WO2016128030A1

WO2016128030A1 - Method and apparatus for managing a communication network

Info

Publication number: WO2016128030A1
Application number: PCT/EP2015/052786
Authority: WO
Inventors: Michael Grimes; Prudhvi DANDU; Raghavendra MAHALINGAPPA; Jan Groenendijk; Premnath KANDHASAMY NARAYANAN
Original assignee: Telefonaktiebolaget LM Ericsson AB
Current assignee: Telefonaktiebolaget LM Ericsson AB
Priority date: 2015-02-10
Filing date: 2015-02-10
Publication date: 2016-08-18
Anticipated expiration: 2017-08-10

Abstract

A method (100) for managing a communication network is disclosed, the communication network capable of supporting a plurality of network functions. The method comprises discovering network functions deployed in the network (1 10), obtaining a measure of network performance (120), evaluating the measure of network performance to identify a network performance condition (130), identifying a network function to address the identified network performance condition (140), and deploying the identified network function (160). Also disclosed are a network element (300, 400) and a computer program product configured to carry out a method for managing a communication network.

Description

METHOD AND APPARATUS FOR MANAGING A COMMUNICATION NETWORK

Technical Field The present invention relates to a method and network element for managing a communication network. The present invention also relates to a computer program product configured, when run on a computer, to carry out a method for managing a communication network. Background

Communication networks support a range of network functions which contribute to the smooth functioning of the network. Examples of network functions may include optimisation, packet inspection, load balancing, Automatic Neighbour Relations (ANR) etc. Network functions may be realised in a centralised, distributed or hybrid manner. Centralised network functions are implemented in centralised network components, including for example a network Operations Support System or central controller. Distributed network functions are implemented outside of the central elements, for example in network elements forming part of an operator access network. Hybrid network functions are shared between centralised and distributed network elements. Figure 1 illustrates example operator networks comprising centralised network functions 1 , distributed network functions 2 and hybrid network functions 3. The choice of how to deploy network functions within a network is typically made by the network operator on the basis of standard design assumptions.

Network Function Virtualisation (NFV) is a network architecture concept involving the virtualisation of network functions into building blocks which may be chained together to provide end to end services. NFV proposes migrating from a proprietary hardware model, in which application specific hardware is deployed to provide network functions, to a virtualised model, in which network functions are realised as virtual machines running different software or processes. These virtual machines may run on industry standard high volume servers, switches and storage, and may for example be housed on cloud computing data centres. Virtualised Network Functions (VNFs) may be Software Defined Network (SDN) functions, Self Organising Network (SON) functions, Autonomous Network Functions (ANFs) or other types of network function. In a virtualised environment, centralised, distributed and hybrid network functions may all be realised as virtual machines on the same industry standard hardware. For example, a centralised network function and a distributed network function could be run as virtual machines or virtual containers on the same cloud data centre. Advantages of NFV include both flexibility and reduced development lead time, as virtualised network functions may be tested and readied for deployment much faster than the corresponding hardware based solutions.

Some network functions may be repeated in both centralised and distributed implementations. Additionally, certain network operators may deploy network functions from different vendors and using different underlying technologies within a single network. This creates an overlapping feature environment for the network operator, and may lead to conflicts when the different network functions attempt to operate at the same time. Conflicts may manifest for example in the form of conflicting results for the same parameter set being produced by a centralised and distributed implementation of overlapping network functions. Individual workarounds are required to avoid such conflicts, imposing a burden of manual tuning whenever a more complex deployment of functions, or multivendor solution is implemented. This manual tuning negates many of the advantages offered by SON, SDF and ANF functions when deployed individually. Summary

It is an aim of the present invention to provide a method, apparatus and computer readable medium which at least partially address one or more of the challenges discussed above.

According to a first aspect of the present invention, there is provided a method for managing a communication network, the communication network capable of supporting a plurality of network functions. The method comprises discovering network functions deployed in the network, obtaining a measure of network performance, evaluating the measure of network performance to identify a network performance condition, identifying a network function to address the identified network performance condition, and deploying the identified network function.

According examples of the invention, the network may be physical, virtual or may combine both physical and virtual elements. The method may run as a process for example in a virtual machine or Linux Container (LXC). According to some examples of the invention, the identified network function may comprise at least one of a Self Organising Network, SON, function, a Software Defined Network, SDN, function, or an Autonomous Network Function, ANF. Example network functions may include optimisation, packet inspection, load balancing, Automatic Neighbour Relations (ANR), Mobility Robustness (MRO), energy saving, transport capacity etc. In some examples, the identified network function may comprise a SON use case as defined in 3GPP TR 36.902. According to some examples of the invention, the identified network function may comprise a Virtualised Network Function, VNF. In some examples, the identified network function may comprise a plurality of VNFs, the plurality of VNFs cooperating to provide the network function. According to some examples of the invention, discovering network functions deployed in the network may comprise discovering at least two network functions having different underlying technology or supplied by different vendors. In some examples, discovering network functions deployed in the network may comprise receiving information from a network element such as an Operations Support System (OSS), probing network elements, or packet inspection. The discovered functions may in some examples be centralised, distributed and/or hybrid.

