 ...549, and 550 may be performed for each area of a set of areas of the wireless communication system 100. In these sub-actions causes that are main contributors to a selected area's KPIs are identified. One or more of the following actions may be performed in any suitable order. Action 510

The network node 110 may select an area for which a respective contribution value C^lause j_{S to ca}|_CU|_at_ec| |_n some examples, the respective contribution value (-cause j_{g on}|y _ca|_CU|_at_ec| f_{or one or m0}re identified specific areas, i.e. those areas considered to be more problematic than other areas. It may also be the case that the respective contribution value C^^ause is calculated for all areas, or all most all areas.

Action 520

In order to calculate the respective contribution value C^^ause , the network node 1 10 may find out, e.g. by input from the operator as explained for the weight values, which KPIs are related to which cause or causes.

This means that for each cause, the network node 110 may select a sub-set, or even all, KPIs of the KPIs used when calculating the respective penalty values in order to obtain a respective influence value i^^ause as determined in action 540, 541 ... 549 below. The selection of sub-sets may be performed for each area as the method of Figure 5 proceeds to the next area in action 550, or this action 520 may be performed before action 510 in order to selected sub-sets in advance for each area. That is to say, this action may not necessarily be repeated on a per each area basis.

Action 530, 531 ... 539

The network node 1 10 may calculate, or determine, which KPIs are relevant for each cause to be analyzed.

Action 540, 541 ... 549

For each of the causes, the network node 1 10 may calculate the respective influence value i^^ause for each cause. The respective influence value i_a ^cause may be given by:

N

KPI?

cause (cause)

la

T_KPl_t

i=l where a identifies the studied polygon area, N indicates the number of analyzed KPIs, Wj is the weight related to KPI i, KPl (cause) is the KPI i in the area a calculated with those events labelled with the studied cause cause and T_KPIi is the "target KPI" of KPI i. This action (and at least the preceding action) may be repeated for every area.

Notably, in some rare cases, I^^ause = P_a when studied cause cause is related to all analyzed KPIs. In other cases, the respective penalty value may be given by P_a = I^^ause +∑ L Wj ^{■ KP!l} , where M is a number of KPIs which are not

T_KPIi

related to the cause cause. Therefore, it may in some cases be beneficial to calculate the respective influence value I^^ause before the respective penalty value P_a.

Action 550

The network node 1 10 may check whether there are further areas for which a respective influence value i^^ause is to be calculated according to action 530 and 540.

In case there are further areas, the method proceeds with action 510. Otherwise, the method proceeds to action 560 below. In some cases, the respective influence values for only the specific area may be determined, if the specific area is known from e.g. action 220. Action 560

Now that the respective influence values i^^ause for all causes to be analyzed have been determined, a contribution of each cause for each area may be determined by:

jcause

r cause ^l where cause identify the studied cause and a identify the selected (studied) area and R indicates a number of analyzed causes. Each cause of the number of analyzed causes R is related to at least one of the KPIs under study, i.e. at least one of the N KPIs. As an example, a most contributing cause may be given by ( cause\

argmax \ L_a I.

Then the appropriate SON function according to the most contributing cause cause_max may be selected.

Figures 6a through 6d illustrates maps for a number of KPIs. In this example, the operator is interested in improving the network performance by analyzing the following KPIs: CSFR (call setup failure rate), DCR (drop call rate), IRAT HO rate (Inter radio access technology handover) and bad coverage percentage. These KPIs are computed based on geolocated traced events in an LTE network with a 75% overlap between test polygons. The test polygons are thus examples of the set of areas.

Figure 6e illustrates relevance of each test polygon in terms of traffic volume within each test polygon.

