JP2019149619A - Communication control device, communication control program, and communication control method - Google Patents

Abstract

To simultaneously use a plurality of redundancies on a network, control and determine the redundancies of the entire network according to network information, and improve an error rate, a data transmission rate, and a delay time in a communication system using a disappearance correction code.SOLUTION: A communication control device according to the present invention communicates a coded packet with disappearance correction encoding applied to a destination terminal via a network. The communication control device includes: disappearance correction encoding means for dividing a packet including a communication signal to be transmitted to a destination terminal and applying disappearance correction encoding processing to each division packet; network information storage means for storing network information on a network; and redundancy control means for comprehensively and sequentially controlling a plurality of redundancies related to transmission of each coded packet with disappearance correction encoding applied according to the network information.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a communication control apparatus, a communication control program, and a communication control method, and can be applied to, for example, a packet communication system using an erasure correction code on a wireless or wired network.

  UDP (User Datagram Protocol) communication is a collectionless communication method in which a transfer channel, a communication path, and the like are not set in advance and a clear handshake is not performed between a transmission source terminal and a destination terminal. Therefore, since packet loss may occur on the network, communication reliability, packet order, and data integrity are not guaranteed. An erasure correction code is applied to ensure such communication reliability.

  The erasure correction code is a technique related to packet loss countermeasures. A data packet transmitted by a transmission source terminal is divided, a redundant data packet is added and transmitted, and when a packet is lost, another data packet or A lost packet can be recovered from a redundant data packet.

  In conventional packet communication, redundancy is individually defined in each layer for a plurality of redundancy (for example, transmission delay, transmission rate, channel frequency, communication path, etc.). Is controlled. For example, technologies in each layer individually control redundancy, such as a transport layer erasure correction code, a network layer routing protocol, a data link layer link aggregation, and a physical layer error correction code.

  For example, Patent Document 1 discloses a technique for distributing and transmitting encoded packets to a plurality of systems and controlling the encoding rate, but is not intended for packet length and transfer path control. Redundancy control is not performed.

JP 2009-278162 A

  However, redundancy regarding redundancy is individually defined in each layer, and if it is attempted to control each redundancy with each layer technology, redundant use of the entire network may be wasted.

  In addition, in order to optimally control multiple redundancy, it is necessary to grasp the network status by periodically sending monitoring packets between the source terminal and the destination terminal, which increases the load on the network. Leads to.

  Therefore, the present invention controls and determines the redundancy of the entire network according to network information by simultaneously using a plurality of redundancy on the network in a communication system using an erasure correction code on a wireless or wired network. Thus, the object is to improve the error rate, data transmission rate, and delay time.

  In order to solve such a problem, a communication control apparatus according to the first aspect of the present invention provides a communication control apparatus that communicates an encoded packet subjected to erasure correction encoding to a destination terminal via a network. Erasure correction encoding means for dividing a packet including a communication signal to be transmitted to a terminal and performing erasure correction encoding processing on each packet; (2) network information storage means for storing network information; and (3) erasure correction code. And a redundancy control unit that comprehensively and sequentially controls each of a plurality of redundancy levels related to transmission of each encoded packet subjected to encoding.

  A communication control program according to a second aspect of the present invention is a computer provided with network information storage means for storing network information in a communication control program for communicating an encoded packet subjected to erasure correction encoding to a destination terminal via a network. (1) erasure correction coding means for dividing a packet including a communication signal to be transmitted to a destination terminal and performing erasure correction coding processing on each packet; and (2) each coded packet subjected to erasure correction coding. Each of a plurality of redundancy levels related to the transmission is made to function as a redundancy level control means for comprehensively and sequentially controlling according to network information.

  A communication control method according to a third aspect of the present invention is a communication control method for communicating a coded packet subjected to erasure correction coding to a destination terminal via a network. (1) The erasure correction coding means is a destination terminal. The packet including the communication signal to be transmitted is divided and the erasure correction encoding process is performed on each packet. (2) The network storage unit stores the network information. (3) The redundancy control unit stores the erasure correction code. Each of a plurality of redundancy levels related to transmission of each encoded packet subjected to encoding is comprehensively and sequentially controlled according to network information.

  According to the present invention, in a communication system using an erasure correction code on a wireless or wired network, the redundancy of the entire network is controlled and determined according to network information by simultaneously using a plurality of redundancy on the network. Thus, the error rate, data transmission rate, and delay time can be improved as compared with the case where the redundancy is individually controlled in each layer.

