CN104322017A

CN104322017A - Method, device, and system for identifying network packet loss type

Info

Publication number: CN104322017A
Application number: CN201280002149.2A
Authority: CN
Inventors: 蓝海青
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2012-12-27
Filing date: 2012-12-27
Publication date: 2015-01-28
Anticipated expiration: 2032-12-27
Also published as: WO2014101047A1; CN104322017B

Abstract

The present invention relates to the field of communications. Embodiments of the present invention provide a method, a device, and a system for identifying a network packet loss type, which can precisely and rapidly determine a network packet loss type. The method comprises: a first network element device sending data packets to a second network element device; the first network element device classifying the data packets sent to the second network element device in a first classification manner; the first network element device counting the number of data packets of various packet lengths in the data packets sent to the second network element device, to generate a first counting result; and the first network element device sending the first counting result to the second network element device, so that the second network element device determines, according to the first counting result and a second counting result, a network packet loss type between the first network element device and the second network element device. The embodiments of the present invention are applied in identification of a packet loss type of network data packets.

Description

Method, device, and system for identifying network packet loss type

The present invention relates to the communications field, more particularly to a kind of Wang Luo Lost Packet types recognition methods, equipment and system for a kind of Network Packet Loss kind identification method, equipment and systems technology field.

Background technology

Two equipment are by the primary demand that go-between transmission packet is that current information communicates, and the basic function of go-between is exactly to make full use of the bandwidth of oneself that the packet of two equipment is as much as possible and accurately and reliably transmit.But it is due to a variety of causes, packet can go out existing Lost bags when go-between is transmitted, so as to influence the communication before two equipment.Lost bags are main as caused by two kinds of reasons before mesh, and one kind is that middle bandwidth is not enough, and network congestion occurs and drawn Lost bags；Another is that go-between the wrong bag of transmission occurs, causes Wang network Zhu Dong Lost bags, and error code, product mistake bag occurs in such as microwave transmission.Wang Luo Lost bags can influence upper-layer service quality, it is necessary to which it is that the network bandwidth to be made full use of is increased economic efficiency to use certain method, also want Jiang Di Lost bag rates, ensure business experience.For the first reason Yin Qi Lost bags, the user equipment at two ends can release congestion by reduction of speed, for second of reason Yin Qi Lost bag, and the wrong bag of the anti-transmission of redundant transmission can be used by occurring end equipment.Premise ustomer premises access equipment must judge it is clearly Lost bags which kind of reason has been drawn, if it is determined that error, causes further degrading transmission on the contrary.If actual conditions are Wu Ma Lost bags, but misjudgement is Yong Sai Lost bags, and equipment attempts to release congestion using reduction of speed, but is actually still unable to Jiang Di Lost bag rates, causes the continuous reduction of speed of equipment, until packet transmission speed breaks off the base.In addition, if actual conditions are Yong Sai Lost bags, but misjudgement is transmission Cuo Bao Lost bags, and equipment two ends use redundant transmission Kang Lost bags, but aggravate network load on the contrary, so as to cause more serious congestion, it is possible to cause go-between to collapse.

But it is blunt it is many in the case of, the reason for terminal device can not coordinate acquisition Wang Luo Lost bags together with go-between, therefore main at present is still congestion Yin Qi Lost bags using manual evaluation in advance, or loading error occurring Yin Qi Lost bags, from the related anti-network congestion of configuration or the function of error resilience bag, inventor has found that existing technology can not accurately sentence the dynamic change of disconnected Lost bags reason and Lost bag reasons, therefore accurately can not quickly sentence circuit network Lost Packet types in above process. Content of the invention embodiments of the invention provide a kind of Wang Luo Lost Packet types recognition methods, equipment and system, can be accurately according to Wang Luo Lost bags reasons Ji the dynamic change of Lost bag reasons, accurately quickly judges Wang Luo Lost Packet types.To reach above-mentioned purpose, embodiments of the invention are adopted the following technical scheme that：

On the one hand there is provided a kind of Wang Luo Lost Packet type recognition methods, including：

First network element device sends packet to the second network element device；First network element device will be sent to the packet of second network element device is classified by the first mode classification, and first mode classification is to classify by the preset range that bag is long or bag is long of the packet；The data packet number of first network element device statistics transmission all kinds of bag length into the packet of second network element device, generates the first statistical result；

First network element device sends first statistical result to second network element device, so that second network element device judges the type of the Wang Luo Lost bags between first network element device and second network element device according to first statistical result and the second statistical result；Or,

First network element device receives second statistical result that second network element device is sent, and first network element device judges the type of the Wang Luo Lost bags between first network element device and second network element device according to first statistical result and second statistical result；

Wherein, second statistical result is after second network element device classifies the packet received by first mode classification, to the statistical result of the data packet number of all kinds of bag length in the packet received.

In the first possible implementation, with reference in a first aspect, first network element device judges that the type of the Wang Luo Lost bags between first network element device and second network element device includes according to first statistical result and second statistical result：Obtained at least one times in first statistical result neutralizes second statistical result The statistical result of the packet of first kind bag length and the packet of Equations of The Second Kind bag length, the average bag that the average bag length of the packet of the first kind bag length is more than the packet of the Equations of The Second Kind bag length is long；

The Shuo Ju Bao Lost bags rates of the first kind bag length and the Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length are calculated, the Shuo Ju Bao Lost bags rate is the ratio that first statistical result neutralizes the number difference of the packet of same class bag length and the number of the packet of same class bag length described in first statistical result in second statistical result；When the number of times that the Shuo Ju Bao Lost bags rate of the first kind bag length is more than the Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length is more than default number of times, judge current network Lost Packet types as Wu Ma Lost bags；When the number of times that the Shuo Ju Bao Lost bags rate of the first kind bag length is less than or equal to the Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length is more than default number of times, judge current network Lost Packet types as Yong Sai Lost bags.

In second of possible implementation, with reference in a first aspect, first network element device judges that the type of the Wang Luo Lost bags between first network element device and second network element device includes according to first statistical result and second statistical result：According to first mode classification in first statistical result neutralizes second statistical result, obtain the statistical result of the packet of the default different long classifications of bag, the Shuo Ju Bao Lost bag rates that each in the described default different long classifications of bag wraps long classification are calculated, the Shuo Ju Bao Lost bags rate is the ratio that first statistical result neutralizes the number of the number difference packet corresponding with same class bag length described in first statistical result of the packet of same class bag length in second statistical result；

The default different packets for wrapping long classification are subjected to ascending sort according to long classification is wrapped, the at least Shuo Ju Bao Lost bags rates including predetermined number arrange for ascending order in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag, then judge current network Lost Packet types as Wu Ma Lost bags；

If the number of the Shuo Ju Bao Lost bag rates arranged in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag according to ascending order is less than the predetermined number, judge current network Lost Packet types as Yong Sai Lost bags.

In the third possible implementation, with reference in a first aspect, the first or second Any one in possible implementation, methods described also includes：

When judging the Lost Packet types of current network as code Lost bags by mistake, the number of at least one long classification of bag described in the average Bao Changyu of the packet of at least one long classification of bag is subjected to linear fit according to bag Lost bags rate；When the slope of a curve that fitting is obtained is ascendant trend or the slope of a curve is the fixed value more than default slope, then judge that current network has Wu Ma Lost bags；When the corresponding Shuo Ju Bao Lost bag rates of all long classifications of bag are all higher than the wealthy value of default Shuo Ju Bao Lost bag rates in the obtained curve of fitting, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；The wherein described obtained curve of being fitted is using the long classification of bag as the abscissa in coordinate space, using the number according to wrapping Lost bags rate as ordinate.In the 4th kind of possible implementation, with reference in a first aspect, any one in the first or second of possible implementation, methods described includes：

When judging current network Lost Packet types as Wu Ma Lost bags, obtain second network element device currently receive at least one packet when extend to the average delay of the second network element device received data packet, when the time delay for currently receiving at least one packet is more than the average delay of the received data packet, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；

When the time delay for currently receiving at least one packet is less than or equal to the average delay of the received data packet, judge that current network has Wu Ma Lost bags.

Second aspect receives the packet that the first network element device is sent there is provided a kind of Wang Luo Lost Packet type recognition methods, including the second network element device；The packet received is classified according to the first mode classification, first mode classification is to classify by the preset range that bag is long or bag is long of the packet；

The data packet number of all kinds of bag length, generates the second statistical result in the packet received described in the second network element device statistics；

Second network element device sends second statistical result to first network element device, so that first network element device judges the type of the Wang Luo Lost bags between first network element device and second network element device according to first statistical result and the second statistical result； Or,

Second network element device receives first statistical result that first network element device is sent, and second network element device judges the type of the Wang Luo Lost bags between first network element device and second network element device according to first statistical result and second statistical result；

Wherein, first statistical result is that first network element device presses transmission to the packet of second network element device after first mode classification classification, to the statistical result of transmission data packet number of all kinds of bag length into the packet of second network element device.In the first possible implementation, with reference to second aspect, second network element device judges that the type of the Wang Luo Lost bags between first network element device and second network element device includes according to first statistical result and second statistical result：Obtain the statistical result of the packet of first kind bag length and the packet of Equations of The Second Kind bag length at least one times in first statistical result neutralizes second statistical result, the average bag that the average bag length of the packet of the first kind bag length is more than the packet of the Equations of The Second Kind bag length is long；

In second of possible implementation, with reference to second aspect, second network element device judges that the type of the Wang Luo Lost bags between first network element device and second network element device includes according to first statistical result and second statistical result：According to described first in first statistical result neutralizes second statistical result Mode classification, obtain the statistical result of the packet of the default different long classifications of bag, the Shuo Ju Bao Lost bag rates that each in the described default different long classifications of bag wraps long classification are calculated, the Shuo Ju Bao Lost bags rate is the ratio that first statistical result neutralizes the number of the number difference packet corresponding with same class bag length described in first statistical result of the packet of same class bag length in second statistical result；The default different packets for wrapping long classification are subjected to ascending sort according to long classification is wrapped, the at least Shuo Ju Bao Lost bags rates including predetermined number arrange for ascending order in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag, then judge current network Lost Packet types as Wu Ma Lost bags；

In the third possible implementation, with reference to second aspect, any one in the first or second of possible implementation, methods described also includes：

When judging the Lost Packet types of current network as code Lost bags by mistake, the number of at least one long classification of bag described in the average Bao Changyu of the packet of at least one long classification of bag is subjected to linear fit according to bag Lost bags rate；When the slope of a curve that fitting is obtained is ascendant trend or the slope of a curve is the fixed value more than default slope, then judge that current network has Wu Ma Lost bags；When the corresponding Shuo Ju Bao Lost bag rates of all long classifications of bag are all higher than the wealthy value of default Shuo Ju Bao Lost bag rates in the obtained curve of fitting, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；The wherein described obtained curve of being fitted is using the long classification of bag as the abscissa in coordinate space, using the number according to wrapping Lost bags rate as ordinate.

