JP2010166244A - Device and method for determining packet loss - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To detect a packet loss of a fragment packet in real time. <P>SOLUTION: A receiving part 11 receives a fragment packet, a packet processing part 13 determines whether the fragment packet is the last fragment packet, and when any fragment packets are not received upon receiving the last fragment packet and when unreceived fragment packets can not be received within a prescribed time, a packet loss determining part 14 determines that the packet loss has occurred. Consequently, it is possible to detect a packet loss in real time while taking into consideration the switching of arrival orders of the fragment packets. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a technique for detecting a packet loss of a received fragment packet.

  There are a plurality of existing methods for determining the loss of RTP (Real-time Transport Protocol) packets. For example, Patent Document 1 describes a loss detection method considering fragment packets. Patent Document 2 describes a transmission rate control method based on a packet loss rate. Patent Document 3 proposes a method for detecting the occurrence of packet loss including burst loss in real time.

JP-A-2005-328131 JP 2001-144802 A JP-A-2005-159807

  However, in Patent Document 1, when the last packet is received, since retransmission of the lost packet in the packet group is requested, it is determined that the packet arriving after the last packet is also lost due to packet reordering or the like. End up.

  Moreover, in patent document 2, it is a system which determines a data transmission rate based on the feedback information from a receiver, and cannot detect a packet loss in real time.

  In Patent Document 3, it is assumed that a sequence number is assigned to each packet, and it is not assumed that the packet is fragmented.

  The present invention has been made in view of the above, and an object thereof is to detect packet loss of a fragment packet in real time.

  A packet loss determination apparatus according to a first aspect of the present invention includes: a receiving unit that receives a fragment packet; and the last fragment packet of a series of fragment packets obtained by dividing the packet into a single packet when the fragment packet is received. A packet processing means for determining whether or not there is a case where all the fragment packets of a series of fragment packets to which the last fragment packet belongs are not received when the last fragment packet is received, And a packet loss determination means for determining that a packet loss has occurred when the unreceived fragment packet cannot be received within a predetermined time after receiving the fragment packet.

  The packet loss determination method according to the second aspect of the present invention includes a step of receiving a fragment packet, and when the fragment packet is received, the fragment packet is the last fragment packet of a series of fragment packets obtained by dividing one packet A step of determining whether or not all fragment packets of a series of fragment packets to which the last fragment packet belongs are received when the last fragment packet is received, and the last fragment packet is And determining that a packet loss has occurred when the unreceived fragment packet cannot be received within a predetermined time after reception.

  In the present invention, when a fragment packet is received and the fragment packet is the last fragment packet and not all the fragment packets have been received, an unreceived fragment packet cannot be received within a predetermined time. Sometimes, by determining that a packet loss has occurred, real-time packet loss detection can be performed in consideration of changing the arrival order of fragment packets.

  Thus, according to the present invention, the packet loss of the fragment packet can be detected in real time.

It is a block diagram which shows the structure of the packet loss detection apparatus in one embodiment. It is a flowchart which shows the flow of the process of the packet loss determination process of the said packet loss detection apparatus. It is a flowchart which shows the flow which processes a fragment packet. It is a figure which shows the header of a fragment packet.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram illustrating a configuration of a packet loss detection apparatus according to the present embodiment. The packet loss detection apparatus 1 shown in FIG. 1 includes a reception unit 11, a timer 12, a packet processing unit 13, and a packet loss determination unit 14. The packet loss detection device 1 may be configured by a computer including an arithmetic processing device, a storage device, a memory, and the like, and the processing of each unit may be executed by a program. This program is stored in a storage device included in the packet loss detection apparatus 1, and can be recorded on a recording medium such as a magnetic disk, an optical disk, or a semiconductor memory, or provided through a network.

  The receiving unit 11 receives a packet such as a fragment packet obtained by dividing a large packet into several.

  The timer 12 measures time for detecting packet loss. For example, when any of a series of fragment packets obtained by dividing one packet into several pieces has not been received, the time to wait for the arrival of the fragment packet is measured.

  The packet processing unit 13 analyzes the received packet such as the type of packet, the data included in the packet, or whether or not it is the last fragment packet divided.

  The packet loss determination unit 14 determines that a packet loss has occurred, for example, when a packet has not arrived within a predetermined time. For example, when any of a series of fragment packets is not received when the last fragment packet is received, and the unaccepted fragment packet cannot be received within a predetermined time after the last fragment packet is received. It is determined that a packet loss has occurred. There is also a method of determining that a packet loss has occurred when a predetermined time has elapsed, regardless of whether or not the last fragment packet has been received. If a sequence number is assigned to the packet, it is determined that a packet loss has occurred when the received sequence numbers are not continuous.

  Next, the flow of packet loss determination processing for fragment packets will be described. FIG. 2 is a flowchart showing the flow of the packet loss determination process.

  First, when a fragment packet is received (step S201), it is determined whether or not the last fragment packet in the series of fragment packets has been received (step S202). If the last fragment packet has not been received, the process returns to step S201 to wait for the arrival of the next packet. Whether or not the received fragment packet is the last can be known from information attached to the fragment packet.

  If the last fragment packet has been received (including the case where the received packet is the last fragment packet), it is confirmed whether or not a series of fragment packets to which the last fragment packet belongs have been received (step S203). ) If all fragment packets have been received, it is determined as “normal”. The same fragment ID is assigned to a series of fragment packets, that is, when a large packet is divided into several fragment packets, the same fragment ID is assigned to each fragment packet. By checking the total reception size or the number of receptions of fragment packets having the same fragment ID, it is possible to determine whether or not a series of fragment packets have been received.

