The instant invention relates to a novel mechanism using data mining in a mobile network to create or adapt self-organizing network (SON) rules and, more particularly, relates to a system and method for performing analytics on “Big Data” mined in a mobile network and/or obtained from other sources, in order to create SON rules and adapt parameters.
Self-Organizing Networks adjust the configuration of their elements automatically. They do this both at installation time and during ongoing network operation. Especially during network operation such configuration changes should reflect the current situation in the network: Load, number of users, user behavior, service level agreements etc.
The ever rising numbers of network elements makes it a pure necessity that SON does almost all that formerly had been pre-planned or adjusted by human intervention.
Currently, SON algorithms are based on pre-analysis of potential problems in the network based on radio engineering knowledge. This means that engineers understanding the specific Radio Access Technology (like LTE), also heavily based on understanding and experiences of preceding RATs (like 3G), anticipate certain effects/problems based on theoretical work and simulations. Example: The network engineer foresees that in the case of misaligned hand-over thresholds a hand-over (HO) may take place too early and describes a rule of the kind: IF number of too early HO is bigger than a specific number of events/minute, THEN increase HO threshold by 2 dB. In a second step, it is necessary to implement this rule in the network: For that, a counter of too early HO events is defined, and the specific value to trigger the increase of the threshold is defined as configurable. It is preferable that these definitions are done in a multi vendor capable way. Therefore they are captured in standard specifications, e.g. at 3GPP (a global standardization body for mobile networks).
This pre- (and ongoing) analysis can be complemented by data mining mechanisms and, thus, can be done in a much more elaborate way. This makes it possible to identify correlations of events which are hardly or not at all detectable by a human. The next step is then to formulate a rule based on the identified correlations and convert the detected rule into behavior of network elements in the network. For efficient and fast implementation of newly detected correlations and corresponding rules, it takes too long to define all of them in specifications. That process may take months, sometimes years.
It is important to remember that all events which are detected as a root cause for the network behavior are detected using data which was originally provided by the network. That means: Measurements or notifications for these events are defined and implemented. Otherwise the event would not be part of the data collection. Therefore each root cause event can be determined based on existing measurements and notifications.
The establishment of a new SON rule today is very time consuming. Currently there is no automatic mechanism to support the design of new SON functions and thus to bring new SON algorithms into the network. If new useful rules for SON algorithms are detected a long chain of work needs to be started, which involves heavy involvement of humans—contradicting the basic SON principles.
While some rule creation could be enabled also mining today's operations support systems (OSS) typical data set sizes, in particular, data mining using “Big Data” will produce abundant new knowledge about occurrence and interdependencies of events and network behavior. Such data mining is usually done for network performance reporting and to create new or improved network plans. It is not currently known to—more or less—directly feed into the network control, instructions on how to prevent or cater to unwanted network behavior or suboptimal network performance based on this data. Additionally, conversion of this data into SON rules acting on the real network will most likely not take place, if no automated mechanism will exist. Consequently, the major capabilities and targets of SON—saving costs and optimizing resource usage—cannot be exploited to their full possible extent.
What is needed is a system and method that automates the conversion of knowledge obtained from data mining system information into SON rules for a real network.
It is accordingly an object of this invention to provide a system and method for using data mined in a radio access network to generate SON rules and/or adapt SON parameters. In one particular embodiment of the invention, the analytics of “Big Data” mined in the network are used to generate and/or adapt SON rules and/or parameters, in realtime, in an automated way (i.e., substantially without human interaction in creating/updating the rules or parameters).
In one particular embodiment of the invention, a method is provided for creating or adapting a rule in a SON network (i.e., the SON network being a mobile network including its radio access network and its transport network), comprising the steps of: obtaining data mined from the SON network; performing analytics on the mined data; automatically creating a new rule or adapting an existing rule, based on the results of the analytics performed in the previous performing step; providing the new or adapted rule to a network entity that will execute the rule; and in accordance with the rule, performing an action to change a configuration in the SON network.
In one particular embodiment of a system for creating or adapting a rule in a SON network, there is provided a network management layer (which can be the network management layer of a “Big Data” system), an element management system layer, and at least one network element in the network element layer. In this embodiment, the system is configured to: obtain data mined from the SON network; perform analytics on the mined data; automatically create a new rule or adapt an existing rule, based on the results of the analytics performed in the previous performing step; provide the new or adapted rule to a network entity that will execute the rule; and execute the rule to change a configuration in the network.