In the 4th kind of possible implementation, with reference in a first aspect, any one in the first or second of possible implementation, methods described also includes：

When judging current network Lost Packet types as Wu Ma Lost bags, obtain currently receive at least one packet when extend to the average delay of received data packet, when the time delay for currently receiving at least one packet is more than the average delay of the received data packet, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously； When the time delay for currently receiving at least one packet is less than or equal to the average delay of the received data packet, judge that current network has Wu Ma Lost bags.

The third aspect there is provided a kind of first network element device, including：Data transmission unit, for sending packet to the second network element device；

Data sorting unit, the packet for the data transmission unit to be sent to second network element device is classified by the first mode classification, and first mode classification is to classify by the preset range that bag is long or bag is long of the packet；

Statistic unit, the data packet number for counting all kinds of bag length in the sorted packet of the data sorting unit generates the first statistical result；

The data transmission unit, first statistical result for being additionally operable to forward the statistic unit is sent to second network element device, so that second network element device judges the type of the Wang Luo Lost bags between first network element device and second network element device according to first statistical result and the second statistical result；Or,

Lost bag recognition units, for receiving second statistical result that second network element device is sent, first statistical result and second statistical result that are forwarded according to the statistic unit judge the type of the Wang Luo Lost bags between first network element device and second network element device；

In the first possible implementation, with reference to the third aspect, Suo Shu Lost bag recognition units, including：

Sampling subelement, statistical result for obtaining the packet of first kind bag length and the packet of Equations of The Second Kind bag length in neutralizing second statistical result in first statistical result at least one times, the average bag that the average bag length of the packet of the first kind bag length is more than the packet of the Equations of The Second Kind bag length is long；

Subelement is handled, for calculating the Shuo Ju Bao Lost bags rates of the first kind bag length and the Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length, the Shuo Ju Bao Lost bags rate is the described first statistics knot Fruit neutralizes the ratio of the number difference of the packet of same class bag length and the number of the packet of same class bag length described in first statistical result in second statistical result；Differentiate subelement, when being more than default number of times for being more than the number of times of Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length when the Shuo Ju Bao Lost bags rate of the first kind bag length, judge current network Lost Packet types as Wu Ma Lost bags；When the number of times that the Shuo Ju Bao Lost bags rate of the first kind bag length is less than or equal to the Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length is more than default number of times, judge current network Lost Packet types as Yong Sai Lost bags.

In second of possible implementation, with reference to the third aspect, Suo Shu Lost bag recognition units include：

Sampling subelement, according to first mode classification in for neutralizing second statistical result in first statistical result, obtain the statistical result of the packet of the default different long classifications of bag, the Shuo Ju Bao Lost bag rates that each in the described default different long classifications of bag wraps long classification are calculated, the Shuo Ju Bao Lost bags rate is the ratio that first statistical result neutralizes the number of the number difference packet corresponding with same class bag length described in first statistical result of the packet of same class bag length in second statistical result；

Contrast subunit, for the default different packets for wrapping long classification to be carried out into ascending sort according to long classification is wrapped, the Shuo Ju Bao Lost bag Shuai Zhong Lost Packet types of the default different long classifications of bag are Wu Ma Lost bags；If the number of the Shuo Ju Bao Lost bag rates arranged in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag according to ascending order is less than the predetermined number, judge current network Lost Packet types as Yong Sai Lost bags.

In the third possible implementation, with reference to the third aspect, any one in the first or second of possible implementation, first network element device also includes：

Fitting unit, for when judging the Lost Packet types of current network as code Lost bags by mistake, the number of at least one long classification of bag described in the average Bao Changyu of the packet of at least one long classification of bag to be carried out into linear fit according to bag Lost bags rate；

Major punishment unit, for when the slope of a curve that fitting is obtained is ascendant trend or the slope of a curve is the fixed value more than default slope, then judging that current network has Wu Ma Lost bags；When the corresponding Shuo Ju Bao Lost bag rates of all long classifications of bag are all higher than the wealthy value of default Shuo Ju Bao Lost bag rates in the obtained curve of fitting, then judge that current network has Wu Ma Lost bags simultaneously And Yong Sai Lost bags；The wherein described obtained curve of being fitted is using the long classification of bag as the abscissa in coordinate space, using the number according to wrapping Lost bags rate as ordinate.

In the 4th kind of possible implementation, with reference to the third aspect, any one in the first or second of possible implementation, first network element device also includes：Time delay major punishment unit, for when judging current network Lost Packet types as Wu Ma Lost bags, obtain second network element device currently receive at least one packet when extend to the average delay of the second network element device received data packet, when the time delay for currently receiving at least one packet is more than the average delay of the received data packet, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；When the time delay for currently receiving at least one packet is less than or equal to the average delay of the received data packet, judge that current network has Wu Ma Lost bags.

Fourth aspect there is provided a kind of second network element device, including：Data receipt unit, for receiving the packet that the first network element device is sent；Data sorting unit, the packet for the data receipt unit to be received is classified according to the first mode classification, and first mode classification is to classify by the preset range that bag is long or bag is long of the packet；

Statistic unit, the data packet number for counting all kinds of bag length in the sorted packet received of the data sorting unit, generates the second statistical result；

Data transmission unit, second statistical result for the statistic unit to be forwarded is sent to first network element device, so that first network element device judges the type of the Wang Luo Lost bags between first network element device and second network element device according to first statistical result and the second statistical result；Or,

Lost bag recognition units, for receiving first statistical result that first network element device is sent, first statistical result and second statistical result that are forwarded according to the statistic unit judge the type of the Wang Luo Lost bags between first network element device and second network element device；

Wherein, first statistical result is that first network element device presses transmission to the packet of second network element device after first mode classification classification, is sent to described to institute State the statistical result of the data packet number of all kinds of bag length in the packet of the second network element device.In the first possible implementation, with reference to the third aspect, Suo Shu Lost bag recognition units include：Sampling subelement, statistical result for obtaining the packet of first kind bag length and the packet of Equations of The Second Kind bag length in neutralizing second statistical result in first statistical result at least one times, the average bag that the average bag length of the packet of the first kind bag length is more than the packet of the Equations of The Second Kind bag length is long；

Handle subelement, for calculating the Shuo Ju Bao Lost bags rates of the first kind bag length and the Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length, the Shuo Ju Bao Lost bags rate is the ratio that first statistical result neutralizes the number difference of the packet of same class bag length and the number of the packet of same class bag length described in first statistical result in second statistical result；Differentiate subelement, when being more than default number of times for being more than the number of times of Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length when the Shuo Ju Bao Lost bags rate of the first kind bag length, judge current network Lost Packet types as Wu Ma Lost bags；When the number of times that the Shuo Ju Bao Lost bags rate of the first kind bag length is less than or equal to the Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length is more than default number of times, judge current network Lost Packet types as Yong Sai Lost bags.

In second of possible implementation, with reference to the third aspect, Suo Shu Lost bag recognition units include：Sampling subelement, according to first mode classification in for neutralizing second statistical result in first statistical result, obtain the statistical result of the packet of the default different long classifications of bag, the Shuo Ju Bao Lost bag rates that each in the described default different long classifications of bag wraps long classification are calculated, the Shuo Ju Bao Lost bags rate is the ratio that first statistical result neutralizes the number of the number difference packet corresponding with same class bag length described in first statistical result of the packet of same class bag length in second statistical result；

Contrast subunit, for the default different packets for wrapping long classification to be carried out into ascending sort according to long classification is wrapped, the Shuo Ju Bao Lost bag Shuai Zhong Lost Packet types of the default different long classifications of bag are Wu Ma Lost bags；If the number of the Shuo Ju Bao Lost bag rates arranged in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag according to ascending order is less than the predetermined number, judge current network Lost Packet types as Yong Sai Lost bags. In the third possible implementation, with reference to fourth aspect, any one in the first or second of possible implementation, Suo Shu Lost bag recognition units include：

Major punishment unit, for when the slope of a curve that fitting is obtained is ascendant trend or the slope of a curve is the fixed value more than default slope, then judging that current network has Wu Ma Lost bags；When the corresponding Shuo Ju Bao Lost bag rates of all long classifications of bag are all higher than the wealthy value of default Shuo Ju Bao Lost bag rates in the obtained curve of fitting, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；The wherein described obtained curve of being fitted is using the long classification of bag as the abscissa in coordinate space, using the number according to wrapping Lost bags rate as ordinate.

In the 4th kind of possible implementation, with reference to fourth aspect, any one in the first or second of possible implementation, first network element device also includes：Time delay major punishment unit, for when judging current network Lost Packet types as Wu Ma Lost bags, obtain currently receive at least one packet when extend to the average delay of received data packet, when the time delay for currently receiving at least one packet is more than the average delay of the received data packet, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；When the time delay for currently receiving at least one packet is less than or equal to the average delay of the received data packet, judge that current network has Wu Ma Lost bags.5th aspect, a kind of network system is provided, including at least one first network element device and second network element device, wherein described first network element device and the second network element device are connected by switching network, first network element device is above-mentioned any first network element device, and second network element device is the second any of the above-described network element device.The recognition methods of embodiments of the invention Wang Luo Lost Packet types, equipment and system, by accurately can quickly judge Wang Luo Lost Packet types by the packet Lost bags rate that bag is long and difference bag is long of the packet transmitted between network element device.

Brief description of the drawings is in order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, the required accompanying drawing used in embodiment or description of the prior art will be briefly described below, it should be evident that drawings in the following description are only some embodiments of the present invention, for For those of ordinary skill in the art, on the premise of not paying creative work, other accompanying drawings can also be obtained according to these accompanying drawings.