Figure 7 provides ratios of each KPI where test polygons are drawn depending if KPI value is high, medium or low. The distributions shown by these figures are inputs to the cost function, together with the KPI weights and the maximum acceptable KPI value. For this example, the defined thresholds and weights given to each KPI are shown in the table below:

Example values of T_KPIi has been provided. The contribution of each KPI the cost function is the same. Therefore, all the weights are set to 1. In other examples, at least two weight values may differ from each other. This allows the operator to fine tune influence of one or more KPIs.

Once the second stage "STAGE 2" has computed the cost function and the penalty of each polygon area has been obtained, the results are provided in visual maps and reports. For this example, some of that information is illustrated in Figure 6. The image shows the penalty of each area classified as "Very Bad Performance", "Bad Performance" or "Good Performance" meaning that worst areas are the ones with "Very Bad Performance". Note that, due to the 75% of overlap, the full area with very bad performance (two areas for this example) is represented with an square (two squares for this example). Additional statistical information would be provided to the engineers about this analysis such as 3D maps or statistics (average, 90^th percentile, etc.). Figures 8a through 8c illustrates contribution of each cause per test polygon for the general case. It is however enough to calculate the contribution of each cause for the most problematic area identified in the preceding stages. In this case, three kinds of root causes have been analyzed: missing neighbors, RF issues and unknown issues.

Next, the appropriate SON function, i.e. one or more optimization methods and/or software tools, is identified based on the contribution of each root cause. In this example, the contribution of each root cause in the most problematic area is depicted in Figure 9, which means that the operator should focus its resources at a SON function that optimizes and troubleshoots RF issues as the penalty of RF issues presents the 90% of the contribution.

Now that the SON function that optimizes and troubleshoots RF issues has been identified, it may be that there are several SON functions to choose between. In this case, one of the several SON functions is selected based on e.g. network topology and/or performance, where the performance again may be given by one or more of the KPIs. Once a single SON function has been identified, any commercial optimization tool available may be loaded to execute design and optimization activities in the most problematic area.

According to some embodiments of the method herein, improved network performance is achieved in the most problematic area, e.g. even without human intervention. These embodiments may be applied to multiple problematic area with low requirements in investments and resources compared to requirements on operators and engineers to manually achieve a similar result. With reference to Figure 10, a schematic block diagram of embodiments of the network node 1 10 of Figure 1 is shown.

The network node 1 10 may comprise a processing module 1001 , such as a means for performing the methods described herein. The means may be embodied in the form of one or more hardware modules and/or one or more software modules

The network node 1 10 may further comprise a memory 1002. The memory may comprise, such as contain or store, a computer program 1003. According to some embodiments herein, the processing module 1001 comprises, e.g. 'is embodied in the form of or 'realized by', a processing circuit 1004 as an exemplifying hardware module. In these embodiments, the memory 1002 may comprise the computer program 1003, comprising computer readable code units executable by the processing circuit 1004, whereby the network node 1 10 is operative to perform the method of Figure 2.

In some other embodiments, the computer readable code units may cause the network node 1 10 to perform the method according to Figure 2 when the computer readable code units are executed by the network node 1 10.

Figure 10 further illustrates a carrier 1005, or program carrier, which comprises the computer program 1003 as described directly above.

In some embodiments, the processing module 1001 comprises an Input/Output module 1006, which may be exemplified by a receiving module and/or a sending module as described below when applicable.

In further embodiments, the processing module 1001 may comprise one or more of an initiating module 1010, an identifying module 1020, a determining module 1030 and a selecting module 1040 as exemplifying hardware modules. In other examples, one or more of the aforementioned exemplifying hardware modules may be implemented as one or more software modules.

Accordingly, the network node 1 10 and/or the processing module 1001 and/or the identifying module 1020 is configured for identifying at least a specific area among a set of areas of a wireless communication system 100, wherein the network node 110 is configured for performing the identifying based on relevance values for the set of areas and on a set of weighted KPIs indicating performance of the wireless communication system 100, wherein the set of weighted KPIs comprises a respective set of weighted KPIs for each area of the set of areas. Each weighted KPI is associated with a respective weight value.

