JP5698107B2 - Network evaluation method and network evaluation system - Google Patents

Description

  The present invention relates to a network evaluation method and a network evaluation system technique.

  A network in which discrete packets are transmitted, such as an MPLS (Multi Protocol Label Switching) packet network, will continue to expand, and various services have been devised and provided. As a network used in MPLS, SDH (Synchronous Digital Hierarchy) / SONET (Synchronous Optical Network) excellent in quality and Ethernet (registered trademark) that can provide a service at a low price are adopted. Further, there is an MPLS network in which SDH / SONET and Ethernet are used together, and it becomes possible to manage signals of different standards centrally in one network.

  As a packet network maintenance function, the number of missing packets passing through the packet network is detected.

  For example, Ether OAM (Operation, Administration and Maintenance) has been recommended as a maintenance function of a packet network (see Non-Patent Document 1). With the LM (Loss Measurement) function of Ether OAM, the packet transmission device stores the number of transmission packets in a CCM (Continuity Check Message) frame and transmits it, and the packet reception device stores the number of received packets and the received CCM frame. The number of transmitted packets is compared, and the number of lost packets is monitored.

Moreover, in the method for diagnosing a transmission path state described in Patent Document 1, not only user packets configured as user traffic between user terminals but also a dummy packet for diagnosis is sent during a blank period when the user packets are not flowing. By continuing, the number of lost packets is monitored from the difference between the total number of transmissions and the total number of receptions of these two types of packets.
In Patent Document 2, similarly to Patent Document 1, a test frame is transmitted to an unused band, and an error rate is calculated.

JP 2011-114380 A JP 2010-16654 A

ITU-T Y.1731: Requirements for OAM functions in Ethernet-based networks and Ethernet services

  However, in the prior art, since the reliability of the packet network is evaluated from the number of missing packets, it cannot be said that the evaluation accuracy is sufficient. Packets flowing from the entrance to the exit of a packet network are often damaged due to deterioration of communication quality on the packet network. Here, the main reason why the evaluation accuracy is not sufficient is that the degree of packet damage is not taken into consideration.

  For example, packet loss means that 100% of packets are lost, but it is not necessary to count the same “missing one packet” regardless of whether the packet damage level is 1% or 99%. The communication quality on the packet network, which causes a packet damage level, is not reflected.

  In addition, even if a packet is damaged, a damaged packet may be repaired by a router device on the route due to an error correction code or the like. Even in this case, the damage degree of the packet is not reflected in the reliability evaluation of the packet network only by counting “one packet passing”.

  In view of the above, the main object of the present invention is to solve the above-described problems and to realize the reliability evaluation of the packet network with high accuracy.

In order to solve the above problems, the present invention is a network evaluation method for evaluating the reliability of an evaluation target network for transferring a test packet from an ingress edge device to an egress edge device,
The previous entry-side edge device,
For a user signal that has flowed into the evaluation target network, a plurality of types of error detection codes for detecting a transmission error of the user signal are generated, each of the generated error detection code fields, and the evaluation target network A user packet encapsulated in the user signal and a field indicating a destination used for transfer in the device as the inspection packet and transmitted to the outgoing edge device ;
In a blank period in which the user packet is not transmitted, a plurality of dummy packet headers including a concatenation identifier indicating that a plurality of dummy signals having a predetermined length that do not flow out of the evaluation target network are concatenated. A dummy packet that is packetized by concatenating a plurality of dummy signals with error detection codes corresponding to types is also generated as the inspection packet and transmitted to the outgoing edge device,
The exit-side edge device, the plurality of error detecting codes respectively taken out received the test packet and decapsulates the error detection operation intended for the user signal and the dummy signal taken out by decapsulation, the plurality The number of error detections for each type is calculated, and the error rate of the evaluation target network is calculated according to the ratio between the calculated number of error detections and the amount of data of the received inspection packet.
Other means will be described later.

  According to the present invention, reliability evaluation of a packet network can be realized with high accuracy.