A kind of Wang Luo Lost Packet type recognition methods schematic flow sheets that Fig. 1 provides for embodiments of the invention；Another Wang Luo Lost Packet type recognition methods schematic flow sheets that Fig. 2 provides for embodiments of the invention；

A kind of Wang Luo Lost Packet type recognition methods schematic flow sheets that Fig. 3 provides for another embodiment of the present invention；A kind of Wang Luo Lost Packet type recognition methods schematic flow sheets that Fig. 4 provides for another embodiment of the present invention；The linear fit schematic diagram of a kind of Shuo Ju Bao Lost bags rates and the long classification of bag that Fig. 5 provides for embodiments of the invention；A kind of first network element device structural representation that Fig. 6 provides for embodiments of the invention；Another first network element device structural representation that Fig. 7 provides for embodiments of the invention；

Another first network element device structural representation that Fig. 8 provides for embodiments of the invention；

Another first network element device structural representation that Fig. 9 provides for embodiments of the invention；

A kind of second network element device structural representation that Figure 10 provides for another embodiment of the present invention；

Another second network element device structural representation that Fig. 11 provides for another embodiment of the present invention；

Another first network element device structural representation that Figure 12 provides for another embodiment of the present invention；

Another first network element device structural representation that Figure 13 provides for another embodiment of the present invention；

Figure 14 shows for a kind of first network element device structure that another embodiment of the present invention is provided It is intended to；

A kind of second network element device structural representation that Figure 15 provides for one more embodiment of the present invention；

A kind of network architecture schematic diagram that Figure 16 provides for embodiments of the invention.The technical scheme in the embodiment of the present invention is clearly and completely described below in conjunction with the accompanying drawing in the embodiment of the present invention for embodiment, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.Based on the embodiment in the present invention, the every other embodiment that those of ordinary skill in the art are obtained under the premise of creative work is not made belongs to the scope of protection of the invention.In existing various communication system net structures, two network element devices are by the primary demand that go-between transmission packet is that current information communicates, and the basic function of go-between is exactly to make full use of the bandwidth of oneself that the packet of two equipment is as much as possible and accurately and reliably transmit.But it is due to a variety of causes, packet Hui Chu Xian Lost bags when go-between is transmitted, so as to influence the communication before two equipment, in order to accurately identify the type of Wang Luo Lost bags in order to the real-time control and regulation transmitted to current network, the present invention provides a kind of Wang Luo Lost Packet type recognition methods, the first network element device side of side is being sent as packet, shown in reference picture 1, including：

101st, the first network element device sends packet to the second network element device.

102nd, the first network element device will send to the packet of the second network element device and be classified by the first mode classification.

First mode classification is to classify by the preset range that bag is long or bag is long of packet, the long classification of bag when being classified here with specific bag long value certainly in the first mode classification is specific bag long value, and when being classified with wrapping long preset range, the long classification of bag can be used as using the average bag long value of the packet in each default long scope of bag, the specific concrete form to wrap long preset range classification can be according to the long sequence of bag according to arithmetic progression by packet, tolerance value is used as using default bit number, with a length of long initial value of bag of the maximum of packet or most parcel, such as with a length of long initial value of bag of the maximum bag of packet, tolerance value is set as n-bit, the packet for being then less than the long all bag length in n bit ranges of maximum bag is set as same class packet, reduce direction successively by each tolerance n-bit according to the bag length of packet accordingly In the range of packet be respectively set as a class, the average bag long value of packet in each margin of tolerance is now used as to the long classification of bag.

103rd, the first network element device statistics sends the data packet number of all kinds of bag length into the packet of the second network element device, generates the first statistical result.

104th, the first network element device sends the first statistical result to the second network element device.So that the second network element device judges the type of the Wang Luo Lost bags between the first network element device and the second network element device according to the first statistical result and the second statistical result.Or,

105th, the first network element device receives the second statistical result that the second network element device is sent, and the first network element device judges the type of the Wang Luo Lost bags between the first network element device and the second network element device according to the first statistical result and the second statistical result.Wherein, the second statistical result is after the second network element device classifies the packet received by the first mode classification, to the statistical result of the data packet number of all kinds of bag length in the packet that receives.

The present invention provides a kind of Wang Luo Lost Packet type recognition methods, and the second network element device side of side is being received as packet, shown in reference picture 2, including：

201st, the second network element device receives the packet that the first network element device is sent.

202nd, the second network element device is classified the packet received according to the first mode classification.Wherein, first mode classification is to classify by the preset range that bag is long or bag is long of packet, the long classification of bag when being classified here with specific bag long value certainly in the first mode classification is specific bag long value, and when being classified with wrapping long preset range, the long classification of bag can be used as using the average bag long value of the packet in each default long scope of bag, the specific concrete form to wrap long preset range classification can be according to the long sequence of bag according to arithmetic progression by packet, tolerance value is used as using default bit number, with a length of long initial value of bag of the maximum of packet or most parcel, such as with a length of long initial value of bag of the maximum bag of packet, tolerance value is set as n-bit, the packet for being then less than the long all bag length in n bit ranges of maximum bag is set as same class packet, the packet in each tolerance n bit ranges is respectively set as a class successively according to the long direction that reduces of bag of packet accordingly, now by data in each margin of tolerance The average bag long value of bag is used as the long classification of bag.

203rd, the second network element device counts the data packet number of all kinds of bag length in the packet received, generates the second statistical result.

204th, the second network element device sends the second statistical result to the first network element device.So that the first network element device judges the type of the Wang Luo Lost bags between the first network element device and the second network element device according to the first statistical result and the second statistical result.Or,

205th, the second network element device receives the first statistical result that the first network element device is sent, and the second network element device judges the type of the Wang Luo Lost bags between the first network element device and the second network element device according to the first statistical result and the second statistical result.Wherein, the first statistical result is that the first network element device presses transmission to the packet of the second network element device after the classification of the first mode classification, the statistical result to sending the data packet number of all kinds of bag length into the packet of the second network element device.The recognition methods of embodiments of the invention Wang Luo Lost Packet types, equipment and system, by accurately can quickly judge Wang Luo Lost Packet types by the packet Lost bags rate that bag is long and difference bag is long of the packet transmitted between network element device.

Specifically, being illustrated below in conjunction with specific embodiment.

Shown in reference picture 3, can be on the basis of the embodiment shown in Fig. 1 and Fig. 2, embodiments of the invention provide a kind of Wang Luo Lost Packet type recognition methods, the specific method differentiated there is provided network element device according to the first statistical result and the second statistical result to Lost Packet types Jian network element device, specifically includes following steps：

301st, the first network element device sends packet to the second network element device；

302nd, the first network element device will send to the packet of the second network element device and be classified by the first mode classification.

First mode classification is to classify by the preset range that bag is long or bag is long of packet.

303rd, the first network element device statistics sends the data packet number of all kinds of bag length into the packet of the second network element device, generates the first statistical result；

304th, the second network element device receives the packet that the first network element device is sent. 305th, the second network element device is classified the packet received according to the first mode classification.

First mode classification is to classify by the preset range that bag is long or bag is long of packet, the long classification of bag when being classified here with specific bag long value certainly in the first mode classification is specific bag long value, and when being classified with wrapping long preset range, the long classification of bag can be used as using the average bag long value of the packet in each default long scope of bag, the specific concrete form to wrap long preset range classification can be according to the long sequence of bag according to arithmetic progression by packet, tolerance value is used as using default bit number, with a length of long initial value of bag of the maximum of packet or most parcel, such as with a length of long initial value of bag of the maximum bag of packet, tolerance value is set as n-bit, the packet for being then less than the long all bag length in n bit ranges of maximum bag is set as same class packet, the packet in the range of each tolerance n-bit is respectively set as a class successively according to the long direction that reduces of bag of packet accordingly, now it regard the average bag long value of packet in each margin of tolerance as the long classification of bag.

306th, the second network element device counts the data packet number of all kinds of bag length in the packet received, generates the second statistical result.

307th, the second network element device sends the second statistical result to the first network element device, so that the first network element device judges the type of the Wang Luo Lost bags between the first network element device and the second network element device according to the first statistical result and the second statistical result.

308th, the first network element device receives the second statistical result that the second network element device is sent, obtain the statistical result of the packet of first kind bag length and the packet of Equations of The Second Kind bag length at least one times in the first statistical result neutralizes the second statistical result, the average bag that the average bag length of the packet of first kind bag length is more than the packet of Equations of The Second Kind bag length is long.

Here according to the specific mode classification employed in step 302, long value is wrapped in the average bag length if classifying with specific bag long value is correspondence classification, if classification is classified with multiple default long scopes of bag, the long average value of bag of a length of default long scope of bag of average bag herein.

Assuming that extract the statistical result obtained to the packet M and the long N of Equations of The Second Kind bag of the first kind bag length packet in the first statistical result and the second statistical result, and the bag indicated by M it is long when it is long more than bag indicated by N.

309th, the first network element device calculates the Shuo Ju Bao Lost bags rates and Equations of The Second Kind of first kind bag length Wrap long Shuo Ju Bao Lost bag rates.Shuo Ju Bao Lost bags rate is the ratio that the first statistical result neutralizes the number difference of the packet of same class bag length and the number of the packet of same class bag length in the first statistical result in the second statistical result.Specifically, obtaining the packet M Lost bag rates X of first kind bag length according to the packet M numbers of first kind bag length in the packet M numbers and the second statistical result of first kind bag length in the first statistical result；The packet N Lost bag rates X of Equations of The Second Kind bag length is obtained according to the packet N numbers of Equations of The Second Kind bag length in the packet N numbers and the second statistical result of Equations of The Second Kind bag length in the first statistical result.

310th, when the number of times that the Shuo Ju Bao Lost bags rate of first kind bag length is more than the Shuo Ju Bao Lost bag rates of Equations of The Second Kind bag length is more than default number of times, judge current network Lost Packet types as Wu Ma Lost bags；When the number of times that the Shuo Ju Bao Lost bags rate of first kind bag length is less than or equal to the Shuo Ju Bao Lost bag rates of Equations of The Second Kind bag length is more than default number of times, judge current network Lost Packet types as Yong Sai Lost bags.