According to some examples of the invention, discovering network functions deployed on the network may further comprise assembling a Network Forwarding Graph, NFG, expressing connectivity between the discovered network functions.

According to some examples of the invention, the method may further comprise updating the NFG with the deployed network function. According to some examples of the invention, identifying a network performance condition may comprise identifying an obtained measure of network performance failing to meet a predefined standard for that measure of network performance. In some examples, the standard may be defined by an operator of the network and may therefore be consistent with operator priorities for the network. In some examples, identifying a network performance condition may comprise identifying a plurality of obtained measures of network performance failing to meet a predefined collective standard for those measures of network performance. In some examples, the plurality of measures may relate to a single aspect of network performance, and the collective standard may be defined for the aspect of network performance.

According to some examples of the invention, identifying a network function to address the identified network performance condition may comprise checking for currently deployed network functions relating to the identified network performance condition, checking for network functions available for deployment, and selecting, in light of the currently deployed network functions, a network function operable to improve the identified network performance condition from among the network functions available for deployment. In some examples, checking for network functions available for deployment may comprise checking a catalogue of network functions available to the network, for example in the form of a software catalogue.

According to some examples of the invention, identifying a network function to address the identified network performance condition may further comprise checking for network functions which have previously been deployed to address the identified network performance condition. In some examples, this may comprise checking a knowledge base including details of historic performance measures and deployed functions, which knowledge base may be internal to the element hosting the process or may be external.

According to some examples of the invention, deploying the identified network function may comprise downloading software corresponding to the identified network function and instantiating the downloaded software. In some examples, the software may be instantiated in any part of the network appropriate to address the identified network performance condition and compatible with the downloaded software. The deployed network function may thus be centralised, distributed or hybrid.

According to some examples of the invention, the method may further comprise monitoring network performance following deployment of the identified network function and scaling the deployment of the network function according to the monitored network performance. Scaling of the deployed network function may comprise scaling up, scaling down, and/or redeploying the network function. Configuration changes resulting from such scaling may be updated in the NFG maintained according to some examples of the invention. According to some examples of the invention, the method may further comprise updating a knowledge base with the network performance condition, the deployed network function and the monitored network performance. This knowledge base may then be consulted to inform future identification of network functions to address similar network performance conditions.

According to some examples of the invention, the method may further comprise discovering an execution environment of the network. An execution environment of the network may comprise an underlying hardware and/or software environment within which the network exists. An execution environment may thus comprise at least one of a hardware environment or a software environment such as a cloud environment.

According to some examples of the invention, discovering an execution environment of the network may comprise obtaining execution environment configurations and capabilities. The process of discovering the execution environment may thus comprise identifying network elements responsible for administration of the network and querying the identified network elements for details of the execution environment. Such network elements may for example include Operations Support Systems (OSS) in the case of a hardware execution environment, or Data Centre Managers in the case of a cloud execution environment.

According to some examples, the execution environment configurations and capabilities may comprise hardware configurations and capabilities, for example obtained from an OSS, or may comprise data centre configurations and capabilities, for example obtained from a data centre manager.

According to some examples of the invention, the method may further comprise identifying resources in the execution environment available for deployment of network functions. According to some examples of the invention, the method may further comprise reserving resources in the execution environment for deployment of the identified function. According to some examples of the invention, the method may further comprise maintaining a record of the discovered execution environment and updating the record of the discovered execution environment following deployment of the identified network function. According to some examples of the invention, identifying a network function to address the network performance condition may comprise consulting the discovered execution environment and identifying a network function appropriate to the discovered execution environment. According to some examples of the invention, the method may further comprise defining a satisfaction criterion for the identified network function; monitoring network performance against the satisfaction criterion following deployment of the identified network function; and, on attainment of the satisfaction criterion, stopping the deployed network function.

According to some examples, stopping the network function may comprise hibernating the function, decommissioning the function or un-deploying the function, for example releasing the resources used for deploying the function and reversing the system configurations made to enable deployment of the function.

According to some examples, the satisfaction criterion may be defined according to a predefined standard for an aspect of network performance.

According to some examples of the invention, the method may further comprise checking performance associated with previous instances of the defined satisfaction criterion and updating the defined satisfaction criterion according to the check. According to some examples, information regarding previous instances of the defined satisfaction criterion may be retrieved from a knowledge base. According to some examples of the invention, the method may further comprise updating a knowledge base with the network performance condition, the deployed network function, the satisfaction criterion and the monitored network performance. According to some examples of the invention, the method may further comprise discovering analytics capabilities in the network.

According to some examples of the invention, the method may further comprise identifying analytics capabilities for the identified network function. The analytics capabilities may for example be identified from a catalogue of available analytics capabilities.

According to some examples of the invention, the method may further comprise checking performance associated with previous instances of the identified analytics capabilities deployed with the identified network function. The information regarding previous instances may for example be retrieved from a knowledge base.

According to some examples of the invention, the method may further comprise deploying the identified analytics capabilities.

