CN101313524A

CN101313524A - Uplink burst performance monitoring method and device of point-to-multipoint access network

Info

Publication number: CN101313524A
Application number: CNA2007800002838A
Authority: CN
Inventors: 林华枫; 赵峻; 黄伟
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2006-04-10
Filing date: 2007-04-10
Publication date: 2008-11-26

Abstract

A method and apparatus for monitoring and controlling the upward outburst performance of the point-to-multipoint access network. The method includes the following steps of: determining the point of time at which the upward outburst data of the current terminal should arrive at the head end according to the time slot appointed to the current terminal in the access network; monitoring the point of time at which the upward outburst data of the current terminal actually arrives at the head end; calculating the bias value between the point of time at which the upward outburst data of the current terminal should arrive at the head end and the point of time at which the upward outburst data of the current terminal actually arrives at the head end; monitoring and controlling the upward outburst performance of the current terminal according to the bias value.; The method and apparatus can monitor and control the upward outburst performance status of the whole network global, and find the terminal with latent fault in time, thereby effectively improving the stability of the network.

Description

The uplink burst performance monitoring method and device of a kind of point-to-multipoint access network

Technical field

The present invention relates to access network field, more particularly to a kind of point-to-multipoint（P2MP, Point to Multi-Point) access network uplink burst performance monitoring method and device.

Background of invention

At present, in access network, it there is a kind of descending use broadcast mode and the up time division multiple acess access using shared bandwidth（TDMA, Time Division Multiple Access) mode network, such as various EPONs（PON, Passive Optical Network), including wideband passive optical fiber network（BPON, Broadband Passive Optical Network), Ethernet passive optical-fiber network（EPON, Ethernet Passive Optical Network) gigabit passive optical networks（GPON, Gigabit Passive Optical Network) etc..Above-mentioned various PONs are the structure of point-to-multipoint, and its system composition includes head end（HE, Head End) distribute net, terminal（TE, Terminal End) and subscriber terminal equipment (End User Device)₀Head end can be optical line terminal（OLT, Optical Line Terminal), terminal can be Guang Wang Network terminals (ONT, Optical Network Terminal), and subscriber terminal equipment can be PC（ PC, Personal Computer )₀Wherein, the data referred to as downlink data that head end is sent to terminal, conversely, the data that terminal head-end is sent are referred to as upstream data.

The topological structure of PON is referring to Fig. 1.It can be seen from figure 1 that head end is descending using time division multiplexing（TDM, Time Division Multiplex) broadcast mode by distribute net branch/coupler send the data to terminal and subscriber terminal equipment, terminal sends upstream data to head end using TDMA modes.Just blunt according to concrete condition, distribution net can be not limited to the one-level branch shown in Fig. 1. One of key technology of this network is the TDMA technologies of up shared bandwidth, and its operation principle is as shown in Figure 2.Under normal circumstances, head end compensates the time delay of each terminal so that the logical reach of each terminal is consistent first by measuring head end to the distance of each terminal；Then Dynamic Bandwidth Allocation is passed through（DBA, Dynamic Bandwidth Attribution) technology unification management, distribution upstream bandwidth；Each terminal sends uplink burst data according to the mandate of head end in defined time slot, so as to be prevented effectively from the conflict of uplink burst data.Numeral 1,2,3 in grid represents to belong to the data of terminal 1, terminal 2 or terminal 3 respectively, including the descending data and terminal uplink for being sent to terminal of head end are sent to the data of head end.As shown in Fig. 2 head end is descending to send data 1,3,1,2,3,1, by distributing net（It is not shown in FIG. 1）Incoming terminal, terminal 1 selects one's own data 1 from received downlink data, and other Shuo Ju Lost are abandoned；Similar, terminal 2 and terminal 3 select the one's own , Lost of data 2 and 3 and abandon other data respectively.When up, terminal then sends data according to the mandate time slot head-end of head end respectively.For example, terminal 1 sends two groups of data in two time slots of the mandate of head end；Then, terminal 2 sends two groups of data in two time slots of head end mandate；Then, terminal 3 sends three groups of data in three time slots of head end mandate.Because the data transmitted by terminal are sent in the time slot of head end mandate, so the uplink burst data order of terminal 1, terminal 2 and terminal 3 reaches head end, the conflict of upstream data will not be produced, so that head end can correctly recover and receive upstream data.