In communication process between network element device, if during Yong Sai Lost bags, Shuo Ju Bao Lost bags rates are unrelated with the bag length of packet, the packet Lost bag rates of all kinds of bag length are in a balanced way, and in the case of error code Yin Qi Lost bags, the long packet Lost bag rates of Bao Changyue are bigger, in addition the statistical result of the packet of first kind bag length and the packet of Equations of The Second Kind bag length is obtained at least one times, when the only statistical result of the packet of first kind bag length of acquisition and the packet of Equations of The Second Kind bag length, it can only characterize in a comparison short period, network Lost Packet types judge current network Lost Packet types as Wu Ma Lost bags when X is more than Y, judge current network Lost Packet types as Yong Sai Lost bags if X is less than or equal to Y；In each stage it is dynamic change yet with the first statistical result in data transmission procedure and the second statistical result, therefore the statistical result of the packet of first kind bag length and the packet of Equations of The Second Kind bag length can also be obtained for K times, according to each sampled result, when the number of times that the long M of first kind bag Shuo Ju Bao Lost bags rate is more than the long N of Equations of The Second Kind bag Shuo Ju Bao Lost bag rates is more than default number of times, judge current network Lost Packet types as Wu Ma Lost bags；Otherwise, it is determined that current network Lost Packet types are Yong Sai Lost bags.

Certainly, step 307 can also be replaced with：First network element device sends the first statistical result to the second network element device, and such second network element device is receiving first statistical result Dotted portion is shown in method that afterwards can also be according to performed by the first statistical result and the second statistical result perform the first network element device in step 308 ~ 3 10, accompanying drawing 3, is specifically repeated no more.

The Wang Luo Lost Packet type recognition methods that embodiments of the invention are provided, the dynamic change of the Chang Ji Lost bag rates of bag for the packet that can be received according to the packet and receiving device for sending equipment transmission, accurately quickly judges Wang Luo Lost Packet types.

Shown in reference picture 4, can be on the basis of the embodiment shown in Fig. 1 and Fig. 2, embodiments of the invention provide a kind of Wang Luo Lost Packet type recognition methods, the specific method differentiated there is provided network element device according to the first statistical result and the second statistical result to Lost Packet types Jian network element device, specifically includes following steps：

401st, the first network element device sends packet to the second network element device；

402nd, the first network element device will send to the packet of the second network element device and be classified by the first mode classification.First mode classification is to classify by the preset range that bag is long or bag is long of packet, the long classification of bag when being classified here with specific bag long value certainly in the first mode classification is specific bag long value, and when being classified with wrapping long preset range, the long classification of bag can be used as using the average bag long value of the packet in each default long scope of bag, the specific concrete form to wrap long preset range classification can be according to the long sequence of bag according to arithmetic progression by packet, tolerance value is used as using default bit number, with a length of long initial value of bag of the maximum of packet or most parcel, such as with a length of long initial value of bag of the maximum bag of packet, tolerance value is set as n-bit, the packet for being then less than the long all bag length in n bit ranges of maximum bag is set as same class packet, the packet in the range of each tolerance n-bit is respectively set as a class successively according to the long direction that reduces of bag of packet accordingly, now it regard the average bag long value of packet in each margin of tolerance as the long classification of bag.

403rd, the first network element device statistics sends the data packet number of all kinds of bag length into the packet of the second network element device, generates the first statistical result；

404th, the second network element device receives the packet that the first network element device is sent.

405th, the second network element device is classified the packet received according to the first mode classification. First mode classification is to classify by the preset range that bag is long or bag is long of packet.

406th, the second network element device counts the data packet number of all kinds of bag length in the packet received, generates the second statistical result.

407th, the second network element device sends the second statistical result to the first network element device, so that the first network element device judges the type of the Wang Luo Lost bags between the first network element device and the second network element device according to the first statistical result and the second statistical result.

408th, the first network element device receives the second statistical result that the second network element device is sent, according to first mode classification in the first statistical result neutralizes the second statistical result, the statistical result of the packet of the default different long classifications of bag is obtained, the Shuo Ju Bao Lost bag rates that each in the default different long classifications of bag wraps long classification are calculated.

Shuo Ju Bao Lost bags rate is the ratio that the first statistical result neutralizes the number of the number difference packet corresponding with same class bag length in the first statistical result of the packet of same class bag length in the second statistical result；Here the statistical result of the packet of the K different long classifications of bag is can use, and the K different numbers for wrapping long classification are calculated according to bag Lost bag rates according to statistical result, corresponding K packet Lost bag rates are obtained.

409th, the default different long classifications of bag are carried out ascending sort by the first network element device according to bag is long, the at least Shuo Ju Bao Lost bags rates including predetermined number arrange for ascending order in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag, then judge current network Lost Packet types as Wu Ma Lost bags；If the number of the Shuo Ju Bao Lost bag rates arranged in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag according to ascending order is less than predetermined number, judge current network Lost Packet types as Yong Sai Lost bags.

Here the K long classification of bag in statistical result is ranked up, then the ordering scenario of the corresponding Shuo Ju Bao Lost bag rates of the long classification of bag of packet is detected, judge that current network Lost Packet types, as Wu Ma Lost bags, otherwise judge current network Lost Packet types as Yong Sai Lost bags when the Shuo Ju Bao Lost bag rates of at least predetermined number in K Lost bag rate are ascending order arrangement.Similarly in the communication process between network element device, if Lost bags rates are unrelated with the bag length of packet during Yong Sai Lost bags, the Shuo Ju Bao Lost bag rates of all kinds of bag length are in a balanced way, and in the case of error code Yin Qi Lost bags, the long Lost bags rates of Bao Changyue are bigger, therefore the number arranged according to the result of sequence according to ascending order exceedes predetermined number according to the number of bag Lost bag rates（Certainly predetermined number here is less than or equal to K's）The reason for more at most causing Wang Luo Lost bags before Dang afterwards is Wu Ma Lost The probability of bag is bigger.Certainly, step 407 can also be replaced with：First network element device sends the first statistical result to the second network element device, such second network element device can also be according to performed by the first statistical result and the second statistical result perform the first network element device in step 408 ~ 409 after receiving first statistical result method, dotted portion is shown in accompanying drawing 4, is specifically repeated no more.The Wang Luo Lost Packet type recognition methods that embodiments of the invention are provided, the dynamic change of the Chang Ji Lost bag rates of bag for the packet that can be received according to the packet and receiving device for sending equipment transmission, accurately quickly judges Wang Luo Lost Packet types.Further, optionally, after judging current network Lost Packet types as Wu Ma Lost bags according to the above embodiments, the program also includes：The Lost bags rate of the Bao Changyu of the packet of at least one long classification of bag at least one packet for wrapping long classification is carried out linear fit by al, the first network element device；

B l, when the obtained slope of a curve of fitting is ascendant trend or slope of a curve is the fixed value more than default slope, then judge that current network has Wu Ma Lost bags；C l, when the corresponding Shuo Ju Bao Lost bag rates of all long classifications of bag are all higher than the wealthy value of default Shuo Ju Bao Lost bag rates in the obtained curve of fitting, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously.Obtained curve is wherein fitted to wrap long classification as the abscissa in coordinate space, using Shuo Ju Bao Lost bags rates as ordinate.

Shown in specific reference picture 5, Shuo Ju Bao Lost bags rates are subjected to linear fit with wrapping the relation of long classification, approximating method is not limited, wherein when wrapping a length of 0, Shuo Ju Bao Lost bags rate is J.When should be now ascendant trend by the slope value for being fitted the curve 1 obtained, then judge that current network only exists Wu Ma Lost bags；Or the obtained curve of fitting is fixed slope K1, that is, be fitted to the straight line for being K1 for slope, when K1 is more than default slope K 2, then judge that current network only exists Wu Ma Lost bags；As shown in Fig. 5, then with i, j is respectively abscissa（Wrap long classification）And ordinate（Shuo Ju Bao Lost bag rates）Coordinate system in, coordinate points（I, j)=(0, J it is) matched curve 1 and the intersection point of axis of ordinates i=0, when the corresponding Shuo Ju Bao Lost bag rates of all long classifications of bag are all higher than the wealthy value of default Shuo Ju Bao Lost bag rates in the obtained curve 1 of fitting, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；When Linear fit obtain for curve when, because slope is ascendant trend, as long as therefore the J at coordinate points (0, J) place is more than the wealthy value of default Shuo Ju Bao Lost bags rates, then judge current network while there is Wu Ma Lost bags and Yong Sai Lost bags.

Certainly according to the above embodiments when judging that current network Lost Packet types are made as Wu Ma Lost bags as the second network element device, step a l, b l, c l can also be performed by the second network element device, and here is omitted.Optionally, after judging current network Lost Packet types as Wu Ma Lost bags according to the above embodiments, the program also includes：

A2, the second network element device obtain the second network element device receive at least one packet when extend to the average delay of the second network element device received data packet, when the time delay for receiving at least one packet is more than the average delay of received data packet, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；When the time delay for currently receiving at least one packet is less than or equal to the average delay of received data packet, judge that current network has Wu Ma Lost bags.The time delay and global average delay that second network element device in one period is now received at least one packet are compared, when the time delay that at least one packet is received in a period is more than global average delay, illustrate that network there is a possibility that to gather around plug Lost bags.

Certain step a2 is that the first network element device performs the time delay for now needing that correlation is obtained from the second network element device side, certainly according to the above embodiments when judging that current network Lost Packet types are made as Wu Ma Lost bags as the second network element device, step a2 also can directly be performed by the second network element device.Cause because Yong Sai Lost bags are mainly network condition factor, therefore if current network Lost bags factors are including Yong Sai Lost bags, then receiving terminal received data packet when postpone a meeting or conference increase, therefore, when the time delay of at least one packet of reception is more than the average delay of the received data packet, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；Otherwise judge that current network has Wu Ma Lost bags.Embodiments of the invention provide a kind of first network element device, first network element device is specifically as follows any node network element device in network architecture, such as switching equipment, Remote Radio Unit, base station, mobile terminal or base station side be used for data pack receiving and transmitting unit, module or entity in network architecture any equipment that can realize data pack receiving and transmitting, the present invention reality Apply in example and the concrete form of network element device is not particularly limited, can realize that any of the above-described Wang Luo Lost Packet type recognition methods that embodiments of the invention are provided is defined, shown in specific reference picture 6, the network element device includes：Data transmission unit 61, for sending packet to the second network element device；Data sorting unit 62, is classified for data transmission unit 61 to be sent to the packet of the second network element device by the first mode classification, and the first mode classification is to classify by the preset range that bag is long or bag is long of packet；Statistic unit 63, for the data packet number of all kinds of bag length in the sorted packet of statistics taxon 62, generates the first statistical result；

The data transmission unit 61, the first statistical result for being additionally operable to forward statistic unit 63 is sent to the second network element device, so that the second network element device judges the type of the Wang Luo Lost bags between the first network element device and the second network element device according to the first statistical result and the second statistical result；Or,

Lost bags recognition unit 64, for receiving the second statistical result that the second network element device is sent, the first statistical result and the second statistical result that are forwarded according to statistic unit 63 judge the type of the Wang Luo Lost bags between the first network element device and the second network element device；Wherein, the second statistical result is after the second network element device classifies the packet received by the first mode classification, to the statistical result of the data packet number of all kinds of bag length in the packet that receives.