It is a block diagram which shows the communication system regarding one Embodiment of this invention. It is explanatory drawing which shows the packet which flows through the communication system regarding one Embodiment of this invention. It is explanatory drawing which shows the creation process of the dummy packet regarding one Embodiment of this invention. It is a block diagram which shows each edge apparatus of the communication system regarding one Embodiment of this invention. It is a flowchart which shows the process of the entrance edge apparatus regarding one Embodiment of this invention. It is a flowchart which shows the process of the outgoing side edge apparatus regarding one Embodiment of this invention.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

FIG. 1A is a configuration diagram showing a communication system. The communication system is configured as an MPLS network 9 (evaluation target network) that connects the transmission-side client device 7 and the reception-side client device 8. In the MPLS network 9, an ingress edge device 1 that accommodates a transmitting client device 7, an egress edge device 3 that accommodates a receiving client device 8, and a core device 2 that does not accommodate a client device are connected to each other. ing.
Each device constituting the communication system is configured as a computer having a CPU (Central Processing Unit), a memory, a hard disk (storage means), and a communication interface, and the computer executes a program read on the memory. Thus, each processing unit is operated.

The MPLS network 9 is configured in a clockwise ring shape (incoming edge device 1 → upper core device 2 → outgoing edge device 3 → lower core device 2 → incoming edge device 1 →). It also has redundancy such as switching the transmission path when a failure occurs. For example, the user packet reaches the outgoing edge device 3 via the upper core device 2, but if there is an abnormality such as a failure in the upper core device 2 or a link connected to the core device 2, A section that passes through the lower core apparatus 2 can also be selected as a transmission path for user packets.
Instead of the MPLS network 9, a network that connects client devices may be formed as an Ethernet network or a network that complies with other standards, recommendations, or methods.

In the communication system, user packets and dummy packets flow as packets for discretely transmitting signals.
The user packet is a packet constituting user traffic exchanged with the client device. For example, in FIG. 1A, a user signal that is an SDH or Ethernet frame transmitted from the transmission-side client device 7 is encapsulated as an MPLS packet (inspection packet) by the ingress edge device 1 of the MPLS network 9. Then, the data is transferred to the outgoing edge device 3 via the core device 2. Then, the outgoing edge device 3 transfers the user signal (SDH or Ethernet frame) obtained by decapsulating the MPLS packet to the receiving client device 8.

  Note that the user packet encapsulation / decapsulation processing in the MPLS network 9 may use, for example, T-MPLS (Transport-Multi Protocol Label Switching). T-MPLS is described in the document “ITU-T (International Telecommunications Union Telecommunications Standardization Sector) G.8121 / Y.1381: Characteristics of Transport MPLS equipment functional blocks”.

FIG. 1B is an explanatory diagram showing a packet flowing through the MPLS network 9 and its Valid information.
The dummy packet is a packet that is generated in a blank period that does not correspond to an IFG (Inter Frame Gap) that is a gap period between packets and that does not flow in the MPLS network 9 and flows in the MPLS network 9. is there. For example, in FIG. 1 (a), it is created by the ingress edge device 1 and transferred to the egress edge device 3 via the upper and lower core devices 2. Unlike the user packet, the dummy packet is not transmitted to the receiving client device 8.
The ingress edge device 1 starts generating the dummy packet after securing the minimum IFG length after the user packet stops flowing, and before the arrival of the next user packet, the dummy packet is transmitted by the minimum IFG length. End generation. Note that when the interval between two consecutive user packets is narrow (the minimum IFG length cannot be secured), the generation of the dummy packet may be omitted.

The Valid information is a control signal indicating a period during which a user packet is flowing in the MPLS network 9. During the period when the Valid information is “ON”, a user packet flows in the MPLS network 9.
In this way, by evaluating the reliability in the MPLS network 9 based not only on the user packet but also on the transmission degree (error rate) of the dummy packet, the reliability in the MPLS network 9 is always maintained (user packet). (Even during a period during which no flow occurs), the failure in the MPLS network 9 can be detected early.