It is an internal block diagram which shows the internal structure of the transmission source terminal which concerns on embodiment. It is a whole lineblock diagram showing the whole communication system composition concerning an embodiment. It is an internal block diagram which shows the internal structure of the destination terminal which concerns on embodiment. It is a flowchart which shows the communication control process in the transmission source terminal which concerns on embodiment. It is a block diagram which shows the structural example of the flame | frame which the transmission source terminal which concerns on embodiment transmits. It is a block diagram which shows the structural example of the flame | frame which the destination terminal which concerns on embodiment transmits. It is explanatory drawing explaining the determination method of the transfer ratio and transfer route which concern on embodiment. It is a relationship figure which shows the relationship between the loss | disappearance probability after decoding which concerns on embodiment, and redundancy. It is explanatory drawing explaining the disappearance number estimation process by the disappearance number estimation part which concerns on embodiment. It is explanatory drawing explaining the modification of the loss | disappearance number estimation process by the loss | disappearance number estimation part which concerns on embodiment. It is explanatory drawing explaining the determination process of the decoding error rate which concerns on embodiment. It is explanatory drawing explaining an example of the adjustment method of the path | route ratio by the transfer path | route determination part which concerns on embodiment. It is explanatory drawing explaining the data regarding the decoding delay which concerns on embodiment. It is a flowchart which shows the communication control process in the transmission source terminal which concerns on deformation | transformation embodiment.

(A) Embodiments Hereinafter, embodiments of a communication control device, a communication control program, and a communication control method according to the present invention will be described in detail with reference to the drawings.

(A-1) Configuration of Embodiment [Configuration of Communication System]
FIG. 2 is an overall configuration diagram illustrating an overall configuration of the communication system according to the embodiment.

  As illustrated in FIG. 2, the communication system NT according to the embodiment includes a transmission source terminal 1 (1-1 to 1-n; n is an integer) that performs packet communication via a plurality of relay devices 3, and a destination terminal 2. (2-1 to 2-m; m is an integer).

  In FIG. 2, for ease of explanation, the source terminal 1 and the destination terminal 2 will be described separately, but in reality, both the source terminal 1 and the destination terminal 2 are various types of source terminals 1 to be described later. Functions and various functions of the destination terminal 2 are provided.

  The communication system NT is a network that employs UDP communication represented by the Internet, for example. Further, in the communication system NT, in order to ensure the reliability of the UDP communication, a loss correction coding technique is adopted so that a lost packet during packet transfer can be restored. The communication line of the communication system NT may be a wired line or may include a wireless line.

  The transmission source terminal 1 is a transmission terminal that transmits a data packet including transmission data (hereinafter simply referred to as “packet”) to the destination terminal 2. The transmission data is not particularly limited, and examples thereof include video, sound (including sound and sound), data signal, and the like. For example, a server, a personal computer, a smartphone, a tablet terminal, a wearable terminal, other portable terminals, or the like can be applied to the transmission source terminal 1.

  Further, the transmission source terminal 1 divides the packet into a plurality of packets, assigns redundant packets using the erasure correction coding technique, and transmits the packets to the destination terminal 2. Further, the transmission source terminal 1 collects and accumulates network information used in past packet communication from the destination terminal 2 and the relay device 3, and uses the accumulated network information to control the redundancy of the entire network. It has the function to do. Here, as the network information, information including part or all of information relating to packet loss, transmission buffer usage rate, traffic type, QoS, and the like is applied. The packet loss information is information indicating the loss of the encoded packet for each path transmitted from the transmission source terminal 1 to the destination terminal 2 via the network. For example, the packet loss rate is exemplified. In this embodiment, the terms “number of lost packets (number of lost packets)” and “packet loss rate” are used in order to easily explain information related to packet loss. A detailed description of processing for controlling redundancy using network information will be given in the section “(A-2) Operation of the embodiment”.

  The destination terminal 2 is a receiving terminal that receives a packet including transmission data from the transmission source terminal 1. As the destination terminal 2, for example, a personal computer, a smartphone, a tablet terminal, a wearable terminal, other portable terminals, or the like can be applied. The destination terminal 2 has a function of analyzing the received packet and performing lost correction decoding using a redundant packet using the lost correction code to restore the lost packet when packet loss has occurred. In addition, the destination terminal 2 obtains a packet loss rate for each route until the packet is relayed and received, and transmits the packet loss rate for each route to the transmission source terminal 1. Thereby, the packet loss rate for each route can be notified to the transmission source terminal 1 as one piece of network information.

  The relay device 3 is a router device, a switch device, an access point, or the like that constitutes a wireless or wired network, and relays packets exchanged between the transmission source terminal 1 and the destination terminal 2.

[Internal configuration of source terminal 1]
FIG. 1 is an internal configuration diagram illustrating an internal configuration of a transmission source terminal 1 according to the embodiment.

  In FIG. 1, a transmission source terminal 1 according to the embodiment includes a transmission / reception unit 10, an erasure correction coding unit 11, a packet processing unit 13, a network information management unit 110, and a redundancy control unit 120.

  The packet processing unit 13 packetizes data supplied (input) from an upper layer (for example, an application layer) (not shown) and supplies the data to the erasure correction encoding unit 11. The packet processing unit 13 is notified of the packet length determined by the redundancy control unit 120 and divides it into packet lengths to be subjected to erasure correction encoding.