However, inventor is after the operation principle of this network is carefully studied, find to there may be certain risk in its running.For example, certain system exception occurs in some terminal, this exception includes two kinds of situations, and one is that terminal own hardware and/or software break down, and two be hackers operation terminal.Occur abnormal terminal no longer will send uplink burst data as normal terminal according to the mandate of head end, but do not send uplink burst data authorizing in time slot, or send uplink burst data in unauthorized time slot, or regardless of whether in mandate time slot, send uplink burst data.It is following to be referred to as rogue's terminal occurring abnormal terminal（Rogue TE ).When rogue's terminal, which is appeared in, does not send the situation of uplink burst data in mandate time slot, portion only can be wasted Divide uplink bandwidth resource, the normal operation of whole network is not influenceed, such case is not repeated below；When rogue's terminal is sending uplink burst data the unauthorized period, uplink burst data is would potentially result in clash on head end receiver, so that spectrum data signal is superimposed, head end be will be unable to be properly received the uplink burst data that each terminal is sent, and the up channel for then resulting in whole system is blocked.

Referring to Fig. 3, terminal 3 is rogue's terminal, is not only sending uplink burst data as defined in head end in time slot, and also send uplink burst data in unauthorized time slot.Then, terminal 3 is sent in unauthorized time slot data and data collision, the collision of the transmission of other terminals, cause head end can not normally receive in the uplink burst data of each terminal, figure "" represent due to the data-signal that can not correctly recover that uplink burst data conflict is caused.

It is typically to take certain remedial measure after the appearance of rogue's terminal in the prior art, the business of whole network has been have impact on yet with now rogue's terminal, so current solution is more passive, it is difficult to be effectively prevented from the influence of rogue's terminal-pair whole network.The content of the invention

In view of this, the embodiment of the present invention provides a kind of uplink burst performance monitoring method of point-to-multipoint access network, can be included with the performance condition of the uplink burst of monitor terminal, this method：

At the time of time slot according to present terminal in access network is assigned to determines that present terminal uplink burst data should reach head end；At the time of monitoring present terminal uplink burst data is actually reached head end；Present terminal uplink burst data is calculated at the time of should reaching head end and is actually reached the deviation at head end moment；

The uplink burst performance for being confused terminal is monitored according to the deviation.

The present invention also proposes a kind of uplink burst performance monitoring device of point-to-multipoint access network, the device is arranged in the headend of the point-to-multipoint access network, for determine each terminal uplink burst data should be to the moment, the uplink burst data for monitoring the terminal is actually reached the moment, meter Calculation should be monitored to the deviation between moment and the moment of being actually reached according to the deviation to the uplink burst performance of terminal.

It can see from above technical scheme, the uplink burst performance monitoring method for the point-to-multipoint access network that the present invention is provided has the advantages that：

1st, the deviation at the time of present invention should reach head end by the uplink burst data to each terminal and at the time of being actually reached head end is analyzed, and the uplink burst performance situation of whole network is grasped in time, decision-making foundation is provided for network manager；

Although the 2, can be monitored using modes such as the monitoring bit error rates to the situation of whole network, but when the bit error rate substantially increases, occurs the conflict of uplink burst data in network, and the present invention starts with from the due in of uplink burst data, it is simple and be easily achieved, it is accurate much sooner compared to using monitor modes such as the monitoring bit error rates；

3rd, the present invention is monitored to the due in of uplink burst data, the terminal for having incipient fault can be found in time, so as to effectively improve the stability of network.Brief Description Of Drawings

Fig. 1 is the network topology figure of point-to-multipoint access network；

Fig. 1 is the TDMA fundamental diagrams of the up shared bandwidth of point-to-multipoint access network；Fig. 3 is state diagram when occurring rogue's terminal in point-to-multipoint access network；

Fig. 4 is according to embodiments of the present invention one global terminal monitoring method flow chart；

Fig. 5 is according to embodiments of the present invention two automatic terminal monitoring and early warning method flow diagram；Fig. 6 is the blunt uplink burst performance monitoring method for early warning flow chart according to the embodiment of the present invention three of.

Implement the mode of the present invention

In order that the features and advantages of the present invention are more clearly understood, with reference to the accompanying drawings with reference to specific The invention will be further described for embodiment.

It was found from the introduction of background technology, take the method for corresponding measure more passive again after occurring rogue's terminal in access network, to this, the scheme that the embodiment of the present invention is proposed is that the uplink burst performance of the terminal of each in access network is monitored, this method is different from usually using by way of monitoring the bit error rate and being alerted, but start with the time of reaching head end from terminal uplink bursty data, at the time of monitoring terminal uplink burst data reaches head end and answer the deviation of due in, deviation is analyzed, realize the performance monitoring happened suddenly to terminal uplink and/or alarm, so as to realize the early warning to rogue's terminal.