The set of weighted KPIs may comprise one or more of:

Drop Call Rate (DCR),

Call Setup Failure Rate (CSFR),

Bad Coverage percentage,

Handover failure rate (HOFR),

Handover drop rate (HODR),

Low Signal-to-Noise Ratio (SNR) samples Rate,

Low CQI samples Rate,

High Uplink Interference samples Rate.

Low cell throughput samples Rate,

Low user throughput samples Rate,

QPSK transmissions Rate,

RANK1 transmissions Rate, and the like.

The network node 1 10 and/or the processing module 1001 and/or the

determining module 1030 may be configured for determining cause scores for at least the specific area based on the set of weighted KPIs, wherein the cause scores comprises a respective cause score for each of a plurality of causes, wherein a respective subset of weighted KPIs is related to the respective cause score, wherein the respective subset of weighted KPIs is selected from the set of weighted KPIs.

Moreover, the network node 1 10 and/or the processing module 1001 and/or the selecting module 1040 may be configured for selecting, based on the cause scores, a SON function to be applied to the specific area, wherein the SON function is selected from among a set of SON functions, wherein each SON function of the set of SON functions is adapted to mitigate performance degradation due to at least one cause of the plurality of causes.

The SON function may comprise one or more of:

a tool for adjusting network parameters to mitigate at least one cause, an activity adjusting network parameters to mitigate at least one cause, a system for adjusting network parameters to mitigate at least one cause, an automatic execution from any network entity adjusting network parameters to mitigate at least one cause, and the like.

The relevance values may comprise a respective relevance value for each area of the set of areas, wherein a set of penalty values comprises a respective penalty value for each area of the set of areas.

The respective relevance value may be indicated by one or more of:

a data volume in said each area per time unit,

a number of wireless communication devices in said each area, and

a number of connections in said each area.

Furthermore, the network node 1 10 and/or the processing module 1001 and/or the identifying module 1020, or another identifying module (not shown), may be configured for identifying the specific area by calculating, for each area of the set of areas, the respective penalty value representing weighted and normalized KPIs multiplied by the respective relevance value for said each area, wherein the respective penalty value for the specific area exceeds a first threshold value for identification of at least the specific area.

The network node 1 10 and/or the processing module 1001 and/or the selecting module 1040 , or another selecting module not shown, may be configured for selecting of the SON function by selecting the SON function for which the respective cause score exceeds a second threshold value for selection of SON function.

The network node 1 10 and/or the processing module 1001 and/or the

determining module 1030, or another determining module not shown, may be configured for determining cause scores by determining respective cause scores for said each area.

Said each area may enclose one or more service areas of the wireless communication system 100.

Each area of the set of area may overlap with at least one other area.

The network node 1 10 and/or the processing module 1001 and/or the identifying module 1020 and the determining module 1030 may be configured for performing the identifying of at least the specific area and the determining of the cause scores and/or the respective cause scores at least partially concurrent. As used herein, the term "node", or "network node", may refer to one or more physical entities, such as devices, apparatuses, computers, servers or the like. This may mean that embodiments herein may be implemented in one physical entity. Alternatively, the embodiments herein may be implemented in a plurality of physical entities, such as an arrangement comprising said one or more physical entities, i.e. the embodiments may be implemented in a distributed manner, such as on a set of server machines of a cloud system.

As used herein, the term "module" may refer to one or more functional modules, each of which may be implemented as one or more hardware modules and/or one or more software modules and/or a combined software/hardware module in a node. In some examples, the module may represent a functional unit realized as software and/or hardware of the node.