Optionally, , Lost bags recognition unit shown in reference picture 7 64, including：

Sampling subelement 641, obtain the statistical result of the packet of first kind bag length and the packet of Equations of The Second Kind bag length in the second statistical result at least one times for being neutralized used in the first statistical result, the average bag that the average bag length of the packet of first kind bag length is more than the packet of Equations of The Second Kind bag length is long；

Handle subelement 642, for calculating the Shuo Ju Bao Lost bags rates of first kind bag length and the Shuo Ju Bao Lost bag rates of Equations of The Second Kind bag length, Shuo Ju Bao Lost bags rate is the ratio that first statistical result neutralizes the number difference of the packet of same class bag length and the number of the packet of same class bag length in the first statistical result in second statistical result； Differentiate subelement 643, when being more than default number of times for being more than the number of times of Shuo Ju Bao Lost bag rates of Equations of The Second Kind bag length when the Shuo Ju Bao Lost bags rate of first kind bag length, judge current network Lost Packet types as Wu Ma Lost bags；When the number of times that the Shuo Ju Bao Lost bags rate of first kind bag length is less than or equal to the Shuo Ju Bao Lost bag rates of Equations of The Second Kind bag length is more than default number of times, judge current network Lost Packet types as Yong Sai Lost bags.

Optionally, , Lost bags recognition unit shown in reference picture 8 64, including：

Sampling subelement 644, according to the first mode classification in for neutralizing the second statistical result in the first statistical result, obtain the statistical result of the packet of the default different long classifications of bag, the Shuo Ju Bao Lost bag rates that each in the default different long classifications of bag wraps long classification are calculated, Shuo Ju Bao Lost bags rate is the ratio that the first statistical result neutralizes the number of the number difference packet corresponding with same class bag length in the first statistical result of the packet of same class bag length in second statistical result；Contrast subunit 645, for the default different packets for wrapping long classification to be carried out into ascending sort according to long classification is wrapped, the at least Shuo Ju Bao Lost bags rates including predetermined number arrange for ascending order in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag, then judge current network Lost Packet types as Wu Ma Lost bags；If the number of the Shuo Ju Bao Lost bag rates arranged in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag according to ascending order is less than the predetermined number, judge current network Lost Packet types as Yong Sai Lost bags.

Further, shown in optional reference picture 9, when 6 Bao Kuo Lost bag recognition units 64 of the first network element device, the first network element device 6 also includes：Fitting unit 65, for when judging the Lost Packet types of current network as code Lost bags by mistake, at least one number for wrapping long classification to carry out linear fit according to bag Lost bags rate by the average Bao Changyu of the packet of at least one long classification of bag；

Major punishment unit 66, for when the slope of a curve that fitting is obtained is ascendant trend or slope of a curve is the fixed value more than default slope, then judging that current network has Wu Ma Lost bags；When the corresponding Shuo Ju Bao Lost bag rates of all long classifications of bag are all higher than the wealthy value of default Shuo Ju Bao Lost bag rates in the obtained curve of fitting, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；Obtained curve is wherein fitted to wrap long classification as the abscissa in coordinate space, using Shuo Ju Bao Lost bags rates as ordinate.

Optionally, shown in reference picture 9, when the Bao Kuo Lost bag recognition units of the first network element device 6 When 64, the first network element device 6 also includes：Time delay major punishment unit 67, for when judging current network Lost Packet types as Wu Ma Lost bags, obtain the second network element device currently receive at least one packet when extend to the average delay of the second network element device received data packet, when the time delay for currently receiving at least one packet is more than the average delay of received data packet, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；When the time delay for currently receiving at least one packet is less than or equal to the average delay of received data packet, judge that current network has Wu Ma Lost bags.The first network element device that embodiments of the invention are provided, the dynamic change of the Chang Ji Lost bag rates of bag for the packet that can be received according to the packet and receiving device of transmission, accurately quickly judges Wang Luo Lost Packet types.

Embodiments of the invention provide a kind of second network element device, second network element device is specifically as follows any node network element device in network architecture, such as switching equipment, Remote Radio Unit, base station, mobile terminal or base station side are used for the unit of data pack receiving and transmitting, any equipment that can realize data pack receiving and transmitting in network architecture such as module or entity, the concrete form of network element device is not particularly limited in an embodiment of the present invention, can realize that any of the above-described Wang Luo Lost Packet type recognition methods that embodiments of the invention are provided is defined, shown in specific reference picture 10, the network element device includes：Data receipt unit 71, for receiving the packet that the first network element device is sent；Data sorting unit 72, the packet for data receipt unit 71 to be received is classified according to the first mode classification, and the first mode classification is to classify by the preset range that bag is long or bag is long of packet；

Statistic unit 73, for the data packet number of all kinds of bag length in the sorted packet received of statistics taxon 72, generates the second statistical result；

Data transmission unit 74, the second statistical result for statistic unit 73 to be forwarded is sent to the first network element device, so that the first network element device judges the type of the Wang Luo Lost bags between the first network element device and the second network element device according to the first statistical result and the second statistical result；Or,

Lost bags recognition unit 75, for receiving the first statistical result that the first network element device is sent, the first statistical result and the second statistical result forwarded according to statistic unit 73 judges the The type of Wang Luo Lost bags between one network element device and the second network element device is wherein, first statistical result is that the first network element device presses transmission to the packet of the second network element device after the classification of the first mode classification, the statistical result to sending the data packet number of all kinds of bag length into the packet of the second network element device.

Optionally, with reference to shown in Figure 11, Lost bags recognition unit 75 includes：

Sampling subelement 75 1, statistical result for obtaining the packet of first kind bag length and the packet of Equations of The Second Kind bag length in neutralizing the second statistical result in the first statistical result at least one times, the average bag that the average bag length of the packet of first kind bag length is more than the packet of Equations of The Second Kind bag length is long；

Handle subelement 752, for calculating the Shuo Ju Bao Lost bags rates of first kind bag length and the Shuo Ju Bao Lost bag rates of Equations of The Second Kind bag length, Shuo Ju Bao Lost bags rate is the ratio that the first statistical result neutralizes the number difference of the packet of same class bag length and the number of the packet of same class bag length described in the first statistical result in the second statistical result；

Differentiate subelement 753, when being more than default number of times for being more than the number of times of Shuo Ju Bao Lost bag rates of Equations of The Second Kind bag length when the Shuo Ju Bao Lost bags rate of first kind bag length, judge current network Lost Packet types as Wu Ma Lost bags；When the number of times that the Shuo Ju Bao Lost bags rate of first kind bag length is less than or equal to the Shuo Ju Bao Lost bag rates of Equations of The Second Kind bag length is more than default number of times, judge current network Lost Packet types as Yong Sai Lost bags.

Optionally, with reference to shown in Figure 12, Lost bags recognition unit 75 includes：

Sampling subelement 754, according to the first mode classification in for neutralizing the second statistical result in the first statistical result, obtain the statistical result of the packet of the default different long classifications of bag, the Shuo Ju Bao Lost bag rates that each in the default different long classifications of bag wraps long classification are calculated, Shuo Ju Bao Lost bags rate is the ratio that the first statistical result neutralizes the number of the number difference packet corresponding with same class bag length described in the first statistical result of the packet of same class bag length in second statistical result；

Contrast subunit 755, for the default different packets for wrapping long classification to be carried out into ascending sort according to long classification is wrapped, the at least Shuo Ju Bao Lost bags rates including predetermined number arrange for ascending order in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag, then judge current network Lost Packet types as Wu Ma Lost bags；If the number of the Shuo Ju Bao Lost bag rates arranged in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag according to ascending order is less than the predetermined number, current network is judged Lost Packet types are Yong Sai Lost bags.

Further alternative, shown in reference picture 13, the second network element device also includes：Fitting unit 76, for when judging the Lost Packet types of current network as code Lost bags by mistake, at least one number for wrapping long classification to carry out linear fit according to bag Lost bags rate by the average Bao Changyu of the packet of at least one long classification of bag；

Major punishment unit 77, for when the slope of a curve that fitting is obtained is ascendant trend or slope of a curve is the fixed value more than default slope, then judging that current network has Wu Ma Lost bags；When the corresponding Shuo Ju Bao Lost bag rates of all long classifications of bag are all higher than the wealthy value of default Shuo Ju Bao Lost bag rates in the obtained curve of fitting, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；Obtained curve is wherein fitted to wrap long classification as the abscissa in coordinate space, using Shuo Ju Bao Lost bags rates as ordinate.

Optionally, shown in reference picture 13, the second network element device also includes：Also include：Time delay major punishment unit 78, for when judging current network Lost Packet types as Wu Ma Lost bags, obtain the second network element device currently receive at least one packet when extend to the average delay of the second network element device received data packet, when the time delay for currently receiving at least one packet is more than the average delay of received data packet, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；When the time delay for currently receiving at least one packet is less than or equal to the average delay of received data packet, judge that current network has Wu Ma Lost bags.The second network element device that embodiments of the invention are provided, accurately can quickly judge Wang Luo Lost Packet types according to the dynamic change for wrapping Chang Ji Lost bag rates for sending the equipment packet sent and the packet received.Embodiments of the invention provide a kind of first network element device, first network element device is specifically as follows any node network element device in network architecture, such as switching equipment, Remote Radio Unit, base station, mobile terminal or base station side are used for the unit of data pack receiving and transmitting, any equipment that can realize data pack receiving and transmitting in network architecture such as module or entity, the concrete form of network element device is not particularly limited in an embodiment of the present invention, can realize that any of the above-described Wang Luo Lost Packet type recognition methods that embodiments of the invention are provided is defined, shown in specific reference picture 14, the network element device includes：Including：At least one processor 81, memory 82, COM1 83 and bus 84, at least one processor 81, memory 82 and the communication interface 83 are connected by bus 84 and complete mutual communication. The bus 84 can be industry standard architecture（Industry Standard Architecture, referred to as ISA) bus, external equipment interconnection (Peripheral Component, referred to as PCI) bus or extended industry-standard architecture（Extended Industry Standard Architecture, referred to as EISA) bus etc..The bus 84 can be divided into address bus, data/address bus, controlling bus etc..For ease of representing, only represented in Figure 14 with a thick line, it is not intended that only one bus or a type of bus.Wherein：

Memory 82 is used to store executable program code, and the program code includes computer-managed instruction.Memory 82 may include high-speed RAM memory, it is also possible to also including nonvolatile memory（Non-volatile memory), for example, at least one magnetic disk storage.Stored in the storage device：Operating system, application program, the program code for realizing the embodiment of the present invention.The operating system is used for the processing function for controlling and realizing that the processing unit is performed.The application program includes program code, such as Word, email softwares.

