The present invention relates to mobile telecommunication systems. In particular, the present invention relates to a novel and improved method and system for reporting network parameters.
Different elements or entities of a network are continuously interchanging information between them. The way in which such exchange is performed will affect to the signalling reporting rate and could saturate the capacity of the interfaces.
For example, in coming multisystem or multilayer networks it is essential to utilise all the systems or layers in the most efficient way. For this reason, a new network element, the Common Radio Resource Management (CRRM), is being developed. The main functionality of the CRRM is to be able to direct the connections in the call set-ups and handovers to the optimum cell within optimum radio access technology (RAT) depending on the Quality of Service (QoS) requirements of the connection. The algorithms of the CRRM for the target cell selection and auto-tuning are based on the input parameters read from the respective interfaces. These parameters represent the status information of the different cells. Parameters can, for example, be the total load, RTLoad (RT, Real Time), SIR (Signal to Interference Ratio) and NRT Delay (NRT, Non Real-Time). Another example in which common measurements have to be reported is the Iur-interface between different Radio Network Controllers (RNC). This is actually one alternative for CRRM and requires event triggered reports.
The ETSI (European Telecommunication Standardisation Institute) TS 125 423 V4.1.0 (2001-06) describes several different methods how the cell measurements can be sent. The scope of the TS 125 423 V4.1.0 (2001-06) document is to specify the radio network layer signalling procedures of the control plane between Radio Network Controllers (RNC) in Universal Terrestrial Radio Access Network (UTRAN). One of the aspects considered at the standard is how the reporting of a dedicated measurement shall be performed between a Serving RNC (SRNC) and a Drift-RNC (DRNC). The different measurement reporting methods proposed at the standard can be grouped in two categories: the measured entity value no-dependent methods and the dependent ones. The first category includes the ‘on-demand’ and ‘periodic’ strategies, which are defined at the standard as:
On the other hand, the methods belonging to the second category, the measured entity value dependent ones, are labelled as ‘Event A’, ‘Event B’, . . . , until ‘Event F’. Among them, the ‘Event C’ and the ‘Event D’ strategies are described in more detail in the following.
If the Report Characteristics IE is set to ‘Event C’, the DRNC shall initiate the Dedicated Measurement Reporting procedure when the measured entity rises by an amount greater than the requested threshold (Mc) within the requested time (Tc). After having reported this type of event, the next C event reporting for the same measurement cannot be initiated before the rising time has elapsed since the previous event reporting.
The periodic strategy is typically the chosen one because of its simplicity. It is obvious that a low period would assure a good decision or subsequent action, but at the same time could saturate the processing capacity of the entity which receives the reports, like for example in a star configuration with a large number of entities connected to a unique one. On the other hand, a high period would reduce the number of messages to be processed by the receiving entity but the decision would be less reliable.
In one possible solution, event C, event D and periodic strategies can be combined.
To overcome such drawback, two alternatives can be used:
However, the simultaneous use of the periodic strategy in conjunction with the ‘Event C’ and ‘Event D’ strategies will produce a inefficient reporting rate. The reason is that they are uncoupled, that is, a periodic report will always be produced each period of time, without taking into account if reports by fast changes in the value of the measured entity have recently been sent (caused by event C or D). Therefore, it is sure that an unnecessary number of periodic reports will be sent, especially if the selected period is low.
Although any of the measurement reporting strategies suggested in the standards can also be used to exchange information between entities, none of them is so efficient as the proposed one in the present invention. The reason is that they are based on the absolute value of the input parameter or how fast the parameter to be reported is varying, that is, in the observed slope.
The present invention describes a flexible reporting method, system and data network entity for interchanging information between different entities of a data network. The data network comprises one or more data network parameters to be reported between different entities.
The present invention optimises and reduces the needed information exchange between network entities through the interfaces which connect them. The main property of the method of the present invention is its flexibility because it is able to change the reporting rate depending on the value of the input parameter in each instant of time. That is, it sends more reports or less reports depending on if the parameter to be reported is changing quickly or slowly, respectively. The main idea of the present invention is that reports of a data network parameter are based on the relative change of the data network parameter value with respect to the previously reported data network parameter value.