As used herein, the term "wireless communication device" may refer to a user equipment, a machine-to-machine (M2M) device, a mobile phone, a cellular phone, a Personal Digital Assistant (PDA) equipped with radio communication capabilities, a smartphone, a laptop or personal computer (PC) equipped with an internal or external mobile broadband modem, a tablet PC with radio communication capabilities, a portable electronic radio communication device, a sensor device equipped with radio

communication capabilities or the like. The sensor may be any kind of weather sensor, such as wind, temperature, air pressure, humidity etc. As further examples, the sensor may be a light sensor, an electronic or electric switch, a microphone, a loudspeaker, a camera sensor etc. The term "user" may indirectly refer to the wireless device.

Sometimes, the term "user" may be used to refer to the user equipment or the like as above. It shall be understood that the user may not necessarily involve a human user. The term "user" may also refer to a machine, a software component or the like using certain functions, methods and similar.

As used herein, the term "program carrier", or "carrier", may refer to one of an electronic signal, an optical signal, a radio signal, and a computer readable medium. In some examples, the program carrier may exclude transitory, propagating signals, such as the electronic, optical and/or radio signal. Thus, in these examples, the carrier may be a non-transitory carrier, such as a non-transitory computer readable medium.

As used herein, the term "processing module" may include one or more hardware modules, one or more software modules or a combination thereof. Any such module, be it a hardware, software or a combined hardware-software module, may be a determining means, estimating means, capturing means, associating means, comparing means, identification means, selecting means, receiving means, sending means or the like as disclosed herein. As an example, the expression "means" may be a module

corresponding to the modules listed above in conjunction with the Figures.

As used herein, the term "software module" may refer to a software application, a Dynamic Link Library (DLL), a software component, a software object, an object according to Component Object Model (COM), a software component, a software function, a software engine, an executable binary software file or the like.

As used herein, the term "processing circuit" may refer to a processing unit, a processor, an Application Specific integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA) or the like. The processing circuit or the like may comprise one or more processor kernels.

As used herein, the expression "configured to/for" may mean that a processing circuit is configured to, such as adapted to or operative to, by means of software configuration and/or hardware configuration, perform one or more of the actions described herein.

As used herein, the term "action" may refer to an action, a step, an operation, a response, a reaction, an activity or the like. It shall be noted that an action herein may be split into two or more sub-actions as applicable. Moreover, also as applicable, it shall be noted that two or more of the actions described herein may be merged into a single action.

As used herein, the term "memory" may refer to a hard disk, a magnetic storage medium, a portable computer diskette or disc, flash memory, random access memory (RAM) or the like. Furthermore, the term "memory" may refer to an internal register memory of a processor or the like.

As used herein, the term "computer readable medium" may be a Universal Serial Bus (USB) memory, a DVD-disc, a Blu-ray disc, a software module that is received as a stream of data, a Flash memory, a hard drive, a memory card, such as a MemoryStick, a Multimedia Card (MMC), Secure Digital (SD) card, etc. One or more of the

aforementioned examples of computer readable medium may be provided as one or more computer program products.

As used herein, the term "computer readable code units" may be text of a computer program, parts of or an entire binary file representing a computer program in a compiled format or anything there between.

As used herein, the expression "transmit" and "send" are considered to be interchangeable. These expressions include transmission by broadcasting, uni-casting, group-casting and the like. In this context, a transmission by broadcasting may be received and decoded by any authorized device within range. In case of uni-casting, one specifically addressed device may receive and decode the transmission. In case of group-casting, a group of specifically addressed devices may receive and decode the transmission.

As used herein, the terms "number" and/or "value" may be any kind of digit, such as binary, real, imaginary or rational number or the like. Moreover, "number" and/or "value" may be one or more characters, such as a letter or a string of letters. "Number" and/or "value" may also be represented by a string of bits, i.e. zeros and/or ones.

As used herein, the term "set of may refer to one or more of something. E.g. a set of devices may refer to one or more devices, a set of parameters may refer to one or more parameters or the like according to the embodiments herein.

As used herein, the expression "in some embodiments" has been used to indicate that the features of the embodiment described may be combined with any other embodiment disclosed herein.