Processor 81 is probably a central processing unit（Central Processing Unit, referred to as CPU), or specific integrated circuit（Application Specific Integrated Circuit, referred to as ASIC), or it is arranged to implement one or more integrated circuits of the embodiment of the present invention.

Communication interface 83, is mainly used in realizing the communication between the device in the present embodiment.The processor 81, for sending packet to the second network element device by COM1 83；

Processor 81 is sent to the packet of the second network element device by COM1 83 is classified by the first mode classification, and the first mode classification is to classify by the preset range that bag is long or bag is long of packet；

Processor 81, for the data packet number of all kinds of bag length in the packet after statistical classification, generates the first statistical result；

Processor 81, for being sent the first statistical result to the second network element device by COM1 83, so that the second network element device judges the type of the Wang Luo Lost bags between the first network element device and the second network element device according to the first statistical result and the second statistical result；Or, Processor 81, for receiving the second statistical result that the second network element device is sent by COM1 83, the type of the Wang Luo Lost bags between the first network element device and the second network element device is judged according to the first statistical result and the second statistical result；

Wherein, the second statistical result is after the second network element device classifies the packet received by the first mode classification, to the statistical result of the data packet number of all kinds of bag length in the packet that receives.Optionally,

Processor 81, the specific statistical result for obtaining the packet of first kind bag length and the packet of Equations of The Second Kind bag length at least one times in the first statistical result neutralizes the second statistical result, the average bag that the average bag length of the packet of first kind bag length is more than the packet of Equations of The Second Kind bag length is long；

Processor 81, for calculating the Shuo Ju Bao Lost bags rates of first kind bag length and the Shuo Ju Bao Lost bag rates of Equations of The Second Kind bag length, Shuo Ju Bao Lost bags rate is the ratio that the first statistical result neutralizes the number difference of the packet of same class bag length and the number of the packet of same class bag length described in the first statistical result in the second statistical result；

Processor 81, when being more than default number of times for being more than the number of times of Shuo Ju Bao Lost bag rates of Equations of The Second Kind bag length when the Shuo Ju Bao Lost bags rate of first kind bag length, judges current network Lost Packet types as Wu Ma Lost bags；When the number of times that the Shuo Ju Bao Lost bags rate of first kind bag length is less than or equal to the Shuo Ju Bao Lost bag rates of Equations of The Second Kind bag length is more than default number of times, judge current network Lost Packet types as Yong Sai Lost bags.

Optionally, processor 81, according to the first mode classification in for neutralizing the second statistical result in the first statistical result, obtain the statistical result of the packet of the default different long classifications of bag, the Shuo Ju Bao Lost bag rates that each in the default different long classifications of bag wraps long classification are calculated, Shuo Ju Bao Lost bags rate is the ratio that the first statistical result neutralizes the number of the number difference packet corresponding with same class bag length in the first statistical result of the packet of same class bag length in the second statistical result；Processor 81, for the default different packets for wrapping long classification to be carried out into ascending sort according to long classification is wrapped, the at least Shuo Ju Bao Lost bags rates including predetermined number arrange for ascending order in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag, then judge current network Lost Packet types as Wu Ma Lost bags；If being arranged in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag according to ascending order The number of the Shuo Ju Bao Lost bag rates of row is less than predetermined number, then judges current network Lost Packet types as Yong Sai Lost bags.

Further, optionally, processor 81, are additionally operable to when judging the Lost Packet types of current network as code Lost bags by mistake, and by the average Bao Changyu of the packet of at least one long classification of bag, at least one number for wrapping long classification carries out linear fit according to bag Lost bags rate；

Processor 81, for when the slope of a curve that fitting is obtained is ascendant trend or slope of a curve is the fixed value more than default slope, then judging that current network has Wu Ma Lost bags；When the corresponding Shuo Ju Bao Lost bag rates of all long classifications of bag are all higher than the wealthy value of default Shuo Ju Bao Lost bag rates in the obtained curve of fitting, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；Obtained curve is wherein fitted using the long classification of bag as the abscissa in coordinate space, it is optional by ordinate of the Shuo Ju Bao Lost bags rates,

Processor 81, for when judging current network Lost Packet types as Wu Ma Lost bags, by communication interface 83 obtain the second network element device currently receive at least one packet when extend to the average delay of the second network element device received data packet, when the time delay for currently receiving at least one packet is more than the average delay of received data packet, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；When the time delay for currently receiving at least one packet is less than or equal to the average delay of received data packet, judge that current network has Wu Ma Lost bags.The first network element device that embodiments of the invention are provided, the dynamic change of the Chang Ji Lost bag rates of bag for the packet that can be received according to the packet and receiving device of transmission, accurately quickly judges Wang Luo Lost Packet types.Embodiments of the invention provide a kind of second network element device, the network element device is specifically as follows any node network element device in network architecture, such as switching equipment, Remote Radio Unit, base station, mobile terminal or base station side are used for the unit of data pack receiving and transmitting, any equipment that can realize data pack receiving and transmitting in network architecture such as module or entity, the concrete form of network element device is not particularly limited in an embodiment of the present invention, can realize that any of the above-described Wang Luo Lost Packet type recognition methods that embodiments of the invention are provided is defined, specific reference picture

Shown in 15, the network element device includes：Including：At least one processor 91, memory 92, COM1 93 and bus 94, at least one processor 91, memory 92 and communication Interface 93 is connected by bus 94 and completes mutual communication.

The bus 94 can be industry standard architecture（Industry Standard Architecture, referred to as ISA) bus, external equipment interconnection (Peripheral Component, referred to as PCI) bus or extended industry-standard architecture（Extended Industry Standard Architecture, referred to as EISA) bus etc..The bus 94 can be divided into address bus, data/address bus, controlling bus etc..For ease of representing, only represented in Fig. 15 with a thick line, it is not intended that only one bus or a type of bus.Wherein：

Memory 92 is used to store executable program code, and the program code includes computer-managed instruction.Memory 92 may include high-speed RAM memory, it is also possible to also including nonvolatile memory（Non-volatile memory), for example, at least one magnetic disk storage.Stored in the storage device：Operating system, application program, the program code for realizing the embodiment of the present invention.The operating system is used for the processing function for controlling and realizing that the processing unit is performed.The application program includes program code, such as Word, email softwares.

Processor 91 is probably a central processing unit（Central Processing Unit, referred to as CPU), or specific integrated circuit（Application Specific Integrated Circuit, referred to as ASIC), or it is arranged to implement one or more integrated circuits of the embodiment of the present invention.

Communication interface 93, is mainly used in realizing the communication between the device in the present embodiment.Processor 91, for receiving the packet that the first network element device is sent by COM1 93；

Processor 91, for the packet received by COM1 93 to be classified according to the first mode classification, the first mode classification is to classify by the preset range that bag is long or bag is long of packet；

Processor 91, for the data packet number of all kinds of bag length in the sorted packet received, generates the second statistical result；

Processor 91, for the second statistical result to be sent to the first network element device by COM1 93, so that the first network element device judges the type of the Wang Luo Lost bags between the first network element device and the second network element device according to the first statistical result and the second statistical result； Or,

Processor 91, for receiving the first statistical result that the first network element device is sent by COM1 93, the type of the Wang Luo Lost bags between the first network element device and the second network element device is judged according to the first statistical result and the second statistical result；Wherein, the first statistical result is that the first network element device presses transmission to the packet of the second network element device after the classification of the first mode classification, the statistical result to sending the data packet number of all kinds of bag length into the packet of the second network element device.

Optionally, processor 91, statistical result specifically for obtaining the packet of first kind bag length and the packet of Equations of The Second Kind bag length in neutralizing the second statistical result in the first statistical result at least one times, the average bag that the average bag length of the packet of first kind bag length is more than the packet of the Equations of The Second Kind bag length is long；

Processor 91, for calculating the Shuo Ju Bao Lost bags rates of first kind bag length and the Shuo Ju Bao Lost bag rates of Equations of The Second Kind bag length, Shuo Ju Bao Lost bags rate is the ratio that the first statistical result neutralizes the number difference of the packet of same class bag length and the number of the packet of same class bag length in the first statistical result in the second statistical result；Processor 91, when being more than default number of times for being more than the number of times of Shuo Ju Bao Lost bag rates of Equations of The Second Kind bag length when the Shuo Ju Bao Lost bags rate of first kind bag length, judges current network Lost Packet types as Wu Ma Lost bags；When the number of times that the Shuo Ju Bao Lost bags rate of first kind bag length is less than or equal to the Shuo Ju Bao Lost bag rates of Equations of The Second Kind bag length is more than default number of times, judge current network Lost Packet types as Yong Sai Lost bags.

Optionally, processor 91, according to the first mode classification in specifically for neutralizing the second statistical result in the first statistical result, obtain the statistical result of the packet of the default different long classifications of bag, the Shuo Ju Bao Lost bag rates that each in the default different long classifications of bag wraps long classification are calculated, Shuo Ju Bao Lost bags rate is the ratio that the first statistical result neutralizes the number of the number difference packet corresponding with same class bag length in the first statistical result of the packet of same class bag length in the second statistical result；

Processor 91, for the default different packets for wrapping long classification to be carried out into ascending sort according to long classification is wrapped, the at least Shuo Ju Bao Lost bags rates including predetermined number arrange for ascending order in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag, then judge current network Lost Packet types as Wu Ma Lost bags；If being arranged in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag according to ascending order The number of the Shuo Ju Bao Lost bag rates of row is less than predetermined number, then judges current network Lost Packet types as Yong Sai Lost bags.