Although the invention is illustrated and described herein as embodied in a system and method self-organizing network rule creation and parameter adaptation by data mining, it is nevertheless not intended to be limited to the details shown, since various modifications and structural changes may be made therein without departing from the spirit of the invention and within the scope and range of equivalents of the claims.
The construction of the invention, however, together with the additional objects and advantages thereof will be best understood from the following description of the specific embodiments when read in connection with the accompanying drawings.
The present invention is illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings, in which like reference numerals refer to similar elements and in which:
The present invention relates to a system and method of using data mining, and in particularly, the analytics performed on “Big Data”, generated from a mobile network—potentially combined with data from other sources—to create or adapt SON rules. In particular, Big Data mining permits correlations between network events, behavior and properties of users and network behavior to be found, as well as, configuration changes needed to improve network performance in these cases.
The invention allows the network operator to define events, conditions and corresponding actions in a way which reduces the need for human interaction and allows automation of SON rules based on these definitions. In one particular embodiment, the invention can be used in a radio access network operating in accordance with the specifications defined by the 3GPP SA5 Telecom Management Working Group (“Information Service”), as provided in effect at the date of filing of the present application. However, this is not meant to be limiting, as it will be appreciated that the system and method of the invention can be used with other radio access network protocols.
Referring now to
For example, other data sources 113 can provide data relating to the personal preferences of mobile users, e.g. an interest in soccer, to the “Big Data” system 110. In accordance with the principles of the present invention, analytics are performed on the mined data to identify occurrence and interdependencies of events and network behavior. Step 320. If data is obtained from other data sources 113, as well, this information can additionally be used to adjust/optimize the network. For example, if there is a big soccer match on a streaming channel and the system determines that many soccer fans are in the cell (based on the data from the other data sources 113), the “Big Data” system 110 can be used to predict that a higher than usual bandwidth will be required, and can adjust the network accordingly.
In one particular embodiment of the invention, a rule creation engine 112 of a “Big Data” system 110 uses the results of analytics (i.e., illustrated by the “data analysis” block 114) performed on “Big Data” and, optionally, on data from other sources 113 outside of the Big Data system 110, to create a new rule or adapt an existing rule affecting a network element 130a, 130b, 130c, operating in the mobile network (i.e., the SON Network). Step 330. The creation of this rule is automated, i.e., it is performed automatically by the rule creation engine 112 in response to the analytics generated from the mined data and/or other data, without human interaction. In the present embodiments, the rule creation engine 112 is provided in the “Big Data” system level or layer of the network.
Subsequently, the resulting rule (new or adapted) produced by the rule creation engine 112 is automatically converted by the system into a formal language identifying parameters associated with the rule, e.g., in a list of event-condition-action (ECA) policies or parameters. In the present invention, one parameter identifies a triggering event, so that it is possible for a rule translation engine to determine if an existing rule should be changed (identifier was used before) or if a new rule should be created (unused identifier); another parameter defines the conditions to be evaluated if the triggering event happens; and another parameter describes the action to be taken in case the triggering event happens. Step 340. The action could be, for example, a change in the configuration of one or several network elements in the SON network.
More particularly, in the present embodiments, the parameters (i.e., the parameters for triggering event, condition to be evaluated and action to be taken), are sent via an interface (i.e., Interface A and/or Interface B), to a so-called “rule translation engine” 122. In one particular embodiment of the invention, the rule translation engine 122 is embodied in software executed as part of the element management system 120a, 120b, as shown more particularly in
Referring back to
Once the change has been successfully performed, the performance of the change is reported via the usual event forwarding mechanisms (i.e., illustrated by “event forwarding” block 136) to one or more of the element management system 120a, 120b or 120c and/or the Big Data system 110, depending on the settings of the event forwarding 136.
One particular example of a protocol neutral specification for defining and implementing one particular embodiment of the invention will now be provided herebelow, wherein capital letters represent section numbers of specifications. It should be noted, however, that the below example is not meant to be limiting, as similar data for SON rules could be configured in different ways from the given example, without departing from the scope of the present invention.
L.M.N Information Object Class SONRule
L.M.N.1 Definition:
This IOC represents a SON rule.
L.M.N.2 Attributes:
L.M.N.3 Notifications:
For Creation, Deletion or attributeValueChange.
L.P.1 Information Attribute Definition and Legal Values
Q.S.1 Operation createSONrule
Q.S.1.1 Definition:
This operation allows the establishment of a new SONrule.