  The erasure correction encoding unit 11 is notified of the encoding rate which is one of the encoding parameters determined by the redundancy control unit 120, and uses the packet acquired from the packet processing unit 13 according to the encoding rate. Erasure correction coding is performed. There are various methods for erasure correction coding. In this embodiment, the erasure correction encoding method is not particularly limited, and various methods can be applied. As a method of erasure correction coding, for example, there is a method of generating a redundant packet (also referred to as “parity packet”) from a packet and adding this parity packet after the packet. In this case, for example, the number of packets Is K, and the number of parity packets is P, the coding rate R is represented by R = K / (K + P). Further, for example, there is a method in which the original divided packet is also a target of encoding and is not distinguished as a parity packet, and is generated as an encoded packet. In this case, for example, the number of original divided packets is K, When the number is N, the coding rate R is represented by R = K / N. In addition, in this embodiment, the case where the former is employ | adopted is illustrated for convenience of explanation.

  The transmission / reception unit 10 adds a header to the packet acquired from the erasure correction encoding unit 11 or a redundant packet (parity packet), and transmits a transmission packet. The transmission / reception unit 10 receives a packet including a packet loss rate, a traffic type, QoS, and the like for each route as network information from the destination terminal 2 and the relay device 3, and gives the network information to the network information management unit 110. is there.

  The network information management unit 110 collects network information with the destination terminal 2 through the transmission / reception unit 10 and accumulates the network information. As shown in FIG. 1, the network information management unit 110 includes a network information acquisition unit 111 that acquires network information, and a network information storage unit 112 that stores the acquired network information.

  Note that the network information management unit 110 may perform statistical processing using the acquired network information and accumulate the result as network information. For example, as will be described later, when a packet loss rate for a predetermined time is acquired from the destination terminal 2, the average value of the packet loss rates is calculated using the past packet loss rate and the latest packet loss rate for a predetermined time. The average packet loss rate may be obtained.

  Further, for example, when the transmission buffer usage rate of the relay device 3 is acquired as network information, it is not limited to storing only the transmission buffer usage rate of the relay device 3 at that time. The transmission buffer usage rate of 3 may also be used to determine the rate of change of the usage rate of the transmission buffer. In other words, it is not limited to a value at a certain time point, but a temporal change rate (change value) may be accumulated as network information.

  The redundancy control unit 120 controls network overall redundancy using network information stored in the network information storage unit 112 of the network information management unit 110.

  Here, the redundancy of the entire network is an extra amount added to check data other than the transmission data so that the data transmitted from the transmission source terminal 1 to the destination terminal 2 can be correctly received by the destination terminal 2. Information. Specifically, the coding rate related to the erasure correction code, the packet length to which the erasure correction encoding is performed, the route to which the packet subjected to the erasure correction encoding is transmitted to the destination terminal 2 via the plurality of relay devices 3 The transfer rate of the route is included.

  The redundancy control unit 120 does not individually control the redundancy related to redundancy defined in each hierarchy in the past, but refers to the accumulated past network information, and transmits each of the information from the transmission source terminal 1 to the destination terminal 2. Estimate the number of discarded packets of the relay device 3 to determine the route, determine the packet length and coding rate related to erasure correction coding from the packet loss rate when transferred by that route, The number of lost packets is estimated, the transmission delay and the transmission rate are determined, and the route that satisfies the request, the encoding parameter, the transmission delay, and the transmission rate are appropriately adapted. In other words, by referring to the past network information, the redundancy related to redundancy defined in each layer is sequentially optimized.

  The redundancy control unit 120 includes a transfer path determination unit 121, a coding rate control unit 122, a packet length control unit 123, an erasure number estimation unit 124, a transmission delay control unit 125, and a transmission rate control unit 126. Detailed descriptions of the processes of the transfer path determination unit 121, the coding rate control unit 122, the packet length control unit 123, the erasure number estimation unit 124, the transmission delay control unit 125, and the transmission rate control unit 126 are described in “(A-2 ) Operation of the embodiment ".

  As the hardware of the redundancy control unit 120, for example, a device having a CPU, ROM, RAM, EEPROM, input / output interface, and the like can be applied, and the CPU stores a processing program (for example, communication control) stored in the ROM. Various functions of the redundancy control unit 120 are realized by executing the program and the like. The communication control program according to the embodiment may be constructed by being installed. Even in this case, the communication control program includes the components illustrated in FIG.

[Internal configuration of destination terminal 2]
FIG. 3 is an internal configuration diagram illustrating an internal configuration of the destination terminal 2 according to the embodiment.

  In FIG. 3, the destination terminal 2 according to the embodiment includes a transmission / reception unit 20, an erasure correction decoding unit 21, a packet processing unit 23, and a network information management unit 210.

  The transmission / reception unit 20 receives a packet transmitted from the transmission source terminal 1 and gives the received packet to the erasure correction decoding unit 21. Further, the transmission / reception unit 20 acquires a packet including the packet loss rate for each path calculated by the packet loss rate calculation unit 213 of the network information management unit 210 described later, and transmits the packet to the transmission source terminal 1.

  The erasure correction decoding unit 21 performs erasure correction decoding processing to restore a lost packet when there is a lost packet for the packet acquired from the transmission / reception unit 20. In order to contribute to the calculation of the packet loss rate, the erasure correction decoding unit 21 gives the number of packet loss to the network information management unit 210. If there is no lost packet, the received packet is given to the packet processing unit 23 as it is.