In a preferred embodiment of the present invention, three different control parameters are used to specify the performance of the method:
The ‘Threshold’ allows us to fix the desired sensibility of the method. If a low value is selected, the algorithm will inform us about light changes in the input data but the reporting rate will be higher. On the other hand, if a large value is allowed, only significant changes will be noticed, although in this case the number of needed measurement reports will be lower.
The ‘MinTBR’ represents the time interval after the last reported value in which no reports are allowed. That is, if an increase/decrease of the input data over/under the fixed threshold occurs between ‘0’ and ‘MinTBR’ seconds after the last reported value, the report is not sent.
The ‘MaxTBR’ represents the maximum time that the entity to be informed can be without knowing the current status of the parameter to be reported. That is, if no report is initiated between ‘MinTBR’ and ‘MaxTBR’, a mandatory one will be sent at MaxTBR. This control parameter does two tasks: not only it guarantees a minimum reporting rate between entities but also serves to check that everything is going well. For example, if no reports are received neither between ‘MinTBR’ and ‘MaxTBR’ nor after ‘MaxTBR’, it is sure that a technical problem has occurred (e.g. the connection may have been lost). Notice also that when the parameter ‘MinTBR’ is equal to ‘MaxTBR’, the event based reporting method works like a periodic reporting rate (no random reports are allowed).
In one embodiment of the present invention data network parameter values are filtered before reporting them. The filtering is done e.g. an Infinite Impulse Response (IIR) of Finite Impulse Response (FIR) filter.
In one embodiment of the present invention the data network is a wireless communication network. The reported wireless network parameter values are reported, e.g. with a radio resource controller of the wireless communication network. Correspondingly, the reported wireless network parameter values are received, e.g. with a radio resource management node of the wireless communication network.
In one embodiment of the present invention, the wireless communication network is the UTRAN, the IP-RAN, the GSM, the GPRS or the EDGE.
The present invention has several advantages over the prior-art solutions. With the new strategy the reporting rate is continuously and automatically optimised according to the current value of the parameter to be reported.
The key of the adaptive method described in the present invention to implement the event based reporting method is that it has a memory. The decision to trigger a measurement report is based on which is the difference, in each instant of time, between the value of the parameter to be reported and the last reported value. In doing so, the following advantages are obtained:
The accompanying drawings, which are included to provide a further understanding of the invention and constitute a part of this specification, illustrate embodiments of the invention and together with the description help to explain the principles of the invention. In the drawings:
a illustrates the evolution of the mean rate of call generation used in the simulations,
b illustrates the load of a cell during a day within the simulations,
Reference will now be made in detail to the embodiments of the present invention, examples of which are illustrated in the accompanying drawings.
The adaptive strategy RRC receives in
When the event based reporting method is applied to the filtered load using the input parameters as rules, the reported values can be sent to the CRRM.
The ‘Threshold’ allows us to fix which is the desired sensibility of the event based reporting method. If a low value is selected, the algorithm will inform us about light changes in the input data, but the reporting rate will be higher. On the other hand, if a large value is allowed, only significant changes will be noticed, although in this case the number of needed measurement reports will be lower.
The ‘MinTBR’ represents the time interval after the last reported value in which no reports are allowed. That is, if an increase/decrease of the input data over/under the fixed threshold occurs between ‘0’ and ‘MinTBR’ seconds after the last reported value, the report is not sent. This delay parameter has been defined in order to introduce a degree of freedom to control an excessive number of measurement reports between the network entities.
The ‘MaxTBR’ represents the maximum time that the entity to be informed can be without knowing the current status of the parameter to be reported. That is, if no report is initiated between ‘MinTBR’ and ‘MaxTBR’, a mandatory one will be sent at ‘MaxTBR’. This control parameter does two tasks: not only it guarantees a minimum reporting rate between entities, but also serves to check that everything is going well. For example, if no reports are received neither between ‘MinTBR’ and ‘MaxTBR’ nor after ‘MaxTBR’, it is sure that a technical problem has occurred (e.g. the connection may have been lost). Notice also that when the parameter ‘MinTBR’ is equal to ‘MaxTBR’, the event based reporting method works like a periodic reporting rate (no random reports are allowed).