According to some examples of the invention, deploying the identified analytics capabilities may comprise downloading software corresponding to the identified analytics capabilities and instantiating the downloaded software. According to some examples of the invention, the method may further comprise updating the NFG with the deployed analytics capabilities.

According to some examples of the invention, the method may further comprise maintaining a record of the discovered analytics capabilities, and updating the record with the deployed analytics capabilities. In some examples, the method may further comprise updating a knowledge base with the network performance condition, the deployed network function, and the deployed analytics capabilities.

According to another aspect of the present invention, there is provided a computer program configured, when run on a computer, to carry out a method according to the first aspect of the present invention. The computer may in some example comprise a cloud computing data centre.

According to another aspect of the present invention, there is provided a computer program product comprising computer readable material, the computer readable material having stored thereon a computer product according to the preceding aspect of the present invention.

According to another aspect of the present invention, there is provided a network element for managing a communication network, the network element comprising a processor and a memory, the memory containing instructions executable by the processor such that the processor is operable to carry out a method according to the first aspect of the present invention. According to another aspect of the present invention, there is provided a network element for managing a communication network, the network element comprising a discovery unit configured to discover network functions deployed in the network, a performance unit configured to obtain a measure of network performance, and an evaluation unit configured to evaluate the measure of network performance to identify a network performance condition. The network element further comprises a function unit configured to identify a network function to address the identified network performance condition, and a deployment unit configured to deploy the identified network function.

Brief description of the drawings

For a better understanding of the present invention, and to show more clearly how it may be carried into effect, reference will now be made, by way of example, to the following drawings in which: Figure 1 is an illustration of centralised, distributed and hybrid network functions;

Figure 2 is a flow chart illustrating steps in a method for managing a communication network; Figure 3 is a representation of an implementation of the method of Figure 2; Figure 4 is a flow chart illustrating steps in another method for managing a communication network;

Figure 5 is a chart illustrating sub steps in a step of the method of Figure 4

Figure 6 is a signalling flow diagram; Figure 7 is another signalling flow diagram; Figure 8 is another signalling flow diagram; Figure 9 is another signalling flow diagram;

Figure 10 is a representation of elements considered according to examples of the method of Figure 4;

Figure 1 1 is a block diagram illustrating a network element; and

Figure 12 is a block diagram illustrating another example of network element.

Detailed Description

Aspects of the present invention provide a method for managing a communication network in which network performance is evaluated, a network performance condition is identified and a network function is deployed to address the network performance condition.

Figure 2 is a flow chart illustrating process steps in a method for managing a communication network, which may be a physical network, a virtual network or may combine both physical and virtual elements. The method may run as a process for example in a virtual machine or Linux Container (LXC). Referring to Figure 2, in a first step 1 10, the method comprises discovering network functions deployed in the network. This may comprise querying appropriate network elements to discover what network functions are already deployed in the network. Discovered functions may include centralised, distributed and/or hybrid network functions and may include functions from more than one vendor or involving more than one underlying technology. The deployed network functions may be SON, SDF or ANF and may or may not be virtualised. In a second step 120, the method comprises obtaining a measure of network performance. This may be obtained for example via querying a suitable network element or may be received as a report. The network performance measure may include multiple network performance measures indicating network performance levels for one or more aspects of network performance. In step 130, the method 100 comprises evaluating the obtained measure of network performance to identify a network performance condition. This may comprise comparing the network performance measure to a threshold for acceptable performance, and identifying a network performance condition when a network performance measure is identified as failing to meet the corresponding threshold. In some examples, a combined threshold for a series of related network performance measures may be defined, and a network performance condition may be identified when the combined threshold is not met. The threshold or thresholds may be predefined by a network operator in accordance with operator priorities for the network. For example, different network operators may prioritise quality of service, capacity, coverage etc. Each network operator may set individual thresholds for different performance measures in line with the operator priorities for service provision in the network. Having identified a network performance condition in step 130, the method 100 then comprises identifying a network function to address the identified network performance condition in step 140. This may comprise checking what network functions were discovered as deployed in step 1 10 and then checking which of the deployed functions relate to the identified network performance condition. Identifying a network function may also comprise checking what network functions are available for deployment, and selecting, in light of the currently deployed network functions, a network function operable to improve the identified network performance measure or measures corresponding to the identified network performance condition. By checking upon already deployed network functions, the method may distinguish between a situation in which no network function has yet been deployed to address a network performance issue, and a situation in which a network function has already been deployed but is perhaps underperforming, or is overwhelmed by current network conditions.

After identifying a network function in step 140, the method 100 comprises deploying the identified network function in step 160. The network function may be deployed in a centralised, distributed or hybrid manner as appropriate to the identified network condition. For example, if the network condition comprises an imbalance between a few overloaded cells and neighbouring under-utilised cells, a distributed load balancing function may be deployed in the affected cells. Other examples appropriate to different network performance conditions may be envisaged.