  The packet processing unit 23 acquires a received packet, extracts data included in the packet, and supplies the extracted data to an upper layer (for example, an application layer). The packet processing unit 23 notifies the network information management unit 210 of route information and traffic type included in the received packet in order to manage the route and traffic type of the received packet. Further, in order to notify the transmission source terminal 1 of the packet loss rate, the packet processing unit 23 acquires the packet loss rate for each route from the network information management unit 210 and packetizes it. The packet including the packet loss rate for each path is given to the transmission / reception unit 20 and transmitted to the transmission source terminal 1.

  The network information management unit 210 manages network information with the transmission source terminal 1. The network information management unit 210 uses a network information acquisition unit 211 that acquires network information, a network information storage unit 212 that stores the acquired network information, and network information stored in the network information storage unit 212. A packet loss rate calculation unit 213 that calculates a packet loss rate for each route.

(A-2) Operation of Embodiment Next, the operation of the communication control process between the transmission source terminal 1 and the destination terminal 2 according to the embodiment will be described in detail with reference to the drawings.

  FIG. 4 is a flowchart showing communication control processing in the transmission source terminal 1 according to the embodiment.

[Network Information Update; Step S300]
In the network information management unit 110 of the transmission source terminal 1, network information on the wireless or wired network is accumulated in the network information accumulation unit 112 by past packet communication (step S300). Each time packet communication is performed between the transmission source terminal 1 and the destination terminal 2, new network information is sequentially accumulated and updated.

  Here, an example of a method for acquiring a packet loss rate, which is one piece of network information, is illustrated.

  FIG. 5 is a configuration diagram illustrating a configuration example of a frame transmitted from the transmission source terminal 1 according to the embodiment, and FIG. 6 is a configuration diagram illustrating a configuration example of a frame transmitted from the destination terminal 2.

  In FIGS. 5 and 6, a packet subjected to erasure correction coding (hereinafter also referred to as “encoded packet”) is transmitted from a plurality of relay apparatuses (herein referred to as “routers”) 3 via a transmission source terminal. An example of transmission from 1 to the destination terminal 2 is illustrated. Further, it is assumed that the encoded packet is transmitted through a route of transmission source terminal → router 3-1 → router 3-2 → router 3-3 → router 3-6 → destination terminal 2.

  As illustrated in FIG. 5, the data frame included in the packet transmitted from the transmission source terminal 1 includes the total number (the number of transmissions) of encoded packets transmitted from the transmission source terminal 1. Specifically, the source terminal 1 stores the total number of encoded packets (the number of transmissions) in the UDP payload of the UDP segment frame in the Ether (registered trademark) frame and transmits it. The total number of encoded packets is not particularly limited as long as it is within the UDP payload. For example, it is stored in the leading area of several bits of the UDP payload. That is, if the total number of encoded packets transmitted by the transmission source terminal 1 is “100”, information indicating “100” is stored.

  In the destination terminal 2, the total number of encoded packets stored in the UDP payload is accumulated in the network information accumulation unit 212. The destination terminal 2 also accumulates the number of received packets received within a predetermined time for each path in order to recognize the number of packets actually received. Therefore, in the network information management unit 210 of the destination terminal 2, the packet loss rate calculation unit 213 determines the packet loss rate for each path (= number of received packets / total number of encoded packets) based on the number of received packets and the total number of encoded packets. ) Is calculated.

  In order to notify the transmission source terminal 1 of the packet loss rate for each route, the destination terminal 2 transmits a packet including the packet loss rate for each route to the transmission source terminal 1 as shown in FIG. Specifically, as illustrated in FIG. 6, the destination terminal 2 stores and transmits the packet loss rate of the target path in the UDP payload of the UDP segment frame in the Ether (registered trademark) frame. For example, when the total number of encoded packets is “100” and the number of received packets at the destination terminal 2 is “95”, the number of lost packets is “5” and the packet loss rate is “0.05”. . Accordingly, in this case, route information for specifying the target route and information indicating the packet loss rate “0.05” of the route are stored in the UDP payload and transmitted to the transmission source terminal 1.

  Here, since the destination terminal 2 does not have a large amount of information indicating the packet loss rate of the target route, the destination terminal 2 sends the route information for specifying the target route and the packet loss rate of the target route to the transmission source terminal 1 every predetermined time. You may make it transmit. However, since the packet including the packet loss rate is sequentially notified to the transmission source terminal 1, the number of packets exchanged on the network may increase, and the traffic on the network may also increase. When the packet loss is high (that is, when the number of lost packets or the packet loss rate is equal to or greater than the threshold), the packet loss rate of the target route may be notified to the transmission source terminal 1.

  As described above, the transmission source terminal 1 can acquire information on the packet loss rate for each route, and store the packet loss rate of the target route in the network information storage unit 112.

  Further, since the network information management unit 110 can acquire the packet loss rate for each predetermined time for each route, the average packet loss rate may be calculated and accumulated for each route using the past packet loss rate. Good.

[Transfer route / ratio selection; Step S301]
In the transmission source terminal 1, the transfer path determination unit 121 of the redundancy control unit 120 obtains the transfer rate between the routers 3 based on the packet loss rate between the routers 3 stored in the network information storage unit 112. Then, referring to the transfer ratio, the transfer route to the destination terminal 2 is determined (step S301). That is, the transfer route determination unit 121 calculates the total value of the packet loss rates between the routers that pass through, and selects a route that minimizes the total value (total route loss value).