  FIG. 2 is an explanatory diagram showing packets flowing through the communication system. In the present embodiment, in the packet diagram, the left side is the beginning of the packet and the right side is the end of the packet.

FIG. 2A shows a standard MPLS packet. The MPLS packet is composed of a MAC (Media Access Control) header of the packet TOP, an MPLS header, a user signal, and an FCS (Frame Check Sequence) footer of the packet END in order from the top.
The MPLS header is composed of a T-LSP indicating a destination in the MPLS network 9 and an HEC (Header Error Control) that is a 1-bit error correction code for the T-LSP. About HEC, it describes in Unexamined-Japanese-Patent No. 2010-206741, for example. This HEC is given to suppress an error in the immediately preceding field (for example, T-LSP) (for example, an error in the packet transmission destination in the case of T-LSP).
The user signal may be an SDH / SONET frame or an Ethernet frame.

FCS indicates that one unit packet ends here, and an error detection code generated by CRC (Cycle Redundancy Check) processing for the first half (MAC, MPLS, user signal) of the packet is stored. Yes.
Then, the CRC processing to the FCS by the egress edge device 3 makes it possible to detect an error of 1 bit or more generated during the packet transmission process, that is, the packet error. The egress edge device 3 may discard the user packet in which an error is detected by FCS, and may request the ingress edge device 1 (which may be the transmission client device 7) to retransmit the discarded user packet. .

  FIG. 2B shows an MPLS packet (user packet) with an extended identifier. The MPLS packet in FIG. 2B is different from the MPLS packet in FIG. 2A in that a BIP (Bit Interleaved Parity) field is provided after the user signal, and a combination of an identifier field and its HEC in the MPLS header. The point of providing is expanded.

  A parity code is stored in the BIP field. This parity code is generated, for example, in the same manner as the BIP field in the SDH frame transmitted for 125 us in the SDH network. For example, if the B1 byte is SOH (Section Over Head), a parity check (bit error detection method) using BIP-8 calculation generates a BIP parity code that can detect errors of up to 8 bits per frame. Is done.

Note that “8” in BIP-8 means that the check target data in one packet is divided into eight partial area data, and one parity bit is added to each partial area data.
By referring to the FCS, a 1-bit error of the packet can be detected, and by referring to the BIP, a multi-bit error of the packet (for example, an error from 1 bit to 8 bits in the case of BIP-8) can be detected.
Here, when the BIP parity code is configured as an even parity code and an even number of bit errors exist in the partial area data, the error in the partial area data is not detected even in the parity check.

The identifier field includes a dummy identifier for distinguishing whether data following the identifier field is a dummy signal (value “1”) or a user signal (value “0”), and after each signal (before FCS). It includes a BIP identifier indicating whether or not a BIP is provided, and a connection identifier indicating that a dummy signal is connected later (the connection will be described later in FIG. 3).
Since the BIP identifier is included, the user packet and the dummy packet may be a packet including the BIP or a packet not including the BIP. Also, since a plurality of user packets that store user signals are not concatenated, no concatenation identifier is required.
Then, after the identifier field, HEC for the identifier field is given.

On the other hand, a dummy signal can be connected to the dummy packet shown in FIG. The difference between the dummy packet and the user packet is that the user signal is replaced with a dummy signal and the value “1” is stored as a dummy identifier.
Since the dummy signal is reliability measurement data, the dummy signal is not transmitted to the transmission-side client device 7 or the reception-side client device 8. Therefore, as the dummy signal, an arbitrary signal such as a signal in which 0 and 1 continue alternately or a PN (Pseudorandom Noise) code may be adopted.

  The dummy packet is processed in the MPLS network 9 in the same way as the user packet (such as label switching processing referring to the T-LSP in the MPLS header), and after being used to calculate the error rate, the outgoing edge device 3 Identified and discarded. The size of the dummy packet (packet length) may be arbitrary as long as the minimum packet length is 64 bytes or more.