Specifically, referring to the drawings and implementation process of the invention is described in detail in following examples.

Embodiment one

Referring to Fig. 4, Fig. 4 is the global terminal monitoring method flow chart of the embodiment of the present invention one.Assuming that there is 32 terminals in access network, and exemplified by being counted to 8000 monitored results of this 32 terminals, idiographic flow is He Bu Sudden are as follows：

Step 401, the global terminal monitoring of startup.

Starting global terminal monitoring can be by setting specific instruction, triggered by specifically instructing, when starting global terminal monitoring, preset monitoring condition, monitoring condition can be monitoring number of times, for example, it is 8000 times to preset the number of times being monitored to each terminal, that is, altogether by obtain at the time of the up Tu Fa Shuo Learn of 8000 X 32 should reach head end and the moment of being actually reached deviation.In addition, monitoring condition can also be the period being monitored to terminal, that is to say, that the deviation at all terminal uplink bursty datas arrival head end moment within the period of setting will be obtained.

Step 402, the uplink burst data for determining and preserving each terminal should be to the moment.

In the access network of point-to-multipoint, head end compensates the time delay of each terminal by ranging first so that the logical reach of each terminal is consistent, and then head end is the slot range that the up sending permission of terminal distribution, i.e. designated terminal send uplink burst data.The method of assigned timeslot scope has many kinds, example Such as dynamic bandwidth allocation algorithm, static bandwidth allocation algorithm.In theory, the slot range that terminal specified by head end sends uplink burst data refers to the starting point of slot range to a period of time between terminating point, at the time of the starting point of the slot range is that the head of uplink burst data that terminal is sent should reach head end, at the time of the terminating point of slot range is that the tail of the uplink burst data that terminal is sent should reach head end.Therefore, what head end can limit the uplink burst data that determines each terminal again according to the up transmission to each terminal distribution should be to the moment, and should be to being saved in respectively constantly in variable Tl Ti T32 by 32 uplink burst datas, wherein, what Ti represented i-th terminal uplink bursty data should be to the moment, and i is from 1 to 32.In practical application, i is the terminal serial number that is currently counted.It should be noted that, some mandate frame in of terminal is sent in head end, be likely to occur the situation that some or some terminals there is no mandate, then this or these there is no mandate the uplink burst data of terminal should all should be 0 to moment and actual arrival moment.

Step 403, the uplink burst data that monitors and preserve each terminal are actually reached the moment.Each terminal head-end sends uplink burst data, head end receives uplink burst data, at the time of the uplink burst data of 32 terminals of record is actually reached head end, and be stored in variable tl ..., ti ..., in t32, wherein, ti represents the moment that is actually reached of i-th terminal uplink bursty data, and i is from 1 to 32.In practical application, i is the terminal serial number that is currently counted.

Step 404, the uplink burst data for calculating and preserving each terminal should to moment and the moment of being actually reached deviation.

Computing terminal uplink burst data to moment Ti and should be actually reached moment ti deviation (Ti-ti), and deviation is stored in A ti, and i is from 1 to 32.In practical application, i is the terminal serial number that is currently counted.

These steps are exactly to obtain the deviation that terminal uplink bursty data reaches the head end moment above.Whether the obtained deviation number of step 405, judgement reaches monitoring number of times set in advance, if not up to, returning and performing step 402;Otherwise, step 406 is performed.

Deviation A ti exception can not illustrate that i-th of terminal has exception or have aggressiveness, need 2 will count to n times Ati, here, N is redefined for 8000, then, the step for is exactly to judge whether to have saved the Ati of 8000 times, if not up to 8000, then return and perform Bu Sudden 402, until collecting and preserving the Δ of 8000, if reaching 8000, the operation of next step can be carried out.

Step 406, calculate and export uplink burst performance desired value.