  The router 3 temporarily stores received packets in a buffer, and transmits the packets stored in order at each transmission timing to the next router 3 or terminal. Since the router 3 cannot transfer the packet due to traffic conditions and the like, the time-out packet is discarded. Therefore, the packet loss rate between the routers 3 affects the coding rate of the erasure correction code, the transmission delay, and the like. Therefore, the transfer route determination unit 121 selects a route with a small packet loss rate based on the packet loss rate (or average packet loss rate) between the routers 3. Information on the packet loss rate between the routers 3 may be acquired from each router 3 or a network management server (not shown), or the packet loss rate of the route notified from the destination terminal 2 is used. You may make it ask.

  FIG. 7 is an explanatory diagram illustrating a method for determining a transfer ratio and a transfer path according to the embodiment.

  Here, for ease of explanation, the router 3-2 and the router 3-4, which are branch paths of the router 3-1, will be described as an example. The packet loss rate between the router 3-1 and the router 3-2 is 0.00 [%], and the packet loss rate between the router 3-1 and the router 3-4 is 0.01 [%]. If there is, the packet loss rate between the router 3-1 and the router 3-2 is low loss. In this case, the transfer ratio of the router 3-1 to the router 3-2 and the transfer to the router 3-4 is “1: 0”. In this way, the packet loss rates between routers passing from the transmission source terminal 1 to the destination terminal 2 are totaled, and a route with the minimum total route loss value is determined. Note that FIG. 7 shows an example in which a single path is the minimum total loss, but multiple paths may be used, and if there is a minimum total loss, transfer is performed at a ratio of the multiple paths. The case may be selected.

  For the initial route determination for the destination terminal 2, for example, only the route with the minimum total route loss value is selected based on the past packet loss rate between the routers 3. Thereafter, when packet communication with the destination terminal 2 is continuously performed and the packet loss rate between the routers 3 is updated, the latest network information (including the packet loss rate between the routers 3) is used. The transfer path may be changed sequentially.

[Selection of packet length and coding rate; steps S302 and S303]
The packet length control unit 123 determines the packet length related to erasure correction coding (step S302), and the coding rate control unit 122 determines the coding rate related to erasure correction coding (step S303).

  FIG. 8 is a relationship diagram illustrating a relationship between the post-decoding erasure probability and the redundancy according to the embodiment.

  FIG. 8 shows the relationship between the error rate and the number of encoded packets for each code related to erasure correction encoding. The vertical axis represents the probability of erasure after decoding (that is, the error rate), and the horizontal axis represents the redundancy (= 1-coding rate). In general, the coding rate and the number of coded packets are in a trade-off relationship. That is, in order to improve the decoding performance (lower the error rate), it is desired to lower the coding rate, but the number of coded packets increases. On the other hand, in order to reduce the number of encoded packets, it is desired to increase the encoding rate, but the decoding performance decreases (the error rate increases). On the other hand, if the packet loss rate is low, even when the coding rate is the same, the shorter the packet length and the greater the number of information packets, the greater the number of received packets and the better the decoding performance.

  Therefore, in packet communication with the destination terminal 2, first, in order to prioritize quality, a short packet length is set and a lower coding rate is set. As the packet length and coding rate for the first communication, a preset packet length and coding rate may be used.

  Note that the number of encoded packets = the number of information packets / the encoding rate can be obtained. In addition, the number of information packets can be calculated as: request rate / packet length.

[Estimation of Number of Packet Loss; Step S304]
Next, the lost number estimation unit 124 estimates the number of lost packets (total lost number) up to the destination terminal 2 based on the packet loss rate for each path (step S304).

  FIG. 9 is an explanatory diagram illustrating the disappearance number estimation process by the disappearance number estimation unit 124 according to the embodiment.

First, the lost number estimation unit 124 calculates the number of lost packets for each router 3 according to the following formula (1), and calculates the total value (total lost number) in each route to the destination terminal 2 at the destination terminal 2. Estimated as the number of disappearances.
Number of lost packets = (number of packets received by the previous router) x (transfer rate) x (packet loss rate) (1)

  For example, in FIG. 9, the loss count estimation unit 124 sets the transfer ratio to the router 3-2 that is the branch path of the router 3-1 and the transfer ratio to the router 3-4 to 0.5: 0.5. In this case, if the packet loss rate is 0.02 when the packet is transferred from the router 3-1 to the router 3-4, the number of lost packets is 1 (= router 3-1) when the packet is transferred to the router 3-4. Of received packets (100) × 0.5 × 0.02).

  As described above, the lost number estimation unit 124 calculates the number of lost packets for each router 3 and adds the number of lost packets of each router 3 that passes through to obtain the total lost number. Then, the number of disappearances at the destination terminal 2 is 2.

  In the first packet communication with the destination terminal 2, the total number of disappearances up to the destination terminal 2 can be obtained according to the equation (1) as described above.