FIG. 3 is an explanatory diagram showing a dummy packet creation process. One dummy packet is configured so that one MAC header and MPLS header are headed, and then a set of one or more “64-byte length dummy signal, BIP, and FCS” is concatenated one after another. Also good.
As described with reference to FIG. 1B, the purpose of the dummy packet is to fill in the blank period of the user packet. Therefore, it is desirable to use a variable-length packet that can be connected in this way. By generating one variable-length concatenated packet instead of creating a plurality of dummy packets, there is no need to prepare MAC headers, MPLS headers, and IFGs for the number of dummy packets. The error rate calculation accuracy can be improved by increasing the line usage rate by the dummy packet (dummy signal).

FIG. 4 is a configuration diagram showing each edge device of the communication system.
The incoming edge device 1 includes a packet recognition unit 11, a dummy packet insertion unit 12, a selector 13, a packet length measurement unit 14, an address control unit 15, a buffer 16, a header insertion unit 17, and a BIP addition unit. 18 and the FCS giving unit 19 are included.

The packet recognition unit 11 notifies the arrival period of the received user packet to the dummy packet insertion unit 12 as Valid information, and transfers the received user packet to the selector 13 at the subsequent stage.
The dummy packet insertion unit 12 generates a dummy packet during a period in which the Valid information is “OFF”, and transfers the dummy packet to the subsequent selector 13.
The selector 13 inserts the transmission packet (user packet and dummy packet) received from the previous stage into the subsequent stage (packet length measurement unit 14) toward the transmission path.

The packet length measurement unit 14 measures the packet length of the transmission packet from the selector 13 and notifies the address control unit 15 of the result. Further, the packet length measurement unit 14 writes the transmission packet in the buffer 16.
The address control unit 15 determines whether or not to transmit each transmission packet in the buffer 16 to the core device 2 by controlling the write pointer (write address) of the buffer 16.
The buffer 16 is a transmission buffer that temporarily stores transmission packets.

The header insertion unit 17 gives an MPLS label to the transmission packet in the buffer 16.
The BIP appending unit 18 appends the BIP calculation result described in FIG. 2B to the transmission packet in the buffer 16. However, what kind of calculation is used as the BIP calculation may be arbitrarily set according to the needs of the user.
The FCS appending unit 19 appends the FCS calculation result described in FIG. 2A to the transmission packet in the buffer 16.

  The outgoing edge device 3 includes a header analysis unit 31, an error detection unit 32, an error rate calculation unit 33, an address control unit 34, a header deletion unit 36, an FCS deletion unit 37, a BIP deletion unit 38, And a buffer 39.

The header analysis unit 31 analyzes the MPLS header information of the received packet, identifies whether it is a dummy packet or a user packet, and identifies the presence or absence of BIP.
The error detection unit 32 includes an FCS error detection unit 32a and a BIP error detection unit 32b in order to detect errors in the received packet.
The FCS error detection unit 32a refers to the FCS of the received packet and detects an error.
The BIP error detection unit 32b refers to the BIP of the received packet and detects an error. Note that BIP error detection processing is omitted for packets without BIP.

The error rate calculation unit 33 calculates the error rate from the number of errors detected by the error detection unit 32 (the details of the calculation formula will be clarified in the description of the flowchart of FIG. 6).
The address control unit 34 controls the write pointer (write address) of the buffer 39 to discard the dummy packet and transfer the user signal depacketed from the user packet to the subsequent receiving client device 8.

The header deletion unit 36 deletes (depackets) the MAC header and the MPLS header for the received packet that has been passed through the error detection unit 32 and written to the buffer 39.
The FCS deletion unit 37 deletes the FCS footer of the received packet processed by the header deletion unit 36.
The BIP deletion unit 38 deletes the BIP field of the received packet processed by the FCS deletion unit 37.
The buffer 39 is a transmission buffer for storing user packets (user signals) transferred to the receiving client device 8 at the subsequent stage.