Figure 3 illustrates an example of the method 100 in operation in an NFV framework. Referring to Figure 2, an NFV Infrastructure 10 provides an execution environment for VNFs 4. A virtual network element running an example of the method 100 for managing a communication network is illustrated at 6. The virtual network element receives measures of performance of the network and identifies a network performance condition in the form of a network performance measure that is failing to meet an operator defined threshold for acceptable performance. The virtual network element identifies a suitable network function to address this condition, and deploys the network function 8. The network function comprises a plurality of individual VNFs chained together as a VNF Forwarding Graph 12. These VNFs together provide an end to end service to address the network performance condition identified by the virtual network element.

In some examples of the present invention, the method may take additional elements into consideration in selection of a suitable network function, and may select and initiate other functions to support the selected network function. For example, in addition to discovering currently deployed network functions, the method may comprise discovering an execution environment of the communication network. The execution environment of a network may comprise the underlying hardware and/or software environment within which the network exists. The execution environment may for example be a hardware environment, or may be a software environment such as a cloud environment. Knowledge of the network execution environment may enable the method to select a network function suited to the environment, and may also enable the method to take into account resources available in the environment for the deployment of the selected network function. In further examples, the method may comprise discovering and deploying analytics capabilities to enable monitoring of the selected network function. The method may for example comprise discovering what analytics capabilities are already available and/or deployed in the network, and may include selecting an analytics capability suited to monitor the performance of the selected network function. The method may then comprise deploying the selected analytics capability to support monitoring of the selected network function when deployed. Finally, the method may select and monitor a satisfaction criterion for the selected network function. A satisfaction criterion for network performance after deployment of the selected network function may be defined or retrieved in line with operator priorities. Network performance following deployment of the selected network function may then be monitored against the satisfaction criterion, enabling the method to identify when the satisfaction criterion has been met and to manage the deployed network function accordingly.

Figure 4 is flow chart illustrating another example of a method 200 for managing a communication network. The method 200 illustrates one way in which the steps of the method 100 may be implemented and augmented to provide the above discussed and additional functionality. In the example illustrated in Figure 4, the communication network being managed is a virtual network and the method is running as a process in a virtual machine implementing a virtual network element to carry out the method. It will be appreciated that this example is merely for the purposes of illustration, and that other examples in which a physical or hybrid network is managed using an example of the methods of the present invention which is implemented other ways may be envisaged. Referring to Figure 4, in a first step 210, the virtual network element discovers network functions deployed in the network, an execution environment of the network and analytics capabilities of the network. As discussed above, the network functions deployed in the network may include centralised, distributed and hybrid functions originating from different vendors and employing different underlying technologies. In the present example, the network is a virtual network, and the execution environment of the network comprises a cloud environment. Discovering the execution environment thus comprises establishing the cloud computing resources available to the network. This may be achieved for example by querying appropriate data centre managers. The virtual network element then assembles the discovered information at step 212. In the case of the discovered network functions, this comprises assembling a Network Forwarding Graph (NFG), representing the currently deployed network functions and their interrelations. In the case of the discovered execution environment and analytics capabilities, this comprises assembling a record of the execution environment, including available resources, and a record of the deployed analytics capabilities. The NFG and environment and analytics records may be referred to and updated by the virtual network element in later method steps.

In step 220, the virtual network element obtains measures of network performance. The virtual network element may obtain a range of measures detailing various different aspects of performance. These measures may be received individually by the virtual network element or may be received in the form of a report, for example assembled by a network monitoring function and forwarded to the virtual network element. In step 230, the virtual network element evaluates the measures of network performance and identifies performance measures failing to meet predefined standards. Predefined standards may vary between operators according to operator priorities for service provision. Different operators may prioritise different aspects of service provision according to their business model. For example, for some operators, network coverage and reliability may be their key priorities, with a reduction in service quality or available bandwidth being an acceptable consequence of maintaining reliable coverage for all users. For other operators, a guaranteed quality of service for key users may be prioritised.

A network performance condition is identified by the virtual network element when one or more measures are identified as failing to meet predefined thresholds. In some examples, a network performance condition may be defined with reference to several performance measures, all related to a particular aspect of network performance. For example, mobility robustness may be represented by several different performance measures. A single one of those performance measures failing to meet a predefined threshold may be treated as an aberration, whereas a plurality of the performance measures all failing to meet their predefined thresholds may be identified as a mobility robustness performance condition.

Having identified a network performance condition, the virtual network element may then identify a network function in step 240 to address the network performance condition. Identifying a network function may comprise several sub steps, as illustrated in Figure 5. Referring to Figure 5, in a first sub step 240a, the virtual network element checks among the discovered network functions for currently deployed network functions relating to the identified network performance measure that is failing to meet its predefined threshold. This enables the virtual network element to establish what network functions are already deployed within the network to manage the relevant network conditions. In sub step 240b, the virtual network element checks for network functions that are available for deployment. This may involve checking a software catalogue for network functions that are available to the network. In sub step 240c, the virtual network function then selects from among the available network functions, a suitable network function to improve the identified network performance measure. Knowing what functions are already deployed in the network and relating to the identified network performance measure may assist the virtual network element with selecting the most appropriate network function for deployment. Referring again to Figure 4, having selected a network function, the virtual network element then checks a knowledge base in step 240d. The knowledge base may include details of previously identified network performance conditions, the network functions that were deployed to address those conditions and any monitoring data relating to the effect of the deployment on the relevant network performance measures. In this manner, the virtual network function may use historical data to verify the selected network function and perhaps identify additional network functions that have proved effective for similar network performance conditions in the past. The virtual network element may also discard a selected network function that has proved ineffective against the identified network performance condition in the past. After checking the sleeted network function, the virtual network element then takes the first step towards deploying the selected network function by downloading the appropriate software for the selected network function in step 260a.