In addition, when packet communication with the destination terminal 2 is continuously performed and the packet loss rate for each route is accumulated from the destination terminal 2, the loss count estimation unit 124 calculates the packet loss rate for each route. By using the equation (2), the total number of disappearances up to the destination terminal 2 may be obtained.
Number of lost packets = (Number of encoded packets from the transmission source terminal) × (Average packet loss rate) (2)

  FIG. 10 is an explanatory diagram illustrating a modified example of the disappearance number estimation process by the disappearance number estimation unit 124 according to the embodiment.

  The loss count estimation process by the loss count estimation unit 124 is not limited to the above, and the packet loss count of each router 3 may be calculated using the transmission buffer usage rate acquired from each router 3. .

For example, the erasure count estimation unit 124 monitors the transmission buffer usage rate of each router 3, and if the transmission buffer usage rate of the router 3 on the route is equal to or greater than a threshold value, the packet passing through the router 3 May be discarded, and the number of lost packets may increase. Therefore, the lost number estimation unit 124 may obtain the number of lost packets of the router 3 using the transmission buffer usage rate of the router 3 according to the following equation (3). In this case, the total number of disappearances is obtained by adding the number of disappearances of each router 3 on the route to the destination terminal 2.
Number of lost packets = (Number of packets received by the previous router) × (Transfer rate) × (Packet loss rate) × (1−Transmission buffer usage rate) (3)

  For example, in the case of the example in FIG. 10, if the threshold for the transmission buffer usage rate is 0.8 and the transmission buffer usage rate of the router 3-4 is 0.8, the number of lost packets is 12.

[Decision of Decoding Error Rate; Step S305]
As described above, the erasure count estimation unit 124 calculates the packet erasure count of each router 3 and estimates the total erasure count up to the destination terminal 2, and the coding rate based on the estimated total erasure count is calculated as a request. It is determined whether the post-decoding erasure probability is satisfied (step S305). If the request is satisfied, the process proceeds to step S308. If the request is not satisfied, the process proceeds to step S306.

[Reselection of coding rate; step S306 => steps S303 and S307]
FIG. 11 is an explanatory diagram illustrating decoding error rate determination processing according to the embodiment.

  FIG. 11 corresponds to the relationship diagram showing the relationship between the post-decoding erasure probability and the redundancy in FIG. Here, description will be made using the relationship between the error rate of code A and the number of encoded packets.

The coding rate control unit 122 obtains the coding rate according to the equation (4) using the estimated total number of erasures of the destination terminal 2.
Coding rate = number of information packets ÷ (number of coded packets−total number of lost terminals 2) (4)

For example, the requested post-decoding erasure probability (error rate) is 1 × 10 ⁻² . Assume that the coding rate (here, the coding rate selected for the first time) in consideration of the estimated total number of erasures in the first packet communication with the destination terminal 2 is around 0.08.

In this case, the post-decoding erasure probability is about 1 × 10 ⁻¹ , and it is determined whether or not the requested post-decoding erasure probability is satisfied as compared with the requested post-decoding erasure probability 1 × 10 ^−2. .

  In this case, since the required post-decoding erasure probability is not satisfied, the coding rate is set low (redundancy on the horizontal axis is increased), and the packet length currently set (ie, In this case, it is determined whether there is an encoding rate candidate that satisfies the request.

  When there is an encoding rate that satisfies the request with the currently set packet length, the process proceeds to step S303, and the encoding rate control unit 122 selects a candidate that satisfies the required post-decoding erasure probability. The coding rate is selected again. Thereafter, the processing after step S303 is repeatedly performed.

  On the other hand, if there is no encoding rate that satisfies the request with the currently set packet length, the process proceeds to step S307.

[Reselection of Packet Length; Step S307⇒Steps S302 and S301]
The packet length control unit 123 determines whether there is a candidate that satisfies the request with the transfer rate on the currently set route (step S307). If the request is satisfied, the process proceeds to step S302. Then, adjust the packet length again. Thereafter, the processing after S302 is repeatedly performed.

  If the request is not satisfied, the process proceeds to step S301, and the transfer route determination unit 121 selects the route to the destination terminal 2 and the transfer rate again. Thereafter, the processing after S301 is repeatedly performed.

  FIG. 12 is an explanatory diagram illustrating an example of a route ratio adjustment method by the transfer route determination unit 121.

  As shown in FIG. 12, when the transfer route determination unit 121 does not satisfy the request, it refers to the network information stored in the network information storage unit 112 and sets the transfer rate again. In this case, For example, the transfer ratio between the routers 3 may be adjusted uniformly. For example, in this example, a case where the transfer ratio between the routers 3 is changed from “1: 0” to “0.8: 0.2” is illustrated.

  As described above, the transfer rate may be changed uniformly, or when the transmission buffer usage rate of each router 3 is taken into consideration, when passing through the router 3 whose transmission buffer usage rate is equal to or higher than the threshold value, You may make it change a setting so that the transfer to the router 3 may decrease.

[Calculation of transmission delay; Step S308]
The transmission delay control unit 125 estimates the transmission delay based on information including the path and transfer rate, the packet length, and the coding rate, which are the selection parameters so far (step S308).