Further, as used herein, the common abbreviation "e.g.", which derives from the Latin phrase "exempli gratia," may be used to introduce or specify a general example or examples of a previously mentioned item, and is not intended to be limiting of such item. If used herein, the common abbreviation "i.e.", which derives from the Latin phrase "id est," may be used to specify a particular item from a more general recitation. The common abbreviation "etc.", which derives from the Latin expression "et cetera" meaning "and other things" or "and so on" may have been used herein to indicate that further features, similar to the ones that have just been enumerated, exist.

Even though embodiments of the various aspects have been described, many different alterations, modifications and the like thereof will become apparent for those skilled in the art. The described embodiments are therefore not intended to limit the scope of the present disclosure.

Claims

1. A method, performed by a network node (1 10), comprising:

identifying (220) at least a specific area among a set of areas of a wireless communication system (100), wherein the identifying (220) is based on relevance values for the set of areas and on a set of weighted Key Performance Indicators "KPIs" indicating performance of the wireless communication system (100), wherein the set of weighted KPIs comprises a respective set of weighted KPIs for each area of the set of areas, wherein each weighted KPI is associated with a respective weight value.

2. The method according to claim 1 , wherein the method further comprises:

determining (230) cause scores for at least the specific area based on the set of weighted KPIs, wherein the cause scores comprises a respective cause score for each of a plurality of causes, wherein a respective subset of weighted KPIs is related to the respective cause score, wherein the respective subset of weighted KPIs is selected from the set of weighted KPIs; and

selecting (240), based on the cause scores, a Self-Organizing-Network "SON" function to be applied to the specific area, wherein the SON function is selected from among a set of SON functions, wherein each SON function of the set of SON functions is adapted to mitigate performance degradation due to at least one cause of the plurality of causes.

3. The method according to claim 1 or 2, wherein the relevance values comprises a respective relevance value for each area of the set of areas, wherein a set of penalty values comprises a respective penalty value for each area of the set of areas, wherein the identifying (220) of the specific area comprises calculating, for each area of the set of areas, the respective penalty value representing weighted and normalized KPIs multiplied by the respective relevance value for said each area, wherein the respective penalty value for the specific area exceeds a first threshold value for identification of at least the specific area.

4. The method according to claim 2 or 3, wherein the selecting (240) of the SON

function comprises selecting the SON function for which the respective cause score exceeds a second threshold value for selection of SON function.

5. The method according to the preceding claim, when dependent on claim 3, wherein the respective relevance value is indicated by one or more of:

a data volume in said each area per time unit,

a number of wireless communication devices in said each area, and a number of connections in said each area.

6. The method according to any one of the preceding claims, wherein the set of

weighted KPIs comprises one or more KPIs relating to a service area and/or one or more KPIs relating geo-located events reported by a user equipment.

7. The method according to any one of the preceding claims, wherein the set of

weighted KPIs comprises one or more of:

Drop Call Rate (DCR),

Call Setup Failure Rate (CSFR),

Bad Coverage percentage,

Handover failure rate (HOFR),

Handover drop rate (HODR),

Low Signal-to-Noise Ratio (SNR) samples Rate,

Low CQI samples Rate,

High Uplink Interference samples Rate.

Low cell throughput samples Rate,

Low user throughput samples Rate,

QPSK transmissions Rate, and

RANK1 transmissions Rate.

8. The method according to any one of the preceding claims, wherein said each area encloses one or more service areas of the wireless communication system (100).

9. The method according to any one of claims 2 to 8, wherein the determining of cause scores further comprises determining respective cause scores for said each area.

10. The method according to any one of the preceding claims, wherein each area of the set of area overlaps with at least one other area.

1 . The method according to any one of the preceding claims, wherein the identifying (220) of at least the specific area and the determining (230) of the cause scores and/or the respective cause scores is performed at least partially concurrent.