FIG. 5 is a flowchart showing processing of the entry side edge device 1.
As S101, the packet recognition unit 11 determines whether or not the present time is a period in which a user packet has arrived. The transfer period from the MAC header, which is the TOP (start) of the user packet, to the FCS footer, which is the END (end), corresponds to the period during which the user packet has arrived (the period when the Valid information is “ON”). If Yes in S101, the process proceeds to S111, and if No, the process proceeds to S102.

In S <b> 102, the dummy packet insertion unit 12 receives notification of Valid information “OFF” from the packet recognition unit 11 and starts creating a dummy packet (header creation). Note that, as described in FIG. 1B, when the period of two consecutive user packets is short, the process may return to S101 without generating a dummy packet.
As S103, the packet recognizing unit 11 determines whether or not the user packet has arrived from the Valid information, as in S101. If Yes in S103, the process proceeds to S105, and if No, the process proceeds to S104.
As S104, the dummy packet insertion unit 12 continues to create dummy packets. When the dummy packet that has been generated becomes 64 bytes or more, the generation is continued as a concatenated packet as shown in FIG.
In S105, the dummy packet insertion unit 12 receives the notification of the valid information “ON” from the packet recognition unit 11, ends the creation of the dummy packet, and proceeds to S112.

In S <b> 111, the selector 13 passes the user packet notified from the packet recognition unit 11 to the packet length measurement unit 14.
As S <b> 112, the selector 13 inserts the dummy packet that has been created in S <b> 105 into the packet length measurement unit 14.

As S121, the packet length measurement unit 14 measures the packet length of the transmission packet (the user packet of S111 or the dummy packet of S112), and the packet length is normal (for example, 64 bytes which is the minimum packet unit). Whether or not). Here, the packet length measurement unit 14 also measures the packet length of the connected packets as one packet.
If Yes in S121, the address control unit 15 writes the transmission packet in the buffer 16 and causes the following processing (S122 to S125) of the transmission packet to be executed. If No in S121, the process proceeds to S126.

As S122, the header insertion unit 17 adds a header to the transmission packet written in S121.
As S123, the BIP assigning unit 18 assigns the BIP to the transmission packet in S122.
In S124, the FCS grant unit 19 grants FCS to the transmission packet in S123.
In S <b> 125, the address control unit 15 transmits the transmission packet in S <b> 124 to the core device 2 that heads for the outgoing edge device 3.

  In S126, the address control unit 15 discards the transmission packet. Here, the address control unit 15 may discard the transmission packet by not writing it to the buffer 16, or after writing the transmission packet to the buffer 16, the address control unit 15 sets a write pointer (write address) to the buffer 16. Even if the current transmission packet is discarded, the transmission packet scheduled to be written next overwrites the memory area of the transmission packet written this time, by moving back by the packet length of the written transmission packet. Good.

FIG. 6 is a flowchart showing processing of the outgoing edge device 3. The flowchart in FIG. 6A is started when the egress edge device 3 receives an MPLS packet from the transmission side.
In S201, the header analysis unit 31 analyzes the header and checks the dummy packet identifier to determine whether the packet is a user packet (whether it is a dummy packet). If Yes in S201, S202 and S203 are executed, and if No in S201, S202 and S204 are executed.

As S202, the egress edge device 3 performs egress processing for each packet (refer to FIG. 6B described later for details). The process of S202 is executed regardless of the determination result of S201.
As S203, the address control unit 34 writes the user signal extracted from the user packet in the buffer 39.
In S204, the address control unit 34 discards the dummy packet. Note that the packet discarding method may be discarded by not writing the dummy signal to the buffer 39 as in S126, or after the dummy signal is written to the buffer 39, the write pointer (write address) is set. The written dummy signal may be discarded by returning it to the previous position.

The flowchart of FIG. 6B is called from S202 of FIG.
As S211, the FCS error detection unit 32a detects an FCS error (1 bit error) by the CRC processing described with reference to FIG.
In S212, the BIP error detection unit 32b detects a BIP error (multi-bit error) by the parity check process described with reference to FIG.
If the dummy packet is connected, the connected packet is distributed to the dummy packet group before connection, and then the error calculation process of S211 and S212 is performed for each dummy packet constituting the dummy packet group. .