For example, when a packet is transferred to a plurality of routes, the transmission delay control unit 125 calculates the maximum transmission delay in the route according to the following equation (5).
Maximum transmission delay = maximum number of hops x transmission delay + decoding delay (5)

  In Expression (5), the maximum number of hops is the maximum number of hops among a plurality of routes from the transmission source terminal 1 to the destination terminal 2. The transmission delay is a value indicating the transmission delay time between the routers 3, and an arbitrary value such as 1 ms can be set. The decoding delay is a delay time required for decoding the packet subjected to erasure correction coding at the destination terminal 2, and can be obtained using, for example, preset data illustrated in FIG.

  FIG. 13 is an explanatory diagram illustrating data related to the decoding delay according to the embodiment.

  FIG. 13 is an explanatory diagram showing the relationship between the packet length and the delay time required for decoding. In this case, the decoding delay can be obtained based on the packet length targeted for the set erasure correction code. For example, when a packet length of 1024 bytes and an encoding rate (for example, 1024 corresponding encoded packets) are selected, the decoding delay is about 500 ms, and this value is substituted into the decoding delay of Equation (5).

[Determination of Transmission Delay; Step S309]
The transmission delay control unit 125 determines whether or not the estimated maximum transmission delay value satisfies a requested transmission delay (hereinafter referred to as “requested transmission delay”) (step S309). If the requested transmission delay is satisfied, the process proceeds to step S310. If not satisfied, the process proceeds to step S307.

  When the request transmission delay is not satisfied, the process proceeds to step S307, and the packet length control unit 123 increases the packet length so as to satisfy the request transmission delay (step S302). That is, the number of encoded packets is reduced to reduce delay.

  If the requested transmission delay is still not satisfied, the process proceeds to step S301, where the transfer route determination unit 121 selects a route with a low hop count, or a route through the router 3 with a low transmission buffer usage rate. Or select.

[Calculation of transmission rate; Step S310]
The transmission rate control unit 126 estimates the transmission rate based on the information including the path and transfer rate, the packet length, the coding rate, and the maximum transmission delay, which are the selection parameters so far (step S310).

More specifically, the transmission rate control unit 126 uses the maximum transmission delay calculated by the transmission delay control unit 125 to calculate the transmission rate according to Equation (6).
Transmission rate [bit / sec] = packet length × number of encoded packets ÷ maximum transmission delay (6)

[Transmission rate determination; Step S311]
The transmission rate control unit 126 determines whether or not the calculated transmission rate satisfies the requested transmission rate (step S311). If the requested transmission rate is satisfied, the process ends. If not, the process proceeds to step S307.

  When the requested transmission rate is not satisfied, the process proceeds to step S307, and the packet length control unit 123 increases the packet length so as to satisfy the requested transmission rate (step S302). That is, the number of encoded packets is reduced to reduce delay.

  If the requested transmission rate is still not satisfied, the process proceeds to step S301, where the transfer route determination unit 121 selects a route with a low hop count, or selects a router 3 with a low transmission buffer usage rate. Select a route to go through.

(A-3) Effect of Embodiment As described above, according to this embodiment, network information (packet loss rate, etc.) collected and stored on the destination side using past packet communication is transmitted as necessary. Notifying the original network while reducing the load on the network, and the overall network redundancy according to the notified network information, traffic type and QoS (encoding packet coding rate, length, transfer path and ratio) Can be controlled and determined.

(B) Other Embodiments In the above-described embodiments, the modified embodiments of the present invention are mentioned, but the present invention can also be applied to the following modified embodiments.

(B-1) Modified Embodiment 1
FIG. 14 is a flowchart showing a communication control process in the transmission source terminal 1 according to a modified embodiment.

  In FIG. 14, the same processes as those in FIG. 4 of the above-described embodiment are denoted by the same reference numerals.

  In the above-described embodiment, in order to give priority to the data quality at the time of erasure correction decoding, the case where the initial packet length and the coding rate are set to be low has been exemplified. However, the decoding delay may be reduced, that is, the decoding delay may be prioritized. In this case, in order to reduce the number of encoded packets, the packet length may be set longer.

  For example, in step S301 of FIG. 14, the transfer route determination unit 121 selects the shortest route that minimizes the number of hops. In step S302, the packet length control unit 123 sets a longer packet length in the first packet communication with the destination terminal 2 in order to reduce the number of received packets and reduce the decoding delay. In step S303, in order to reduce the number of encoded packets and reduce the decoding delay, the encoding rate is initially set higher.

  Similar to the above-described embodiment, the transmission delay control unit 125 calculates the maximum transmission delay (step S308), determines whether the maximum transmission delay satisfies the request (step S309), and satisfies the request. If yes, the process proceeds to step S304. If not, the process proceeds to step S306.

  Here, in step S306, it is determined whether it is possible to set a higher coding rate than the current one (whether there is a candidate). If there is a candidate, the coding rate control unit 122 sets a coding rate higher than the current rate (step S303), and repeatedly performs the processing after step S303. On the other hand, if there is no candidate, the process proceeds to step S307, and it is determined whether or not the packet length can be set longer than the current one (whether there is a candidate). If there is a candidate, the packet length control unit 123 sets a packet length longer than the current length (step S302), and repeatedly executes the processing from step S302.