Uplink burst performance desired value can reflect the performance condition of the transmission uplink burst data of each terminal, the tranmitting data register shake of such as terminal is more severe, then the mean square deviation of its Ν times deviation will be than larger, therefore, the uplink burst performance desired value of each terminal obtained can be estimated to the performance that terminal sends uplink burst data, so as to monitor the performance of uplink burst.The computational methods of uplink burst performance desired value can be the statistical method of mathematics, i.e., the deviation to pre-determined number carries out the Mathematical Statistics Analysis with physical significance, such as the average, mean square deviation, root mean square that calculate deviation.Here, 8000 deviations to each terminal carry out statistical analysis, can calculate average, mean square deviation etc..Statistical analysis to 8000 deviations can reflect that terminal uplink bursty data reaches the quality condition of head end well, in other words, reflect the performance that terminal sends uplink burst data.Further, for the terminal that the poor such as mean square deviation of uplink burst performance is larger, system manager can be as needed, further trigger being separately monitored to the uplink burst performance of the terminal, other monitoring technology can also be aided in if necessary, such as monitoring luminous power, the bit error rate, or system manager can also directly take the terminal that corresponding measure processing has rogue's terminal to tend to.

The process and method that the individual terminal second-rate to uplink burst data is monitored are similar with the global terminal monitoring method described by the present embodiment, it is properly termed as pinpointing terminal monitoring method, unlike unique, global terminal monitoring is that all terminals in network are monitored, and it is the Ν monitoring and statistical analysis carried out for one or more terminals to pinpoint terminal monitoring.

In addition, before step described above is carried out, the condition of triggering alarm can also be preset, for example set the result threshold value of statistical analysis, after exported statistic analysis result is obtained, can also further by the statistic analysis result of resulting each terminal and result it is wealthy be worth into Row compares, and when the statistic analysis result for having terminal is more than result threshold value, triggers alarm event, notifies network manager.

The process that global monitoring is carried out to the uplink burst performance of each terminal in access network is described in detail in above step, from above step it can be seen that, at the time of the present embodiment collects uplink burst data arrival head end in time, by comparison terminal uplink burst data should to moment and the moment of being actually reached deviation, grasp the performance of the global uplink burst of whole network, decision-making foundation is provided for network manager, find the terminal for having incipient fault in time simultaneously, effectively improve the stability and attack tolerant of network.

Embodiment two

Automatic terminal monitoring and early warning method is described below, automatic terminal monitoring and early warning method is similar with global terminal monitoring method, be also to obtain the deviation that terminal uplink bursty data reaches the head end moment first, then acquired deviation is analyzed.Automatic terminal monitoring and early warning method monitors each bandwidth allocation of head end and the deviation with the uplink burst of each terminal of this bandwidth allocation correlation in real time, and the deviation for all terminals being collected into is compared with deviation threshold set in advance one by one every time, if wherein the uplink burst deviation of some or some terminals is more than or equal to deviation threshold, then automatic triggering is separately monitored to this or these terminal, if the statistical result being separately monitored of n times is still above result threshold value set in advance, it is then automatic to trigger alarm event, notify network manager.Automatic terminal monitoring and early warning method is that triggering is restarted in electricity or reset on head end.

Specifically, in the present embodiment, there are 32 terminals in access network, the times N that is separately monitored of fixed point terminal is set to 1000 times, and deviation threshold Tcon and result threshold value can be preset according to specific standard or agreement.The monitoring in real time of automatic pointing terminal early warning mechanism and the deviation of each bandwidth allocation correlation, once the deviation of one or more terminals is more than or equal to the wealthy value Tcon of deviation, then trigger and this or these terminal be separately monitored for 1000 times, and statistical analysis is carried out to the deviation of 1000 monitorings, if the result of statistical analysis is more than or equal to result threshold value, from trend network management module or Element management system（EMS, Element Management System) alarm is sent, it is on the contrary, then it is assumed that terminal is normal, terminates and this time pinpoints being separately monitored for terminal, is returned To bandwidth allocation, start automatic terminal monitoring next time, so as to set up automatic terminal monitoring early warning mechanism in the entire network.Referring to Fig. 5, Fig. 5 is according to embodiments of the present invention two automatic terminal monitoring and early warning method flow diagram, and operating process is as follows：

Electricity or reset start triggering and monitored automatically in step 501, head end.

Now, each terminal is not on by the stage of independent fixed point monitoring, specifically, in the present embodiment by setting mark flag [m]=0 to represent that m-th of terminal is not on the stage being separately monitored；If on the contrary, flag [m]=l, can represent that m-th of terminal is in the stage that coverlet is solely monitored.In addition, deviation threshold Tcon and result threshold value are initialized to default value set in advance, of course, it is possible to by ordering modification deviation threshold Tcon and the wealthy value of result.

Step 502, head end be each terminal distribution bandwidth, it is determined that and preserve each terminal uplink burst data should be to the moment.