Q.S.1.2 Input Parameters:
Q.S.1.3 Output Parameters:
Q.S.1.4 Pre-Condition:
No such rule exists.
Q.S.1.5 Post-Condition:
SONrule is made known to the system, which prepares a possible activation. son RuleStatus is “suspended”. Output parameter result is set to success.
Q.S.1.6 Exceptions:
Rule Identifier is Already in Use:
Creation is rejected. Output parameter result is set to notUniqueIdentifier.
Q.S.2 changeSONruleStatus
Q.S.2.1 Definition:
This operation allows changing the status of a SONrule.
Q.S.2.2 Input Parameters:
Q.S.2.3 Output Parameters:
Q.S.2.4 Pre-Condition:
Identified SONrule exists.
Q.S.2.5 Post-Condition:
SONrule is active or suspended as requested via input parameter sonRuleStatus. sonRuleStatus in information object SONrule reflects this value. Output parameter result is set to success.
Q.S.2.6 Exceptions:
Rule Identifier is not in Use:
Change is rejected. Output parameter result is set to noSuchldentifier.
Q.S.3 changeSONruleStatus
Q.S.3.1 Definition:
This operation allows changing the status of a SONrule.
Q.S.3.2 Input Parameters:
Q.S.3.3 Output Parameters:
Q.S.3.4 Pre-Condition:
Identified SONrule exists.
Q.S.3.5 Post-Condition:
SONrule is changed as requested via input parameters triggeringCondition and triggeredAction. sonRuleStatus stays unchanged. Output parameter result is set to success.
Q.S.3.6 Exceptions:
Rule Identifier is not in Use:
Change is rejected. Output parameter result is set to noSuchldentifier.
Please note that, although described herein in connection with SON rules detected and/or derived for systems in which Big Data is analyzed, the invention is not intended to be limited only thereto. Rather, the present invention, wherein a system is informed of rules on how to behave in case of specific events, could also be used for SON rules that are detected for system not analyzed by Big Data mechanisms. This holds true for new systems that do not provide sufficient amounts of data for a Big Data mechanism, or where the first set of rules comes from predictions made by the system designers. In these cases, the parameters describing triggering events and triggered actions, and the rule identifier, can be assigned in a non-automatic way.
Referring now to
Additionally, it should be understood that the network devices or network elements and their functions described herein may be implemented by software, e.g. by a computer program product for a computer, or by hardware. In any case, for executing their respective functions, correspondingly used devices, such as the user equipment, access nodes, MME, S-GW, P-GW, CEM, location server, etc., include several means and components (not shown) which are required for control, processing and communication/signaling functionality. Such means may comprise, for example, a processor unit for executing instructions, programs and for processing data, memory means for storing instructions, programs and data, for serving as a work area of the processor and the like (e.g. ROM, RAM, EEPROM, and the like), input means for inputting data and instructions by software (e.g. USB memory stick, CD-ROM, EEPROM, and the like), user interface means for providing monitor and manipulation possibilities to a user (e.g. a screen, a keyboard, a mouse, a touchscreen and the like), interface means for establishing links and/or connections under the control of the processor unit (e.g. wired and wireless interface means, an antenna, etc.) and the like.
For the purpose of the present invention as described herein above, it should be noted that:
Many modifications and other embodiments of the inventions set forth herein will come to mind to one skilled in the art to which these inventions pertain having the benefit of the teachings presented in the foregoing descriptions and the associated drawings. Therefore, it is to be understood that the invention is not to be limited to the specific embodiments disclosed and that modifications and other embodiments are intended to be included within the scope of the appended claims. Moreover, although the foregoing descriptions and the associated drawings describe example embodiments in the context of certain example combinations of elements and/or functions, it should be appreciated that different combinations of elements and/or functions may be provided by alternative embodiments without departing from the scope of the appended claims. In this regard, for example, different combinations of elements and/or functions other than those explicitly described above are also contemplated as may be set forth in some of the appended claims. Although specific terms are employed herein, they are used in a generic and descriptive sense only and not for purposes of limitation.
It should be noted, that reference signs in the claims shall not be construed as limiting the scope of the claims. Additionally, although the invention is illustrated and described herein as embodied in a system and method for self-organizing network rule creation and parameter adaptation by data mining g, it is nevertheless not intended to be limited to only these details shown, as various modifications and structural changes may be made therein without departing from the spirit of the invention and within the scope and range of equivalents of the claims.
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/EP13/55639 | 3/19/2013 | WO | 00 |