  As a result of the above iterative processing, if the requested transmission delay is still not satisfied and there is no candidate in step S307, the transfer route determination unit 121 changes the transfer rate to the route with a low buffer usage rate step by step (step S301). , S301 and the subsequent processes are repeated.

  In step S304, as in the above-described embodiment, the loss count estimation unit 124 estimates the destination packet loss count from the packet loss rate and the transmission buffer usage rate, and takes the estimated packet loss count of the destination terminal 2 into account. (Step S305). If the required quality is satisfied, the process proceeds to step S310.

  On the other hand, when the required quality is not satisfied, the process proceeds to step S306, and the coding rate control unit 122 adjusts the coding rate to be lowered (step S306⇒step S303), as in the above-described embodiment. ), The packet length control unit 123 adjusts the packet length to be shortened (step S307 → step S302), and the transfer path determination unit 121 adjusts the transfer rate to the low-loss path so as to increase stepwise (step S307 = step S302). Step S307 => Step S301).

  In steps S310 and S311, the transmission rate control unit 126 may calculate the transmission rate and process whether or not the required rate is satisfied, as in the above-described embodiment.

(B-2) Modified Embodiment 2
About the processing of the redundancy control unit 120 in the above-described embodiments and modified embodiments, using the machine learning algorithm, the coding rate of the erasure correction code, the packet length, the transfer path, and the optimal part or all of the transfer rate You may make it implement a conversion process.

  1 (1-1 to 1-n): transmission source terminal, 2 (2-1 to 2-m): destination terminal, 3 ... relay device, NT: communication system, 10: transmission / reception unit, 11: erasure correction coding , 13 ... Packet processing unit, 110 ... Network information management unit, 111 ... Network information acquisition unit, 112 ... Network information storage unit, 120 ... Redundancy control unit, 121 ... Transfer path determination unit, 122 ... Coding rate control unit , 123 ... packet length control unit, 124 ... erasure number estimation unit, 125 ... transmission delay calculation unit, 126 ... transmission rate control unit, 20 ... transmission / reception unit, 21 ... erasure correction decoding unit, 23 ... packet processing unit, 210 ... network Information management unit 211... Network information acquisition unit 212. Network information storage unit 213. Packet loss rate calculation unit

Claims (10)

In a communication control apparatus that communicates an encoded packet subjected to erasure correction encoding to a destination terminal via a network,
Erasure correction coding means for dividing a packet including a communication signal to be transmitted to the destination terminal and performing erasure correction coding processing on each packet;
Network information storage means for storing network information of the network;
And a redundancy control unit that comprehensively and sequentially controls each of a plurality of redundancy related to transmission of each encoded packet subjected to erasure correction encoding, according to the network information. .   The redundancy control means estimates the number of lost encoded packets transmitted through the route to the destination terminal and the decoding delay time at the destination terminal, estimated using the network information, the required quality and the required transmission delay. 2. The communication control device according to claim 1, wherein each of the plurality of redundancy levels is controlled by determining whether or not the above is satisfied.   The redundancy control means determines whether or not a transmission rate satisfies a required transmission rate through a route to the destination terminal calculated using the network information, and controls each of the plurality of redundancy The communication control apparatus according to claim 2.   The redundancy control means controls any or all of a packet length, a coding rate, a transfer path, and a branch transfer ratio of the path related to the erasure correction coding according to the result of each determination. The communication control device according to claim 2 or 3.   The communication control apparatus according to claim 2, wherein the redundancy control unit estimates the number of lost encoded packets based on a packet loss rate at the destination terminal included in the network information.   3. The communication control according to claim 2, wherein the redundancy control means estimates the number of lost encoded packets based on a transmission buffer usage rate of a relay device on a path included in the network information. apparatus.   4. The communication control apparatus according to claim 2, wherein the redundancy control means determines each of the plurality of redundancy based on a traffic type and / or QoS of request data included in the network information.   The communication control apparatus according to claim 1, wherein the redundancy control unit optimizes each of the plurality of redundancy using a machine learning algorithm. In a communication control program for communicating an encoded packet subjected to erasure correction encoding to a destination terminal via a network,
A computer comprising network information storage means for storing network information of the network,
Erasure correction coding means for dividing a packet including a communication signal to be transmitted to the destination terminal and performing erasure correction coding processing on each packet;
A communication characterized by causing each of a plurality of redundancy related to transmission of each encoded packet subjected to erasure correction encoding to function as redundancy control means for comprehensively and sequentially controlling according to the network information. Control program. In a communication control method for communicating an encoded packet subjected to erasure correction encoding to a destination terminal via a network,
An erasure correction encoding unit divides a packet including a communication signal to be transmitted to the destination terminal and performs an erasure correction encoding process on each packet.
Network storage means stores network information of the network,
A communication control method, wherein the redundancy control means comprehensively and sequentially controls each of a plurality of redundancy related to transmission of each encoded packet subjected to erasure correction encoding according to the network information.

Images