In a parallel process flow to that concerned with selection of a network function, the virtual network element also identifies resources in the execution environment available for deployment of network functions in step 242. This identification of available resources may inform the selection of a network function which takes place in step 240. The virtual network function then reserves resources required for deployment of the selected network function in step 244.

The virtual network element also defines a satisfaction criterion for the selected network function in step 246. This criterion may be defined by the virtual network function itself or may be retrieved from a memory. The satisfaction criterion may be checked with a network administrator or a knowledge base and updated accordingly in step 248. The virtual network element also identifies analytics capabilities appropriate to the selected network function and discovered execution environment in step 250. The virtual network element may check against existing deployed analytics capabilities to determine whether capabilities suitable for the selected network function are already deployed. In step 252 the virtual network element checks against a knowledge base for previous performance of the identified analytics capabilities with the identified network function and then downloads the appropriate software for the identified analytics capabilities in step 254.

In step 260b, the virtual network function instantiates the downloaded software for the selected network function and analytics capabilities, and configures or reconfigures the necessary resources from the execution environment, completing the deployment of the selected network function and analytics capabilities. In step 262, the NFG and records for the execution environment and analytics capabilities are updated. The updated NFG reflects the newly deployed network function and analytics capabilities, the environment record reflects the new use and configuration of resources and the analytics record reflects the newly available analytics capabilities. In step 264, the virtual network element monitors network performance and in step 266, the virtual network element checks whether or not scaling is required in light of the monitored performance. Scaling may include scaling up or down of the deployed network function, in accordance with the evolving network performance measures. If scaling is required, the virtual network element returns to step 260b, reconfiguring environment resources to enable the appropriate scaling, and then updating the NFG and records accordingly. If scaling is not required, the virtual network element then checks whether or not the satisfaction criterion has been met. If the satisfaction criterion has not been met, then a repeat check as to the necessity for scaling is made and the virtual network element continues to monitor network performance and scale the deployed network function accordingly. If the satisfaction criterion has been met, then the software instantiated in step 260b is stopped in step 270. This may include hibernating the deployed function, decommissioning the function or un-deploying the function, for example releasing the resources used for deploying the function and reversing the system configurations made to enable deployment of the function. Once the instantiated software has been stopped, the NFG and environment and analytics records are updated in step 272 and a knowledge base is updated in step 274. The knowledge base may be the knowledge base consulted in checking steps 240d and 252, and may be updated with the identified network performance condition, the network function and analytics capabilities deployed, and the resulting network performance. Maintaining a record of this information enables the virtual network element to learn from past performances, incorporating insight from past function deployments to inform future deployment selections.

It will be appreciated that the virtual network element is capable of running in any execution environment and manages functions within the network regardless of their type, the nature of their deployment or their original vendor. In this manner, the methods of the present invention enables an integrated solution to the problem of overlapping network functions; one single management process discovers all deployed functions in the network, evaluates a need for further deployments and manages those deployments, selecting new functions for deployment from a catalogue of available functions on the basis of network performance.

As discussed above, the network function identified and deployed to address a network performance condition may be a SON function, an SDF function, an ANF function etc. In some examples of the invention, the network function may comprise a use case for a particular network architecture, as defined in appropriate telecommunications standards. Taking the example of SON network architecture, various use cases that might be identified and deployed are discussed below. Mobility Robustness (MRO) may be appropriate for example when network performance measures indicate that handovers between neighboring cells in the network are taking place too early, too late or with errors. The handovers may involve the same or different radio access technology. Load Balancing (LB) may be appropriate for example when the network performance measures indicate that a few cells are overloaded with remaining cells significantly underused.

Automatic Neighbor Relation (ANR) may be appropriate when the network performance measures indicate that sufficient notifications/measurement reports/detected sets are available from User Equipment devices (UEs) about detected neighboring cells.

RACH Optimization may be appropriate if significant numbers of uplink re-attempts are indicated by the network performance measures. Automated configuration of Physical Cell Identity (PCI) may be appropriate if PCI conflicts are reported by the nodes.

Similarly, energy saving, capacity optimization, coverage optimization, core elements load balancing and transport capacity use may all also be appropriate according to the network performance measures received.

It will be appreciated that the above use cases are merely illustrative of the different types of use case that may be envisaged; different use cases for SON SDN, ANF and other types of network function may be envisaged.