By should being stored in the moment in variable T [m] for the uplink burst data of terminal, i.e. at the time of variable T [m] represents that the uplink burst data of m-th of terminal should reach head end.

Step 503, head end receive the uplink burst data of each terminal, and the uplink burst data for recording and preserving each terminal is actually reached the moment.

The uplink burst data of terminal is actually reached and is stored in constantly in variable t [m], be i.e. at the time of variable t [m] represents that the uplink burst data of m-th of terminal is actually reached head end.

Step 504, the uplink burst data for calculating and preserving each terminal should be stored in variables A t [m] to the deviation between moment and the moment of being actually reached, and by the deviation (T [m]-t [m]).

Step 505, whether present terminal is judged in being separately monitored the stage, if it is, performing step 506;Otherwise, step 510 is performed.

Because having indicated whether that terminal is in by setting flag marks in the present embodiment is separately monitored the stage, it need to only judge that the flag of present terminal is identified whether as 1 here, if it is, performing step 506;Otherwise, step 510 is performed.

Step 506, the deviation for preserving present terminal. Step 507, judge whether whether the number of saved deviation reaches the quantity threshold pre-set, if it is, performing step 508;Otherwise, step 512 is performed.

Here, number of times set in advance is 1000, in order to calculate the uplink burst performance desired value of present terminal, it is necessary to collect the deviation of 1000 times.

Step 508, termination are separately monitored to present terminal, and the deviation to collected pre-determined number carries out statistical analysis.

Deviation for have collected 1000 times, then be separately monitored for 0 by the flag set of ^ ^ present terminals to terminate to the terminal, then calculates the performance index value of uplink burst；The performance index value is the result that statistical analysis is carried out to deviation.

Whether step 509, the result for judging statistical analysis if it is, send alarm, then perform step 512 more than result threshold value;Otherwise, step 512 is directly performed.

If statistical result is equal to greatly result threshold value, then illustrate that present terminal is possible to exception occur, it is likely to become rogue's terminal, then, alarm is sent to system manager or EMS, to point out keeper or EMS further to monitor the terminal, other monitoring technology can also be aided in if necessary, such as monitoring luminous power, the bit error rate, or system manager can also directly take corresponding measure（For example, suspending the transmission of the terminal uplink data）Handle the terminal；If statistical result is less than result threshold value set in advance, it is believed that exception does not occur in present terminal, step 512 is performed.

Step 510, judge that the uplink burst data of present terminal reaches whether the deviation at head end moment is more than or equal to deviation threshold, if it is, performing Bu Sudden 511;Otherwise, step 512 is performed.

Step 511, present terminal is labeled as in being separately monitored the stage；

The deviation that the uplink burst data of present terminal reaches the head end moment is more than or equal to the wealthy value of deviation, then illustrate that the terminal has carried out the transmission of the uplink burst data of a subsidiary risk, it is probably that will appear from abnormal omen, therefore need to trigger individually to pinpoint monitoring to the terminal, to judge the size for abnormal possibility occur, here its corresponding flag is set to 1 to realize. Step 512, judge whether to have handled the deviation of all terminals, step 502 is performed if it is, returning;Otherwise, using next terminal as present terminal, return and perform step 505.

Illustrate the scheme of the embodiment of the present invention two by taking implementing in GPON networks as an example again below.Provided in GPON standards：Realized using Ranging-Time PLOAM messages to each terminal ONT balanced time delay or be propagation delay time（EqD; Equalizaiton-Delay fine setting); therefore the deviation for the uplink burst data arrival head end OLT that each terminal ONT is sent can be calculated; the present embodiment is using the obtained deviation of calculating, and the protection time slot in the OLT time burst expense methods of salary distribution recommended with standard（Guard Time) it is that trigger condition is that deviation threshold Tcon is judged and statistical analysis, so as to obtain the uplink burst performance of whole network, and realize the monitoring to the uplink burst performance of some specific terminal.The recommendation method of salary distribution of specific OLT time burst expenses is referring to table 1 (specifically refer to G.984.2 annex 1); it is 1244.16Mbit/s that upstream data rate is taken in the present embodiment, therefore the half of selection protection time slot is that 16bit is deviation threshold Tcon.Similarly, the concrete condition for netting Network can be occupied with ^, 3/4 grade of selection protection time slot is used as the wealthy value Tcon of deviation.