As S213, the error rate calculation unit 33 calculates an error rate based on the number of FCS errors in S211 and the number of BIP errors in S212. As the error rate calculation formula, the following first formula or second formula is used.
First, an example of an error rate calculation formula (first formula) for a packet without BIP is shown.
Error rate = Σ (number of FCS errors) / Σ (packet length [bit])
Here, the symbol “Σ” indicates the sum calculation of the processing results when the processing of S213 is performed a plurality of times (a plurality of packets).

Next, an example of an error rate calculation formula (second formula) for a packet with BIP is shown.
Error rate = Σ (number of FCS errors or BIP errors) / Σ (packet length [bit])
In the second equation, the error rate is calculated by complementing BIP and FCS. Therefore, even if BIP cannot detect an error due to an even bit error on the same rail, there is an error in the packet in FCS. I can confirm that.
Here, the numerator of the second formula (the number of errors of FCS or the number of errors of BIP) is, for example, if the number of detected (non-zero) errors of both types of errors is one ( For example, the number of errors of FCS = 1, the number of errors of BIP = 0, the number of detected errors is adopted (numerator = 1), and the number of detected errors is two (for example, the number of errors of FCS = 1, the number of BIP errors = 4) and the number of BIP errors (numerator = 4).

As S <b> 221, the header deletion unit 36 deletes the MAC header and the MPLS header from the received packet written in the buffer 39.
In S222, the FCS deletion unit 37 deletes the FCS from the received packet written in the buffer 39.
As S223, the BIP deletion unit 38 deletes the BIP from the received packet written in the buffer 39. Thereby, the user signal is extracted from the user packet. Then, the user signal is transferred from the outgoing edge device 3 to the receiving client device 8.

According to the present embodiment as described above, the error detection unit 32 detects an error for each partial area of the packet (multiple bits) by the error detection process for each bit referring to the FCS of the MPLS packet and the parity check referring to the BIP. By using the processing together (two types are combined, but three or more types may be used), the degree of damage of the MPLS packet due to the network quality of the MPLS network 9 can be detected with high accuracy.
As a result, the error rate calculation unit 33 can calculate the reliability of the packet network with high accuracy by calculating the high accuracy error rate using the detection result of the error detection unit 32.

Further, the dummy packet insertion unit 12 performs error rate calculation processing by the error rate calculation unit 33 using both the user packet and the dummy packet by inserting the dummy packet during a period when the user packet is not flowing. Let
Thereby, the calculation result of the highly accurate error rate by the error rate calculation unit 33 can be obtained with a fine time density (always).

As described above, by improving the evaluation accuracy and the evaluation frequency of the network quality of the MPLS network 9, for example, while reinforcing or repairing only a portion where the state of the transmission line is extremely bad, the state of the transmission line It is possible to construct a low-priced and high-quality packet network that suppresses excessive investment in areas with good quality.
Therefore, since the investment efficiency of the equipment cost of the same network is improved, a packet network with high network quality can be realized. In addition, services that require network quality can be accommodated on the packet network, and various services can be provided.

DESCRIPTION OF SYMBOLS 1 Incoming edge apparatus 2 Core apparatus 3 Outgoing edge apparatus 7 Transmission side client apparatus 8 Reception side client apparatus 9 MPLS network (network to be evaluated)
DESCRIPTION OF SYMBOLS 11 Packet recognition part 12 Dummy packet insertion part 13 Selector 14 Packet length measurement part 15 Address control part 16 Buffer 17 Header insertion part 18 BIP provision part 19 FCS provision part 31 Header analysis part 32 Error detection part 32a FCS error detection part 32b BIP error Detection unit 33 Error rate calculation unit 34 Address control unit 36 Header deletion unit 37 FCS deletion unit 38 BIP deletion unit 39 Buffer

Claims (6)