Regardless of the use case or other network function identified as appropriate for the current network conditions, examples of the methods of the present invention may identify the required system configurations for the function, based on the discovered execution environment, identify the appropriate analytics capabilities according to available analytics software, and determine an appropriate satisfaction criterion to be met by the function based on the current network conditions. Once deployed, the selected function may be optimised or expanded according to performance and current or expected traffic, and may be shut down and its resources released once the satisfaction criterion has been met.

Referring again to Figure 4, it will be appreciated that examples of the present invention may involve consideration of four elements contributing to the overall implementation of the method. These include network functions, network execution environment, network analytics capabilities, and satisfaction criteria. Examples of the invention may be "aware" of all these elements, and may take each element into account in implementation of the method, and in some cases in managing others of the elements. Example signalling flow relating to each of these elements is separated out from the example process flow of Figure 4 and illustrated in the signalling diagrams of Figures 6 to 9.

Figure 6 illustrates signalling flow relating to network functions. In step 1 , the network element running an example method according to the invention discovers deployed network functions with support of a "Network Information" module (for example an OSS) and builds an NFG internally. If a Network Information module is not available, the network element could for example identify nearby network elements by probing or by deep packet inspection. In step 2, the network element evaluates network information in the form of performance measures in light of the currently deployed network functions, and identifies whether any additional network function, such as a SON use case, should be introduced. In step 3, the network element checks a knowledge base, which may for example be an external expert system, to verify the network function or functions identified for deployment, and to check for any additional functions which may have been effective in the past for similar network circumstances. In step 4 appropriate software, including for example SON modules, dependent libraries, platform/operating system layers, and their version information, is obtained. In step 5, the NFG and internal records are updated with the deployed functions.

Figure 7 illustrates signalling flow relating to analytics capabilities. In step 1 , the network element discovers Analytics capability information from an Analytics Catalogue. In step 2, the network element identifies analytics capabilities appropriate to the selected network function from an Available Analytics module. This information is then validated in step 3 by checking for prior experience with the identified analytics capabilities and selected network function in a knowledge base. In step 4 the required analytics software and version information for the identified analytics capabilities is obtained. This may include modules, dependent libraries, platform/operating system layers etc. In step 5, the NFG and analytics capabilities records are updated with the appropriate information for the deployed capabilities.

Figure 8 illustrates signalling flow relating to satisfaction criteria. In step 1 the network element identifies a satisfaction criterion for the or each network function to be deployed. In step 2, a check is made with a knowledge base to evaluate previous success rates with the satisfaction criterion and to check for side effects. The satisfaction criterion may be updated in light of this check in step 3.

Figure 9 illustrates signalling flow relating to network execution environment. In steps 1 and 2, the network element retrieves execution environment information. In the case of a hardware environment, the information is retrieved from an OSS or similar node. In the case of a cloud environment, the information is retrieved from data centre managers. The information may include configuration information for hardware or cloud data systems. In step 3, the network element determines available configuration for deploying additional network functions from the hardware system modules for example using SNMP GET, or using appropriate cloud systems such as REST APIs. In step 4, the network element updates the appropriate configuration details in execution environment records.

The four elements which may be taken into account in the course of methods according to the present invention are illustrated in Figure 10, together with a summary of actions that may be conducted according to different examples of the present invention.

As discussed above, the methods of the present invention, as explained with reference to Figures 2 to 10, may be conducted in a network element, which may be a virtual or a physical network element. The method may be implemented on receipt of suitable computer readable instructions, which may be embodied within a computer program running on a network element. Figure 1 1 illustrates a first example of a network element which may execute the methods of the present invention, for example on receipt of suitable instructions from a computer program. Referring to Figure 1 1 , the network element 300 comprises a processor 301 and a memory 302. The memory 302 contains instructions executable by the processor 301 such that the network element 300 is operative to conduct the steps of the methods of Figures 2 or 4 and the signalling flow of Figures 6 to 9. The memory may also store measurements and processing data generated during the process.

Figure 12 illustrates functional units in another example of network element 400 which may execute the process of the present invention, for example according to computer readable instructions received from a computer program. It will be understood that the units illustrated in Figure 12 are functional units, and may be realised in any appropriate combination of hardware and/or software.

Referring to Figure 12, the network element comprises a discovery unit 410 configured to discover network functions deployed in the network, a performance unit 420 configured to obtain a measure of network performance, an evaluation unit 430 configured to evaluate the measure of network performance to identify a network performance condition, a function unit 440 configured to identify a network function to address the identified network performance condition, and a deployment unit 460 configured to deploy the identified network function. The discovery unit 410 may be configured to discover at least two network functions having different underlying technology or supplied by different vendors, to assemble a Network Forwarding Graph (NFG) expressing connectivity between the discovered network functions, and to update the NFG with the deployed network function. The evaluation unit 430 may be configured to identify a network performance condition by identifying a measure of network performance failing to meet a predefined standard for that measure of network performance.

The function unit 440 may be configured to identify a network function to address the identified network performance condition by checking for currently deployed network functions relating to the identified measure of network performance, checking for network functions available for deployment, and selecting, in light of the checked currently deployed network functions, a network function operable to improve the identified network performance measure from among the network functions available for deployment. The function unit 440 may be further configured to check for network functions which have previously been deployed to address the identified network condition.