First, it is 16 to set deviation threshold to table 1, and sets the wealthy value of result of statistical analysis；Next, each ONT uplink burst datas reach OLT deviation in monitoring network in real time, and deviation is compared with deviation threshold 16, once certain ONT deviation is more than or equal to 16, then triggers and the ONT is monitored；Then, the EqD according to as defined in GPON standards is recalculated and is updated EqD Method, finely tune the EqD of the ONT;Time deviation to the ONT calculated in previous step uplink burst data is recorded and preserved；The step of being finely tuned more than repeating and preserve deviation several times, counts to result several times, then compared with the wealthy value of the result pre-set, if greater than or equal to threshold value, then send alarm to EMS；Otherwise, it is believed that ONT is normal, does not deal with.

Above step is described in detail carries out automatic pointing monitoring and early warning method to each terminal, from above step it can be seen that, the method that the present embodiment is provided can be monitored automatically in real time to the uplink burst performance of whole network, and the situation for grasping the network terminal in time for system manager provides very big help.

Embodiment three

The monitoring and early warning method of the embodiment of the present invention three is described below, the monitoring and early warning method of the present embodiment is similar with embodiment one and embodiment two, start with the time of being all and reach head end from terminal uplink bursty data, the deviation that terminal uplink bursty data reaches head end is monitored, embodiment one is to collect n times deviation, then carries out statistical analysis to collected deviation；Embodiment two is to collect n times deviation for the terminal that once deviation is more than deviation threshold set in advance, then carries out statistical analysis to collected deviation again；In the present embodiment, directly using deviation threshold, the terminal that continuous n times deviation is more than deviation threshold set in advance is alerted.

Still by taking implementing in GPON networks as an example, referring to Fig. 6, Fig. 6 is according to embodiments of the present invention three uplink burst performance monitoring method for early warning flow chart.In the present embodiment, just the blunt concrete condition according to protection time slot and/or network presets the wealthy value of deviation and times N, when implementing, can set a counter or variable, and the times N of deviation threshold is more than for counting continuous deviation.Comprise the following steps that：

What step 601- steps 603, head end determined the uplink burst data of terminal according to the slot range specified should be to the moment；The uplink burst data that receiving terminal is sent, obtain uplink burst data is actually reached the moment；Determine terminal uplink burst data should be to moment and actual arrival moment it is inclined Difference；

Here, terminal can be all terminals in access network or the particular terminal specified.Whether step 604, judgment bias are more than the wealthy value of deviation set in advance, if it is, performing step 605-607;Otherwise, counter or variable are resetted and reset, returned and perform step 601;Wherein, any deviation, which is less than deviation threshold, will all cause counter or variable to reset, and the continuous deviation of statistics of starting from scratch again is more than the number of times of deviation threshold.

Step 605-607, for count continuous deviation more than variance thresholds number of times counter or variable plus 1, and judge whether the value of counter or variable reaches count threshold set in advance, if, then alerted, reset simultaneously and reset counter or variable, then EqD adjustment is carried out to terminal, returns and perform step 601;Otherwise, EqD adjustment directly is carried out to terminal, returns and perform step 601.

Wherein, if reaching, the i.e. continuous n times deviation of pre-determined number is more than deviation threshold, then illustrate that EqD adjustment several times is invalid, the ONT is after EqD adjustment, its uplink burst can not be still returned in normal deviation range, it is necessary to alert, clearing counter or variable while alarm is sent, here, alarm can be sent to network management module or EMS when continuous n times deviation is more than deviation threshold.

Three above embodiment respectively describes the method analyzed in point-to-multipoint access network the deviation that terminal uplink bursty data reaches the head end moment, so that the monitoring to network uplink burst performance and/alarm are realized, and the early warning to rogue's terminal.Method provided by the present invention is equally applicable to the monitoring to bursty data performance in other networks.

Method to realize the various embodiments described above, is required to transform the headend in point-to-multipoint access network, increases uplink burst performance monitoring module.The module be used for determine each terminal uplink burst data should be to the moment, the uplink burst data for monitoring the terminal is actually reached the moment, calculating should be monitored to the deviation between moment and the moment of being actually reached according to the deviation to the uplink burst performance of terminal. Specifically, the uplink burst performance monitoring module comprises at least such as lower unit：Should be to moment computing unit, for according to each terminal distribution up sending permission determine each terminal uplink burst data should be to the moment, by it is identified should be to being stored in constantly in variable Ti, and the preservation should be sent to the variable at moment to deviation computing unit；Wherein, i represents the sequence number of terminal；

Moment monitoring unit is actually reached, the moment is actually reached as the uplink burst data at the time of for recording corresponding to head end reception uplink burst data, described be actually reached is stored in variable ti constantly, and variable ti is sent to deviation computing unit；Wherein, i represents the sequence number of terminal；

Deviation computing unit, for calculating the deviation between variable Ti and variable ti (Ti-ti), and deviation is stored in variables A ti, the variables A ti is sent to performance monitoring unit；

Performance monitoring unit, is estimated, and just take corresponding measure to terminal according to assessment result for just blunt according to variable Δ to the uplink burst performance of terminal.