Further, the RNC, IP-BTS and BSS comprise a timer T1 . . . Tn for each data network parameter to be reported, the timer T1 . . . Tn being started after the previous data parameter report, means for rejecting RM the report if a data network parameter to be reported changes more than the threshold value TH, and the timer T1 . . . Tn value is less than the minimum time interval MIN, means for reporting SM the current data network parameter value when a data network parameter to be reported changes more than the threshold value TH, and the value of the timer T1 . . . Tn exceeds the minimum time interval MIN and the value of the timer T1 . . . Tn is less than the maximum time interval MAX and means for restarting RE the timer T1 . . . Tn.
The RNC, IP-BTS and BSS comprise also means for reporting SM the current data network parameter value when no report has been sent after the previous data network parameter report and when the timer T1 . . . Tn reaches the maximum time interval MAX and means for filtering FM data network parameter values received with the first interface IF1 before reporting data network parameter values through the second interface IF2.
The aforementioned means are in a preferred embodiment arranged in a radio resource controller RRC. Furthermore, the aforementioned means are implemented in a preferred embodiment with software and/or hardware components.
a and 8b illustrate simulation scenarios during the simulations.
Under the aforementioned conditions, the evolution of the instantaneous cell load during a day (sampled at each second) that was obtained is plotted in
To quantify the performance of the different reporting methods two quality factors are defined: Total Number of Reports in a Day and Total Number of NonReported Alarms in a Day. The first one is a way to measure the reporting rate required by the different reporting strategies, while the second one informs about the number of times that the corresponding reporting method would have to have reported but it did not do it. In the simulation, it was counted the total number of times that the load, after a report and before the next one, rose/fell an amount greater/lower than the specific trigger threshold respect to the last reported value. For example, for the event based reporting method, if the ‘MinTBR’ control parameter is set to zero, there will never be non-reported alarms, while for the standard's combination (‘Event C+Event D+Periodic’) there are two sources for non-reported alarms:
The previously defined quality factors obtained for the event based reporting method and for the combined strategy, ‘Event C+Event D+Periodic’ representative, are shown in the
The explanation of the observed performance for the ‘Event C+Event D+Periodic’ strategy is the following. For a fixed and low value of ‘Tc’ and ‘Td’, if the period or distance between periodic reports is increased (
It should be noticed that when ‘Tc’ and ‘Tc’ are very large (greater than 30 seconds) the aforementioned performance is inverted, that is, the number of reports diminishes again, and the number of NonReported Alarms grows again. Such change is easily understood because, although an increment of ‘Tc’ and ‘Td’ have a clear positive effect over the number of NonReported Alarms produced by the slow changes, it has also a negative one over the fast changes that can take place just after a report, specifically during the next ‘Tc’ and ‘Td’ seconds.
On the other hand, the observed performance for the event based reporting method, disclosed in the present invention, is the following. The number of NonReported Alarms is drastically reduced to a very low value (0 and 4 alarms for ‘MinTBR’=0 at ‘MinTBR’=5 seconds), as illustrated in
The price to pay for such excellent performance is an increment in the Total Number of Reports. Nevertheless, compared with the situation in which the ‘Event C+Event D+Periodic’ strategy reached a minimum in the number of NonReported Alarms (5000 Alarms for ‘Tc’=‘Td’=30 seconds, as illustrated in
The present invention describes a new and powerful strategy for reporting through the respective interfaces the needed parameters. Although the new strategy would imply to change the philosophy used so far by including a new format of reporting of event triggered type, it has been shown that its performance is superior to the reporting methods suggested by the ETSI standard TS 125 423 V4.1.0 (2001-06). The main property of the event based reporting method is its flexibility and adaptability because it is able to change the reporting rate depending on the status of the network. That is, it sends more reports or less reports depending on if the status (the load for example) of the network is changing quickly or slowly, respectively.
It is obvious to a person skilled in the art that with the advancement of technology, the basic idea of the invention may be implemented in various ways. The invention and its embodiments are thus not limited to the examples described above, instead they may vary within the scope of the claims.
Number | Date | Country | Kind |
---|---|---|---|
20020916 | May 2002 | FI | national |
Number | Date | Country | |
---|---|---|---|
Parent | PCT/FI03/00363 | May 2003 | US |
Child | 10970277 | Oct 2004 | US |