The deployment unit 460 may be configured to download software corresponding to the identified network function and to instantiate the downloaded software. The performance unit 420 may be configured to monitor network performance following deployment of the identified network function and the deployment unit 460 may further be configured to scale the deployment of the network function according to the monitored network performance.

The discovery unit 410 may be further configured to discover an execution environment of the network, which may comprise obtaining execution environment configurations and capabilities. The network element 400 may further comprise an environment unit 442 configured to identify resources in the execution environment available for deployment of network functions and to reserve resources in the execution environment for deployment of the identified function. The environment unit 442 may be further configured to maintain a record of the discovered execution environment and to update the record of the discovered execution environment following deployment of the identified network function. The network element 400 may further comprise a criterion unit 446 configured to define a satisfaction criterion for the identified network function. The performance unit 420 may be configured to monitor network performance against the satisfaction criterion following deployment of the identified network function, and the deployment unit 460 may be configured to stop the deployed network function on attainment of the satisfaction criterion. The criterion unit 446 may be configured to check performance associated with previous instances of the defined satisfaction criterion and to update the defined satisfaction criterion according to the check. The discovery unit 410 may be further configured to discover analytics capabilities in the network, and the network element may further comprise an analytics unit 450 configured to identify analytics capabilities appropriate for the identified network function. The analytics unit 450 may be further configured to check performance associated with previous instances of the identified analytics capabilities deployed with the identified network function and the deployment unit 460 may be configured to deploy the identified analytics capabilities. The function unit 440 may be configured to update the NFG with the deployed analytics capabilities, and the analytics unit 450 may be configured to maintain a record of the discovered analytics capabilities and to update the record with the deployed analytics capabilities.

Embodiments of the present invention thus provide a method for managing a communication network that overcomes integration issues between centralised, distributed and hybrid functions and between overlapping functions from different vendors and with different underlying technologies. In a single management process, network functions of different types and different deployments are discovered and managed through network performance monitoring and identification and deployment of appropriate network functions. By managing functions through a single process, duplication of feature implementation in different parts of the network can be avoided. Software and integration costs may be reduced owing to the autonomous nature of the method and its capacity to manage different types of SON, SDN and ANF functions.

A wide range of network functions may be managed and deployed including for example Analytics, SON functions such as ANR, MRO, LB, CCO, CCLB etc., SDN functions for focused group of networks and Deep Packet Inspection. In addition, on demand supporting functions for tracing, debugging etc can be identified and activated. Upgrade and migration of autonomous functions, as well as introduction of new functions is comparatively easy, as such updates or introductions need only be added to a software catalogue available to the entity running the management method.

Sudden changes in network behaviors may be mitigated using examples of the methods of the present invention. Initial deployment of network functions may involve only basic functionality required for connectivity and service delivery, with additional functionality auto-deployed later on the basis of network conditions.

Finally, multivendor realisation of functions can be managed without manual tuning, through the distribution of appropriate adaptation layer software that is recognisable and configurable by the entity running examples of the methods of the present invention.

The methods of the present invention may be implemented in hardware, or as software modules running on one or more processors. The methods may also be carried out according to the instructions of a computer program, and the present invention also provides a computer readable medium having stored thereon a program for carrying out any of the methods described herein. A computer program embodying the invention may be stored on a computer-readable medium, or it could, for example, be in the form of a signal such as a downloadable data signal provided from an Internet website, or it could be in any other form.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design many alternative embodiments without departing from the scope of the appended claims. The word "comprising" does not exclude the presence of elements or steps other than those listed in a claim, "a" or "an" does not exclude a plurality, and a single processor or other unit may fulfil the functions of several units recited in the claims. Any reference signs in the claims shall not be construed so as to limit their scope.

Claims

1 . A method for managing a communication network, the communication network capable of supporting a plurality of network functions; the method comprising:

discovering network functions deployed in the network;

obtaining a measure of network performance;

evaluating the measure of network performance to identify a network performance condition;

identifying a network function to address the identified network performance condition; and

deploying the identified network function.

2. A method as claimed in claim 1 , wherein the identified network function comprises at least one of a Self Organising Network, SON, function, a Software Defined Network, SDN, function, or an Autonomous Network Function, ANF.

3. A method as claimed in claim 1 or 2, wherein the identified network function comprises a Virtualised Network Function, VNF.

4. A method as claimed in any one of the preceding claims, wherein discovering network functions deployed in the network comprises discovering at least two network functions having different underlying technology or supplied by different vendors.

5. A method as claimed in any one of the preceding claims, wherein discovering network functions deployed in the network further comprises assembling a Network

Forwarding Graph, NFG, expressing connectivity between the discovered network functions.

6. A method as claimed in claim 5, further comprising updating the NFG with the deployed network function.

7. A method as claimed in any one of the preceding claims, wherein identifying a network performance condition comprises identifying an obtained measure of network performance failing to meet a predefined standard for that measure of network performance.