According to the scheme of embodiment one, performance monitoring unit can include following subelement：Deviation number counts subelement, for counting correspondence particular terminal, number Ν and corresponding deviation that deviation occurs, and judge whether Ν is more than the wealthy value of deviation number set in advance, or timing statisticses reach the time pre-set noisy value, and subelement is counted if then sending the deviation recorded to performance indications；

Performance indications count subelement, for carrying out the Mathematical Statistics Analysis with physical significance to the deviation received, obtain uplink burst performance desired value, gained uplink burst performance desired value is sent to treatment measures subelement；

Treatment measures subelement, for according to uplink burst performance desired value and the current monitoring situation to terminal, corresponding processing being carried out to terminal and is operated.The processing operation potentially includes triggering and the terminal is separately monitored, and triggers other auxiliary monitorings, sends warning information, or stop terminal transmission uplink burst data.

According to embodiments of the present invention two scheme is, it is necessary in the uplink burst performance monitoring of embodiment one On the base of module unit, increase monitoring trigger element, the unit connects deviation computing unit and performance monitoring unit, for judging whether the deviation from deviation computing unit is more than the deviation threshold pre-set, if, then the deviation from same terminal is received by after to send to performance monitoring unit, the terminal is separately monitored by performance monitoring unit；Otherwise, then performance monitoring unit is made to keep idle condition.

The scheme of embodiment three is then that the performance monitoring unit in the uplink burst performance monitoring modular in the scheme of embodiment one is simplified, and the performance monitoring unit in the embodiment includes following subelement：Deviation count sub-element, for being counted to the continuous deviation more than deviation threshold, the then notifier processes measure subelement when count value is more than count threshold set in advance, and by counting value returns；If there is the deviation less than deviation threshold, counting value returns are laid equal stress on New count；

Treatment measures subelement, for after the notice of deviation count sub-element is received, to terminal

EqD is adjusted, and records the adjustment number of times to the terminal；If adjustment number of times exceedes the adjustment threshold value pre-set, alarm information is sent.

The foregoing is merely illustrative of the preferred embodiments of the present invention,, it is not intended to limit the invention, any modifications, equivalent substitutions and improvements made within the spirit and principles of the invention etc. should be included in the scope of the protection.

Claims

Claims

1st, a kind of uplink burst performance monitoring method of point-to-multipoint access network, it is characterised in that this method includes：At the time of reaching head end；At the time of monitoring present terminal uplink burst data is actually reached head end；Present terminal uplink burst data is calculated at the time of should reaching head end and the deviation at head end moment is actually reached；

The uplink burst performance of the terminal is monitored according to the deviation.

2nd, according to the method described in claim 1, it is characterised in that before the uplink burst performance that the terminal is monitored according to the deviation, further comprise：

The number of calculated deviation and deviation is counted, judges whether the number of the deviation reaches whether the quantity threshold pre-set, or timing statisticses have reached the time threshold pre-set, if then performing the subsequent step.

3rd, according to the method described in claim 1, it is characterised in that before the uplink burst performance that the terminal is monitored according to the deviation, further comprise：Judge whether the deviation reaches the deviation threshold pre-set, if then performing the subsequent step.

4th, method according to claim 2, it is characterised in that the uplink burst performance for monitoring the terminal according to the deviation includes：

According to the deviation counted, calculate and export uplink burst performance desired value.

5th, method according to claim 3, it is characterised in that the uplink burst performance for monitoring the terminal according to the deviation includes：

Whether the terminal is judged in the stage is separately monitored, if it is not, then determining whether whether the deviation currently calculated is more than the wealthy value of deviation pre-set, if being then to be separately monitored the stage by the terminal label；If then counting calculated deviation and deviation number；

] 6 Judge whether the deviation number that has counted reaches the quantity threshold pre-set, if then terminating to be separately monitored the terminal, and calculate according to the deviation counted the uplink burst performance desired value of the terminal.