  On the other hand, if there is a fragment packet that has been received but has not yet been received, a predetermined time is set and the process returns to step S201 to wait for an unreceived fragment packet to arrive. At this time, if the unreceived fragment packet cannot be received within the set time, it is determined that a packet loss has occurred (step S201).

  Here, determination of whether or not all of a series of fragment packets have been received will be described. If it is determined in step S202 that the received fragment packet is the last fragment packet, or the last fragment packet of the same series of fragment packets as the received fragment packet has been received, in step S203 It is determined whether or not all the series of fragment packets have been received.

  For example, the total size of a series of fragment packets is calculated from the offset value and payload length of the last fragment packet, and all the series of fragment packets are received by comparing with the total size of the fragment packets actually received. It can be determined whether or not. The offset value indicates to which position of the packet before fragmentation the data that the fragment packet has. The payload length indicates the size of data that the fragment packet has. Note that the total size of a series of fragment packets, that is, the size of a packet before division, can also be obtained from the header of the packet before division.

  Alternatively, the number of fragment packets is estimated from the MTU (Maximum Transmission Unit) size, and it can be determined as “normal” when the number of fragment packets having the same fragment ID is received.

  Also, as a method of checking whether or not an unreceived fragment packet has arrived, it waits using the offset value of the unreceived fragment packet as a key, and receives a fragment packet having a desired fragment ID and offset value. Can be determined.

  Next, the process of the packet processing unit 13 of the packet loss detection apparatus 1 will be described with a specific example.

  FIG. 3 is a flowchart showing a processing flow of the packet processing unit 13. In the example shown in the figure, an example of receiving an RTP packet divided into a plurality of fragment packets by IPv6 is shown. Of course, an RTP packet that is not divided may be received. FIG. 4 shows an example of analyzing a fragment packet header.

  First, when a packet is received, it is determined whether or not the received packet is IPv6 (step S301). If the received packet is not IPv6, the arrival of the next packet is awaited. Whether or not the received packet is IPv6 can be determined by examining the Type item in FIG.

  If the received packet is IPv6, it is determined whether or not the packet is a fragment packet (step S302). If it is not a fragment packet, the arrival of the next packet is awaited. Whether or not the packet is a fragment packet can be determined by examining the Next header item in FIG.

  Subsequently, it is checked whether or not the fragment packet is the first fragment packet (step S303). When Offset in the Fragmentation Header in FIG. 4 is 0, the fragment packet is the first fragment packet. The first fragment packet is never the last fragment packet.

  When the received packet is the first fragment packet, the RTP sequence number is obtained from the RTP packet stored in the data part of the UDP packet included in the packet, and is continuously with the RTP sequence number of the previously received RTP packet. It is checked whether or not it is being performed (step S304). If the RTP sequence numbers are not consecutive, it is determined that the packet is lost.

  Then, the fragment ID and payload length value are acquired from the first fragment packet and stored in the storage means (step S305). The fragment ID is the value of Identification of Fragmentation Header in FIG. The Payload length value indicates the length of the payload not including the IPv6 header.

  Further, the UDP Length value included in the first fragment packet is acquired and stored in the storage means (step S306). The acquired Length value indicates the length of the divided packet, that is, the data size when all fragment packets are received.

  On the other hand, if the received packet is not the first fragment packet, the fragment ID and the payload length value are acquired from the fragment packet and stored in the storage means (step S307), and whether or not the last fragment packet has been received is determined. Determination is made (step S308). Whether or not it is the last fragment packet can be determined by examining the More Fragment in the Fragmentation Header of FIG. If More Fragment is 0, the fragment packet is the last fragment packet. If the last fragment packet has not been received yet, it waits for the arrival of the next packet.

  When the last fragment packet has been received or when the last fragment packet has been received, it is determined whether or not all the fragment packets have been received (step S309). Whether or not all fragment packets have been received is determined by comparing the Lenghth value of UDP acquired in step S306 with the sum of the payload length values of the same fragment ID acquired in steps S305 and S307. If the sum of the UDP Length value and the Payload length value is the same, all fragment packets have been received, and the next packet is awaited.

  When all the fragment packets have not been received, a timer for measuring a predetermined time is started to wait for the arrival of the next packet. If an unreceived fragment packet cannot be received within a predetermined time, it is determined as a packet loss (step S201 in FIG. 2).

  As described above, according to the present embodiment, the reception unit 11 receives a fragment packet, the packet processing unit 13 determines whether the fragment packet is the last fragment packet, and packet loss determination When the last fragment packet is received by the unit 14, it is determined that a packet loss has occurred when all the fragment packets have not been received and an unreceived fragment packet cannot be received within a predetermined time. As a result, real-time packet loss detection can be performed while considering the change of the arrival order of the fragment packets.

DESCRIPTION OF SYMBOLS 1 ... Packet loss detection apparatus 11 ... Reception part 12 ... Timer 13 ... Packet processing part 14 ... Packet loss determination part

Claims (2)

Receiving means for receiving fragmented packets;
Packet processing means for determining whether the fragment packet is the last fragment packet in a series of fragment packets obtained by dividing one packet when the fragment packet is received;
When the last fragment packet is received, not all the fragment packets of the series of fragment packets to which the last fragment packet belongs have been received, A packet loss determination means for determining that a packet loss has occurred when the unreceived fragment packet cannot be received within the time;
A packet loss determination apparatus comprising: Receiving a fragment packet; and
When receiving the fragment packet, determining whether the fragment packet is the last fragment packet of a series of fragment packets obtained by dividing one packet;
When the last fragment packet is received, not all the fragment packets of the series of fragment packets to which the last fragment packet belongs have been received, Determining that a packet loss has occurred when the unreceived fragment packet cannot be received in time; and
A packet loss determination method comprising:

Images