It is further alternative, processor 91, for when judging the Lost Packet types of current network as code Lost bags by mistake, the number of at least one long classification of bag described in the average Bao Changyu of the packet of at least one long classification of bag to be carried out into linear fit according to bag Lost bags rate；

Processor 91, for when the slope of a curve that fitting is obtained is ascendant trend or slope of a curve is the fixed value more than default slope, then judging that current network has Wu Ma Lost bags；When the corresponding Shuo Ju Bao Lost bag rates of all long classifications of bag are all higher than the wealthy value of default Shuo Ju Bao Lost bag rates in the obtained curve of fitting, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；Obtained curve is wherein fitted using the long classification of bag as the abscissa in coordinate space, using Shuo Ju Bao Lost bags rates as ordinate.

Optionally, processor 91, for when judging current network Lost Packet types as Wu Ma Lost bags, obtain currently receive at least one packet when extend to the average delay of received data packet, when the time delay for currently receiving at least one packet is more than the average delay of received data packet, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；When the time delay for currently receiving at least one packet is less than or equal to the average delay of received data packet, judge that current network has Wu Ma Lost bags.The second network element device that embodiments of the invention are provided, accurately can quickly judge Wang Luo Lost Packet types according to the dynamic change for wrapping Chang Ji Lost bag rates for sending the equipment packet sent and the packet received.

Shown in reference picture 16, a kind of network system that embodiments of the invention are provided, including at least one first network element device 101 and second network element device 102, wherein the first network element device 101 and the second network element device 102 are connected by switching network, any first network element device that first network element device 101 provides for the said equipment embodiment, any second network element device that the second network element device 102 provides for the said equipment embodiment.The network element device that the network system that certain embodiments of the invention are provided can be provided using embodiments of the invention realizes the Wang Luo Lost Packet type recognition methods that embodiments of the invention are provided, wherein, the structure for each network element device that the system is included, and the detailed process for the Wang Luo Lost Packet type recognition methods realized by each network element device is repeated no more here with reference to the description of the embodiment above with respect to method and apparatus. The network system that embodiments of the invention are provided, the dynamic change of the Chang Ji Lost bag rates of bag for the packet that can be received according to the packet and receiving device for sending equipment transmission, accurately quickly judges Wang Luo Lost Packet types.One of ordinary skill in the art will appreciate that：Realizing all or part of step of above method embodiment can be completed by the related hardware of programmed instruction, foregoing program can be stored in a computer read/write memory medium, the program upon execution, performs the step of including above method embodiment；And foregoing storage medium includes：ROM, RAM, magnetic disc or CD etc. are various can be with the medium of store program codes.

It is described above; only embodiment of the invention, but protection scope of the present invention is not limited thereto, any one skilled in the art the invention discloses technical scope in; change or replacement can be readily occurred in, should be all included within the scope of the present invention.Therefore, protection scope of the present invention should be based on the protection scope of the described claims.

Claims

Claims

1st, a kind of Wang Luo Lost Packet type recognition methods, it is characterised in that including：

First network element device sends packet to the second network element device；

First network element device will be sent to the packet of second network element device is classified by the first mode classification, and first mode classification is to classify by the preset range that bag is long or bag is long of the packet；

The data packet number of first network element device statistics transmission all kinds of bag length into the packet of second network element device, generates the first statistical result；

First network element device sends first statistical result to second network element device, so that second network element device judges the type of the Wang Luo Lost bags between first network element device and second network element device according to first statistical result and the second statistical result；Or,

First network element device receives second statistical result that second network element device is sent, and first network element device judges the type of the Wang Luo Lost bags between first network element device and second network element device according to first statistical result and second statistical result；Wherein, second statistical result is after second network element device classifies the packet received by first mode classification, to the statistical result of the data packet number of all kinds of bag length in the packet received.

2nd, the method according to claim 1, it is characterised in that first network element device judges that the type of the Wang Luo Lost bags between first network element device and second network element device includes according to first statistical result and second statistical result：

Obtain the statistical result of the packet of first kind bag length and the packet of Equations of The Second Kind bag length at least one times in first statistical result neutralizes second statistical result, the average bag that the average bag length of the packet of the first kind bag length is more than the packet of the Equations of The Second Kind bag length is long；The Shuo Ju Bao Lost bags rates of the first kind bag length and the Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length are calculated, the Shuo Ju Bao Lost bags rate is the ratio that first statistical result neutralizes the number difference of the packet of same class bag length and the number of the packet of same class bag length described in first statistical result in second statistical result；

When the Shuo Ju Bao Lost bags rate of the first kind bag length is more than the data of the Equations of The Second Kind bag length When the number of times of Bao Lost bag rates is more than default number of times, judge current network Lost Packet types as Wu Ma Lost bags；

When the number of times that the Shuo Ju Bao Lost bags rate of the first kind bag length is less than or equal to the Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length is more than default number of times, judge current network Lost Packet types as Yong Sai Lost bags.

3rd, the method according to claim 1, it is characterised in that first network element device judges that the type of the Wang Luo Lost bags between first network element device and second network element device includes according to first statistical result and second statistical result：

According to first mode classification in first statistical result neutralizes second statistical result, obtain the statistical result of the packet of the default different long classifications of bag, the Shuo Ju Bao Lost bag rates that each in the described default different long classifications of bag wraps long classification are calculated, the Shuo Ju Bao Lost bags rate is the ratio that first statistical result neutralizes the number of the number difference packet corresponding with same class bag length described in first statistical result of the packet of same class bag length in second statistical result；

The default different packets for wrapping long classification are subjected to ascending sort according to long classification is wrapped, the at least Shuo Ju Bao Lost bags rates including predetermined number arrange for ascending order in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag, then judge current network Lost Packet types as Wu Ma Lost bags；If the number of the Shuo Ju Bao Lost bag rates arranged in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag according to ascending order is less than the predetermined number, judge current network Lost Packet types as Yong Sai Lost bags.

4th, the method according to any one of claim 1 ~ 3, it is characterised in that methods described also includes：

When judging the Lost Packet types of current network as code Lost bags by mistake, the number of at least one long classification of bag described in the average Bao Changyu of the packet of at least one long classification of bag is subjected to linear fit according to bag Lost bags rate；

When the slope of a curve that fitting is obtained is ascendant trend or the slope of a curve is the fixed value more than default slope, then judge that current network has Wu Ma Lost bags；

When the corresponding Shuo Ju Bao Lost bag rates of all long classifications of bag are all higher than the wealthy value of default Shuo Ju Bao Lost bag rates in the obtained curve of fitting, then judge that current network has Wu Ma Lost bags and gathered around simultaneously Sai Lost bags；The wherein described obtained curve of being fitted is using the long classification of bag as the abscissa in coordinate space, using the number according to wrapping Lost bags rate as ordinate.

5th, the method according to any one of claim 1 ~ 3, it is characterised in that methods described also includes：

When judging current network Lost Packet types as Wu Ma Lost bags, obtain second network element device currently receive at least one packet when extend to the average delay of the second network element device received data packet, when the time delay for currently receiving at least one packet is more than the average delay of the received data packet, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；When the time delay for currently receiving at least one packet is less than or equal to the average delay of the received data packet, judge that current network has Wu Ma Lost bags.

6th, a kind of Wang Luo Lost Packet type recognition methods, it is characterised in that including

Second network element device receives the packet that the first network element device is sent；

The packet received is classified according to the first mode classification, first mode classification is to classify by the preset range that bag is long or bag is long of the packet；

The data packet number of all kinds of bag length, generates the second statistical result in the packet received described in the second network element device statistics；

Second network element device sends second statistical result to first network element device, so that first network element device judges the type of the Wang Luo Lost bags between first network element device and second network element device according to first statistical result and the second statistical result；

Or,

Second network element device receives first statistical result that first network element device is sent, and second network element device judges the type of the Wang Luo Lost bags between first network element device and second network element device according to first statistical result and second statistical result；Wherein, first statistical result is that first network element device presses transmission to the packet of second network element device after first mode classification classification, to the statistical result of transmission data packet number of all kinds of bag length into the packet of second network element device.

7th, method according to claim 6, it is characterised in that second network element device judges that the type of the Wang Luo Lost bags between first network element device and second network element device includes according to first statistical result and second statistical result： Obtain the statistical result of the packet of first kind bag length and the packet of Equations of The Second Kind bag length at least one times in first statistical result neutralizes second statistical result, the average bag that the average bag length of the packet of the first kind bag length is more than the packet of the Equations of The Second Kind bag length is long；The Shuo Ju Bao Lost bags rates of the first kind bag length and the Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length are calculated, the Shuo Ju Bao Lost bags rate is the ratio that first statistical result neutralizes the number difference of the packet of same class bag length and the number of the packet of same class bag length described in first statistical result in second statistical result；

When the number of times that the Shuo Ju Bao Lost bags rate of the first kind bag length is more than the Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length is more than default number of times, judge current network Lost Packet types as Wu Ma Lost bags；

When the number of times that the Shuo Ju Bao Lost bags rate of the first kind bag length is less than or equal to the Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length is more than default number of times, judge current network Lost Packet types as Yong Sai Lost bags.

8th, method according to claim 6, it is characterised in that second network element device judges that the type of the Wang Luo Lost bags between first network element device and second network element device includes according to first statistical result and second statistical result：

According to first mode classification in first statistical result neutralizes second statistical result, obtain the statistical result of the packet of the default different long classifications of bag, the Shuo Ju Bao Lost bag rates that each in the described default different long classifications of bag wraps long classification are calculated, the Shuo Ju Bao Lost bags rate is the ratio that first statistical result neutralizes the number of the number difference packet corresponding with same class bag length described in first statistical result of the packet of same class bag length in second statistical result；

The default different packets for wrapping long classification are subjected to ascending sort according to long classification is wrapped, the at least Shuo Ju Bao Lost bags rates including predetermined number arrange for ascending order in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag, then judge current network Lost Packet types as Wu Ma Lost bags；If the number of the Shuo Ju Bao Lost bag rates arranged in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag according to ascending order is less than the predetermined number, judge current network Lost Packet types as Yong Sai Lost bags.