6th, the method according to claim 4 or 5, it is characterised in that the calculating and after exporting uplink burst performance desired value, further comprises：Judge whether the uplink burst performance desired value is more than the result threshold value pre-set, if so, then taking one of following treatment measures or its any combination：

Send alarm information；

Further individually monitoring is carried out to the terminal；

The terminal is monitored using auxiliary monitoring means；

Time delay EqD adjustment is transmitted to the terminal

Stop the terminal head-end and send uplink burst data.

7th, according to the method described in claim 1, it is characterised in that the uplink burst performance for monitoring the terminal according to the deviation includes：

Judge whether calculated deviation is more than the wealthy value of deviation pre-set, if then being counted to the continuous deviation more than the wealthy value of deviation, and perform next step；Otherwise the step of being zeroed and return at the time of the calculating uplink burst data should reach head end is counted by described；

Judge whether the counting reaches the count threshold pre-set, if then carrying out EqD adjustment to the terminal and re-starting flow of the present invention；Otherwise the step of returning to the calculating uplink burst data at the time of should reaching head end.

8th, method according to claim 7, it is characterised in that described to carry out further comprising after EqD adjustment to the terminal：

Judge whether pre-determined number has been reached to the EqD adjustment number of times of the terminal, if then sending alarm information.

9th, the method according to claim 3,5 or 7, it is characterised in that the deviation is wealthy Protection time slot in the method for salary distribution is preset.

10th, the uplink burst performance monitoring device of a kind of point-to-multipoint access network, it is characterized in that, the device is arranged in the headend of the point-to-multipoint access network, for determine each terminal uplink burst data should be to the moment, the uplink burst data for monitoring the terminal is actually reached the moment, calculating should be just blunt that the uplink burst performance of terminal is monitored according to the deviation to the deviation between moment and the moment of being actually reached.

11st, device according to claim 10, it is characterised in that the device further comprises：Should be to moment computing unit, for according to each terminal distribution up sending permission determine each terminal uplink burst data should be to the moment, by it is identified should be to being stored in constantly in variable Ti, and the preservation should be sent to the variable at moment to deviation computing unit；

Moment monitoring unit is actually reached, the moment is actually reached as the uplink burst data at the time of for recording corresponding to head end reception uplink burst data, described be actually reached is stored in variable ti constantly, and variable ti is sent to deviation computing unit；

Deviation computing unit, for calculating the deviation between variable Ti and variable ti (Ti-ti), and deviation is stored in variable, the variable Δ is sent to performance monitoring unit；

Performance monitoring unit, is estimated for just blunt according to variables A ti to the uplink burst performance of terminal；I described above is terminal serial number.

12nd, according to claim 11 described device, it is characterised in that the performance monitoring unit further comprises：

Deviation number counts subelement, for counting correspondence particular terminal, number N and corresponding deviation that deviation occurs, and judge whether N is more than deviation number threshold value set in advance, or timing statisticses reach the time threshold pre-set, subelement is counted if then sending the deviation recorded to performance indications；

Performance indications count subelement, for carrying out the deviation received with physical significance Mathematical Statistics Analysis, obtains and exports uplink burst performance desired value.

13rd, device according to claim 12, it is characterised in that the device further comprises：Treatment measures subelement, for counting subelement gained uplink burst performance desired value and the current monitoring situation to terminal according to performance indications, corresponding processing operation is carried out to terminal.

14th, device according to claim 11, it is characterised in that the device further comprises：Monitor trigger element, the unit connects deviation computing unit and performance monitoring unit, for judging whether the deviation from deviation computing unit is more than the deviation threshold pre-set, if, then the deviation from same terminal is received by after to send to performance monitoring unit, the terminal is separately monitored by performance monitoring unit；Otherwise, then performance monitoring unit is made to keep idle condition.

15th, device according to claim 11, it is characterised in that the performance monitoring unit further comprises：

Deviation count sub-element, for being counted to the continuous deviation more than the wealthy value of deviation, the then notifier processes measure subelement when count value is more than count threshold set in advance, and by counting value returns；If there is the deviation less than the wealthy value of deviation, counting value returns are laid equal stress on New count；

Treatment measures subelement, for after the notice of deviation count sub-element is received, being adjusted to the EqD of terminal.

16th, device according to claim 15, it is characterized in that, the treatment measures subelement further comprises alerting subelement, for judging whether the number of times of the progress EqD adjustment to particular terminal reaches adjustment threshold value set in advance, if then sending alarm information.

17th, the device according to any one of claim 10 to 16, it is characterised in that the access network is passive optical-fiber network, the head end is optical line terminal, and the terminal is ONT Optical Network Terminal.