2. The method according to any one of claims 2 to 11 , wherein the SON function comprises one or more of:

a tool for adjusting network parameters to mitigate at least one cause, an activity adjusting network parameters to mitigate at least one cause, a system for adjusting network parameters to mitigate at least one cause, and an automatic execution from any network entity adjusting network parameters to mitigate at least one cause.

13. A network node (1 10) configured for:

identifying at least a specific area among a set of areas of a wireless communication system (100), wherein the identifying (220) is based on relevance values for the set of areas and on a set of weighted Key Performance Indicators "KPIs" indicating performance of the wireless communication system (100), wherein the set of weighted KPIs comprises a respective set of weighted KPIs for each area of the set of areas, wherein each weighted KPI is associated with a respective weight value.

14. The network node (110) according to claim 13, wherein the network node (110) is configured for:

determining cause scores for at least the specific area based on the set of weighted KPIs, wherein the cause scores comprises a respective cause score for each of a plurality of causes, wherein a respective subset of weighted KPIs is related to the respective cause score, wherein the respective subset of weighted KPIs is selected from the set of weighted KPIs; and

selecting, based on the cause scores, a Self-Organizing-Network "SON" function to be applied to the specific area, wherein the SON function is selected from among a set of SON functions, wherein each SON function of the set of SON functions is adapted to mitigate performance degradation due to at least one cause of the plurality of causes.

15. The network node (1 10) according to claim 13 or 14, wherein the relevance values comprises a respective relevance value for each area of the set of areas, wherein a set of penalty values comprises a respective penalty value for each area of the set of areas, wherein the network node (1 10) is configured for identifying the specific area by calculating, for each area of the set of areas, the respective penalty value representing weighted and normalized KPIs multiplied by the respective relevance value for said each area, wherein the respective penalty value for the specific area exceeds a first threshold value for identification of at least the specific area.

16. The network node (1 10) according to claim 14 or 15, wherein the network node (1 10) is configured for selecting of the SON function by selecting the SON function for which the respective cause score exceeds a second threshold value for selection of SON function.

17. The network node (1 10) according to the preceding claim, when dependent on claim 15, wherein the respective relevance value is indicated by one or more of:

a data volume in said each area per time unit,

18. The network node (1 10) according to any one of claims 13-17, wherein the set of weighted KPIs comprises one or more KPIs relating to a service area and/or one or more KPIs relating geo-located events reported by a user equipment.

19. The network node (1 10) according to any one of claims 13-18, wherein the set of weighted KPIs comprises one or more of:

Drop Call Rate (DCR),

Call Setup Failure Rate (CSFR),

Bad Coverage percentage,

Handover failure rate (HOFR),

Handover drop rate (HODR),

Low Signal-to-Noise Ratio (SNR) samples Rate,

Low CQI samples Rate, High Uplink Interference samples Rate.

Low cell throughput samples Rate,

Low user throughput samples Rate,

QPSK transmissions Rate, and

RANK1 transmissions Rate.

20. The network node (110) according to any one of claims 13-19, wherein said each area encloses one or more service areas of the wireless communication system (100).

21 . The network node (110) according to any one of claims 14-20, wherein the network node (110) is configured for determining cause scores by determining respective cause scores for said each area.

22. The network node (110) according to any one of claims 13-21 , wherein each area of the set of area overlaps with at least one other area.

23. The network node (110) according to any one of claims 14-22, wherein the network node (1 10) is configured for performing the identifying at least the specific area and the determining the cause scores and/or the respective cause scores at least partially concurrent.

24. The network node (110) according to any one of claims 14-23, wherein the SON function comprises one or more of:

25. A computer program (1003), comprising computer readable code units which when executed on a network node (1 10) causes the network node (1 10) to perform the method according to any one of claims 1-12.

26. A carrier (1005) comprising the computer program according to the preceding claim, wherein the carrier (1005) is one of an electronic signal, an optical signal, a radio signal and a computer readable medium.