A network evaluation method for evaluating the reliability of an evaluation target network that transfers inspection packets from an ingress edge device to an egress edge device,
The entry edge device is:
For a user signal that has flowed into the evaluation target network, a plurality of types of error detection codes for detecting a transmission error of the user signal are generated, each of the generated error detection code fields, and the evaluation target network A user packet encapsulated in the user signal and a field indicating a destination used for transfer in the device as the inspection packet and transmitted to the outgoing edge device ;
In a blank period in which the user packet is not transmitted, a plurality of dummy packet headers including a concatenation identifier indicating that a plurality of dummy signals having a predetermined length that do not flow out of the evaluation target network are concatenated. A dummy packet that is packetized by concatenating a plurality of dummy signals with error detection codes corresponding to types is also generated as the inspection packet and transmitted to the outgoing edge device,
The egress edge device performs the error detection operation on the user signal and the dummy signal that are decapsulated and extracted for each of a plurality of error detection codes that are decapsulated and extracted from the received inspection packet. Calculating the number of error detections of each type, and calculating the error rate of the evaluation target network according to a ratio between the calculated number of error detections and the amount of data of the received inspection packet Evaluation method. The ingress edge device generates the plurality of types of error detection codes for the user signal and the dummy signal, respectively .
As one of the plurality of types, an FCS (Frame Check Sequence) that detects an error in 1-bit units for each of the entire user signal and the entire dummy signal is generated,
As the other of the plurality of types, generating a BIP (Bit Interleaved Parity) for detecting an error for each partial region for each partial region obtained by dividing the user signal and the dummy signal into a plurality of sections. The network evaluation method according to claim 1. The ingress edge device adds an identifier indicating whether the inspection packet is the user packet or the dummy packet, encapsulates the inspection packet,
The egress edge device refers to the identifier decapsulated from the received inspection packet and identifies whether the received inspection packet is the user packet or the dummy packet. The decapsulated and extracted user signal is caused to flow out of the evaluation target network, and when it is the dummy packet, the decapsulated and extracted dummy signal is discarded.
The network evaluation method according to claim 1 or 2 . The ingress edge device measures the packet length of the inspection packet transmitted to the egress edge device, and when the abnormal inspection packet with a measurement result less than the minimum packet length is created, The network evaluation method according to any one of claims 1 to 3 , wherein the abnormal inspection packet is discarded without being transmitted. A network evaluation system for evaluating the reliability of an evaluation target network that transfers inspection packets from an ingress edge device to an egress edge device,
For a user signal that has flowed into the evaluation target network, a plurality of types of error detection codes for detecting a transmission error of the user signal are generated, each of the generated error detection code fields, and the evaluation target network A user packet encapsulated in the user signal and a field indicating a destination used for transfer in the device as the inspection packet and transmitted to the outgoing edge device ;
In a blank period in which the user packet is not transmitted, a plurality of dummy packet headers including a concatenation identifier indicating that a plurality of dummy signals having a predetermined length that do not flow out of the evaluation target network are concatenated. The ingress edge device that generates a dummy packet that is packetized by concatenating a plurality of dummy signals with error detection codes for various types and transmits the dummy packet to the egress edge device; and
For each of a plurality of error detection codes extracted by decapsulating the received inspection packet, the number of error detections of each of the plurality of types is obtained by error detection calculation for the user signal and the dummy signal extracted by decapsulation. The output-side edge device that calculates and calculates the error rate of the network to be evaluated according to a ratio between the calculated error detection count and the amount of data of the received inspection packet. Network evaluation system. Entering-side edge device, as one of when the user signal and the dummy signal each of the error detecting code generating plural kinds min, for each of the user signals across and the dummy signal whole detecting an error in 1bit unit On the other hand, FCS (Frame Check Sequence) is generated. On the other hand, BIP (Bit Interleaved Parity) for detecting an error for each partial area is obtained for each partial area obtained by dividing the user signal and the dummy signal into a plurality of sections. A detection code adding unit that generates and adds the generated error detection code to the inspection packet;
The network evaluation system according to claim 5 .

Images