8. A method as claimed in any one of the preceding claims, wherein identifying a network function to address the identified network performance condition comprises: checking for currently deployed network functions relating to the identified network condition;

checking for network functions available for deployment; and

selecting, in light of the checked currently deployed network functions, a network function operable to improve the identified network performance condition from among the network functions available for deployment.

9. A method as claimed in any one of the preceding claims, wherein identifying a network function to address the identified network performance condition further comprises checking for network functions which have previously been deployed to address the identified network performance condition.

10. A method as claimed in any one of the preceding claims, wherein deploying the identified network function comprises downloading software corresponding to the identified network function and instantiating the downloaded software.

1 1 . A method as claimed in any one of the preceding claims, further comprising monitoring network performance following deployment of the identified network function and scaling the deployment of the network function according to the monitored network performance.

12. A method as claimed in any one of the preceding claims, further comprising discovering an execution environment of the network.

13. A method as claimed in claim 12, wherein discovering an execution environment of the network comprises obtaining execution environment configurations and capabilities.

14. A method as claimed in any one of claims 12 to 13, further comprising identifying resources in the execution environment available for deployment of network functions.

15. A method as claimed in claim 14, further comprising reserving resources in the execution environment for deployment of the identified function.

16. A method as claimed in any one of claims 12 to 15, further comprising:

maintaining a record of the discovered execution environment; and

updating the record of the discovered execution environment following deployment of the identified network function.

17. A method as claimed in any one of the preceding claims, further comprising: defining a satisfaction criterion for the identified network function;

monitoring network performance against the satisfaction criterion following deployment of the identified network function; and

on attainment of the satisfaction criterion, stopping the deployed network function.

18. A method as claimed in claim 17, further comprising checking performance associated with previous instances of the defined satisfaction criterion and updating the defined satisfaction criterion according to the check.

19. A method as claimed in any one of the preceding claims, further comprising discovering analytics capabilities in the network.

20. A method as claimed in claim 19, further comprising identifying analytics capabilities for the identified network function.

21 . A method as claimed in claim 20, further comprising checking performance associated with previous instances of the identified analytics capabilities deployed with the identified network function.

22. A method as claimed in claim 20 or 21 , further comprising deploying the identified analytics capabilities.

23. A method as claimed in claim 22, wherein deploying the identified analytics capabilities comprises downloading software corresponding to the identified analytics capabilities and instantiating the downloaded software.

24. A method as claimed in claim 23, when dependent upon claim 5, further comprising updating the NFG with the deployed analytics capabilities.

25. A method as claimed in any one of claims 22 to 24, further comprising:

maintaining a record of the discovered analytics capabilities; and

updating the record with the deployed analytics capabilities.

26. A computer program configured, when run on a computer, to carry out a method according to any one of the preceding claims.

27. A computer program product comprising computer readable material, the computer readable material having stored thereon a computer product as claimed in claim 26.

28. A network element for managing a communication network, the network element comprising a processor and a memory, the memory containing instructions executable by the processor such that the network element is operable to:

discover network functions deployed in the network;

obtain a measure of network performance;

evaluate the measure of network performance to identify a network performance condition;

identify a network function to address the identified network performance condition; and

deploy the identified network function.

29. A network element as claimed in claim 28, wherein the network element is further operable to identify a network function to address the identified network performance condition by:

checking for currently deployed network functions relating to the identified network condition;

checking for network functions available for deployment; and

30. A network element as claimed in claim 28 or 29, wherein the network element is further operable to:

define a satisfaction criterion for the identified network function; monitor network performance against the satisfaction criterion following deployment of the identified network function; and

on attainment of the satisfaction criterion, stop the deployed network function.

31 . A network element as claimed in any one of the claims 28 to 30, wherein the network element is further operable to:

discover analytics capabilities in the network;

identify analytics capabilities for the identified network function; and

deploy the identified analytics capabilities.

32. A network element for managing a communication network, the network element comprising:

a discovery unit configured to discover network functions deployed in the network;

a performance unit configured to obtain a measure of network performance; an evaluation unit configured to evaluate the measure of network performance to identify a network performance condition;

a function unit configured to identify a network function to address the identified network performance condition; and

a deployment unit configured to deploy the identified network function.

33. A network element as claimed in claim 32, wherein the function unit is configured to identify a network function to address the identified network performance condition by:

checking for currently deployed network functions relating to the identified measure of network performance;

checking for network functions available for deployment; and

selecting, in light of the checked currently deployed network functions, a network function operable to improve the identified network performance measure from among the network functions available for deployment.

34. A network element as claimed in claim 32 or 33, further comprising:

a criterion unit configured to define a satisfaction criterion for the identified network function; wherein the performance unit is configured to monitor network performance against the satisfaction criterion following deployment of the identified network function; and

wherein the deployment unit is configured to stop the deployed network function on attainment of the satisfaction criterion.

35. A network element as claimed in any one of claims 32 to 34, wherein the discovery unit is further configured to discover analytics capabilities in the network, the network element further comprising an analytics unit configured to identify analytics capabilities appropriate for the identified network function, and wherein the deployment unit is configured to deploy the identified analytics capabilities.