9th, the method according to any one of claim 6 ~ 8, it is characterised in that the side Method also includes：

When judging the Lost Packet types of current network as code Lost bags by mistake, the number of at least one long classification of bag described in the average Bao Changyu of the packet of at least one long classification of bag is subjected to linear fit according to bag Lost bags rate；

When the slope of a curve that fitting is obtained is ascendant trend or the slope of a curve is the fixed value more than default slope, then judge that current network has Wu Ma Lost bags；

When the corresponding Shuo Ju Bao Lost bag rates of all long classifications of bag are all higher than the wealthy value of default Shuo Ju Bao Lost bag rates in the obtained curve of fitting, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；The wherein described obtained curve of being fitted is using the long classification of bag as the abscissa in coordinate space, using the number according to wrapping Lost bags rate as ordinate.

10th, the method according to any one of claim 6 ~ 8, it is characterised in that methods described also includes：

When judging current network Lost Packet types as Wu Ma Lost bags, obtain currently receive at least one packet when extend to the average delay of received data packet, when the time delay for currently receiving at least one packet is more than the average delay of the received data packet, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；

When the time delay for currently receiving at least one packet is less than or equal to the average delay of the received data packet, judge that current network has Wu Ma Lost bags.

11, a kind of first network element device, it is characterised in that including：

Data transmission unit, for sending packet to the second network element device；

Data sorting unit, the packet for the data transmission unit to be sent to second network element device is classified by the first mode classification, and first mode classification is to classify by the preset range that bag is long or bag is long of the packet；

Statistic unit, the data packet number for counting all kinds of bag length in the sorted packet of the data sorting unit generates the first statistical result；

The data transmission unit, first statistical result for being additionally operable to forward the statistic unit is sent to second network element device, so that second network element device judges the type of the Wang Luo Lost bags between first network element device and second network element device according to first statistical result and the second statistical result； Or,

Lost bag recognition units, for receiving second statistical result that second network element device is sent, first statistical result and second statistical result that are forwarded according to the statistic unit judge the type of the Wang Luo Lost bags between first network element device and second network element device；Wherein, second statistical result is after second network element device classifies the packet received by first mode classification, to the statistical result of the data packet number of all kinds of bag length in the packet received.

12nd, the equipment according to claim 11, it is characterised in that Suo Shu Lost bag recognition units, including：

Sampling subelement, statistical result for obtaining the packet of first kind bag length and the packet of Equations of The Second Kind bag length in neutralizing second statistical result in first statistical result at least one times, the average bag that the average bag length of the packet of the first kind bag length is more than the packet of the Equations of The Second Kind bag length is long；

Handle subelement, for calculating the Shuo Ju Bao Lost bags rates of the first kind bag length and the Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length, the Shuo Ju Bao Lost bags rate is the ratio that first statistical result neutralizes the number difference of the packet of same class bag length and the number of the packet of same class bag length described in first statistical result in second statistical result；

Differentiate subelement, when being more than default number of times for being more than the number of times of Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length when the Shuo Ju Bao Lost bags rate of the first kind bag length, judge current network Lost Packet types as Wu Ma Lost bags；When the number of times that the Shuo Ju Bao Lost bags rate of the first kind bag length is less than or equal to the Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length is more than default number of times, judge current network Lost Packet types as Yong Sai Lost bags.

13rd, the equipment according to claim 11, it is characterised in that Suo Shu Lost bag recognition units include：

Sampling subelement, according to first mode classification in for neutralizing second statistical result in first statistical result, obtain the statistical result of the packet of the default different long classifications of bag, the Shuo Ju Bao Lost bag rates that each in the described default different long classifications of bag wraps long classification are calculated, the Shuo Ju Bao Lost bags rate is that first statistical result neutralizes the number difference of the packet of same class bag length and same class bag described in first statistical result in second statistical result The ratio of the number of long corresponding packet；

Contrast subunit, for the default different packets for wrapping long classification to be carried out into ascending sort according to long classification is wrapped, the at least Shuo Ju Bao Lost bags rates including predetermined number arrange for ascending order in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag, then judge current network Lost Packet types as Wu Ma Lost bags；If the number of the Shuo Ju Bao Lost bag rates arranged in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag according to ascending order is less than the predetermined number, judge current network Lost Packet types as Yong Sai Lost bags.

14th, the equipment according to any one of claim 11 ~ 13, it is characterised in that first network element device also includes：

Fitting unit, for when judging the Lost Packet types of current network as code Lost bags by mistake, the number of at least one long classification of bag described in the average Bao Changyu of the packet of at least one long classification of bag to be carried out into linear fit according to bag Lost bags rate；

Major punishment unit, for when the slope of a curve that fitting is obtained is ascendant trend or the slope of a curve is the fixed value more than default slope, then judging that current network has Wu Ma Lost bags；When the corresponding Shuo Ju Bao Lost bag rates of all long classifications of bag are all higher than the wealthy value of default Shuo Ju Bao Lost bag rates in the obtained curve of fitting, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；The wherein described obtained curve of being fitted is using the long classification of bag as the abscissa in coordinate space, using the number according to wrapping Lost bags rate as ordinate.

15th, the equipment according to any one of claim 11 ~ 13, it is characterised in that first network element device also includes：

Time delay major punishment unit, for when judging current network Lost Packet types as Wu Ma Lost bags, obtain second network element device currently receive at least one packet when extend to the average delay of the second network element device received data packet, when the time delay for currently receiving at least one packet is more than the average delay of the received data packet, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；When the time delay for currently receiving at least one packet is less than or equal to the average delay of the received data packet, judge that current network has Wu Ma Lost bags.

16th, a kind of second network element device, it is characterised in that including

Data receipt unit, for receiving the packet that the first network element device is sent；

Data sorting unit, for the packet that receives the data receipt unit according to One mode classification is classified, and first mode classification is to classify by the preset range that bag is long or bag is long of the packet；

Statistic unit, the data packet number for counting all kinds of bag length in the sorted packet received of the data sorting unit, generates the second statistical result；

Data transmission unit, second statistical result for the statistic unit to be forwarded is sent to first network element device, so that first network element device judges the type of the Wang Luo Lost bags between first network element device and second network element device according to first statistical result and the second statistical result；

Or,

Lost bag recognition units, for receiving first statistical result that first network element device is sent, first statistical result and second statistical result that are forwarded according to the statistic unit judge the type of the Wang Luo Lost bags between first network element device and second network element device；Wherein, first statistical result is that first network element device presses transmission to the packet of second network element device after first mode classification classification, to the statistical result of transmission data packet number of all kinds of bag length into the packet of second network element device.

17th, equipment according to claim 16, it is characterised in that Suo Shu Lost bag recognition units include：

Sampling subelement, statistical result for obtaining the packet of first kind bag length and the packet of Equations of The Second Kind bag length in neutralizing second statistical result in first statistical result at least one times, the average bag that the average bag length of the packet of the first kind bag length is more than the packet of the Equations of The Second Kind bag length is long；

Handle subelement, for calculating the Shuo Ju Bao Lost bags rates of the first kind bag length and the Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length, the Shuo Ju Bao Lost bags rate is the ratio that first statistical result neutralizes the number difference of the packet of same class bag length and the number of the packet of same class bag length described in first statistical result in second statistical result；

Differentiate subelement, when being more than default number of times for being more than the number of times of Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length when the Shuo Ju Bao Lost bags rate of the first kind bag length, judge current network Lost Packet types as Wu Ma Lost bags；When the number of times that the Shuo Ju Bao Lost bags rate of the first kind bag length is less than or equal to the Shuo Ju Bao Lost bag rates of the Equations of The Second Kind bag length is more than default number of times, judge current Network Lost Packet types are Yong Sai Lost bags.

18th, equipment according to claim 16, it is characterised in that Suo Shu Lost bag recognition units include：

Sampling subelement, according to first mode classification in for neutralizing second statistical result in first statistical result, obtain the statistical result of the packet of the default different long classifications of bag, the Shuo Ju Bao Lost bag rates that each in the described default different long classifications of bag wraps long classification are calculated, the Shuo Ju Bao Lost bags rate is the ratio that first statistical result neutralizes the number of the number difference packet corresponding with same class bag length described in first statistical result of the packet of same class bag length in second statistical result；

Contrast subunit, for the default different packets for wrapping long classification to be carried out into ascending sort according to long classification is wrapped, the at least Shuo Ju Bao Lost bags rates including predetermined number arrange for ascending order in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag, then judge current network Lost Packet types as Wu Ma Lost bags；If the number of the Shuo Ju Bao Lost bag rates arranged in the Shuo Ju Bao Lost bag rates of the default different long classifications of bag according to ascending order is less than the predetermined number, judge current network Lost Packet types as Yong Sai Lost bags.

19th, the equipment according to any one of claim 16 ~ 18, it is characterised in that second network element device includes：

Fitting unit, for when judging the Lost Packet types of current network as code Lost bags by mistake, the number of at least one long classification of bag described in the average Bao Changyu of the packet of at least one long classification of bag to be carried out into linear fit according to bag Lost bags rate；

Major punishment unit, for when the slope of a curve that fitting is obtained is ascendant trend or the slope of a curve is the fixed value more than default slope, then judging that current network has Wu Ma Lost bags；When the corresponding Shuo Ju Bao Lost bag rates of all long classifications of bag are all higher than the wealthy value of default Shuo Ju Bao Lost bag rates in the obtained curve of fitting, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；The wherein described obtained curve of being fitted is using the long classification of bag as the abscissa in coordinate space, using the number according to wrapping Lost bags rate as ordinate.

20th, the equipment according to any one of claim 16 ~ 18, it is characterised in that second network element device also includes：

Time delay major punishment unit, for when judging current network Lost Packet types as Wu Ma Lost bags, Obtain currently receive at least one packet when extend to the average delay of received data packet, when the time delay for currently receiving at least one packet is more than the average delay of the received data packet, then judge that current network has Wu Ma Lost bags and Yong Sai Lost bags simultaneously；When the time delay for currently receiving at least one packet is less than or equal to the average delay of the received data packet, judge that current network has Wu Ma Lost bags.

21st, a kind of network system, it is characterized in that, including at least one first network element device and second network element device, wherein described first network element device and the second network element device are connected by switching network, first network element device is the network element device described in any one of claim 11 ~ 15, and second network element device is the network element device described in any one of claim 16 ~ 20.