JP4679495B2 - COMMUNICATION CONTROL DEVICE, RADIO COMMUNICATION DEVICE, COMMUNICATION CONTROL METHOD, AND RADIO COMMUNICATION METHOD - Google Patents

Description

  According to the present invention, a plurality of different wireless communication paths can be used, and a bandwidth that is insufficient in one wireless communication path is compared to a required bandwidth of an application having real-time characteristics (hereinafter, abbreviated as an application as appropriate). The present invention relates to a novel communication control device, a wireless communication device, a communication control method, and a wireless communication method that can complement and execute communication using other wireless communication paths.

  For example, in a wireless communication network using an Internet protocol (IP) group (hereinafter, appropriately abbreviated as “wireless IP network”), so-called mobile IP is defined in order to improve mobility of a wireless communication device (for example, Non-patent document 1).

In the mobile IP, a care-of IP address (Care of Address) that is dynamically assigned according to the position of the wireless communication device is used.
C. Perkins, “IP Mobility Support (RFC2002)”, [online], October 1996, IETF, [March 15, 2006 search], Internet <URL: http: //www.ietf.org / rfc /rfc2002.txt

  By the way, recently, an environment in which a wireless communication apparatus can use a plurality of wireless IP networks (for example, a mobile phone network and a wireless LAN network) is being provided.

  If such an environment is provided, it is conceivable to perform communication using a plurality of wireless IP networks at the same time so as to supplement the bandwidth shortage of one wireless IP network with another wireless IP network.

  However, when one wireless IP network is used as a master route, another wireless IP network is used as a slave route, and the bandwidth that is insufficient in the master route is supplemented by the slave route, the allowable bandwidth in the slave route satisfies the bandwidth that needs to be supplemented. Otherwise, the packets corresponding to the supplementary bandwidth to be transmitted to the slave path are temporarily stored in a buffer existing on the path and sequentially transmitted because the bandwidth is insufficient. For this reason, on the receiving side, a delay occurs between the packet received via the master path and the packet received via the slave path. This delay appears as an overtaking of SN (sequence number) in VoIP packets.

  In real-time applications such as VoIP, a jitter buffer is provided to absorb fluctuations such as packet overtaking in the network. However, if the delay between paths increases and exceeds the buffer capacity of the jitter buffer, the capacity will be exceeded. If the packet is delayed, it is discarded even if it is received. For this reason, if the transmission of packets exceeding the allowable bandwidth is continued, the amount of packets accumulated in the buffer existing on the path will increase (up to the limit amount of the buffer), resulting in a delay difference corresponding to the accumulation amount. Eventually, all the packets sent out as a complement will be discarded at the receiving end.

  In particular, in the wireless communication path, the allowable bandwidth depends on a change in the propagation environment such as fading. Therefore, if a packet is transmitted in the same way as when the propagation environment is bad and normal, a delay difference is generated.

  On the other hand, the jitter buffer capacity in a real-time application such as VoIP can be set according to only the route because the packet is delayed depending on only that route when communication is performed using only one route. .

  However, if there are two or more routes with different wireless communication methods, simply select multiple routes, one of which is the master route, the other is the slave route, and the slave route supplements the bandwidth shortage in the master route. When trying to do so, the propagation characteristics of each path are different, so that a slave path having a delay characteristic larger than that of the master path may be selected, or a slave path having a delay characteristic much smaller than that of the master path may be selected.

  For this reason, when a slave path having a delay characteristic larger than that of the master path is selected, a packet that has arrived via the slave path has a delay equal to or greater than that of the jitter buffer, and is discarded even if it arrives. When a slave route with a higher delay characteristic than the master route is selected, the packet arrives earlier than the buffer capacity of the jitter buffer and is discarded as an overflow. There is a concern that the possibility is high.

  Accordingly, an object of the present invention made in view of such a point is to select a plurality of different wireless communication paths that are most suitable for an application, and perform communication by complementing a band that is insufficient in one wireless communication path with another wireless communication path. The present invention provides a communication control device, a wireless communication device, a communication control method, and a wireless communication method that can effectively reduce the amount of packets that have been discarded.

The invention according to claim 1, which achieves the above object, selects a plurality of different wireless communication paths with a wireless communication apparatus and uses one wireless communication path for a requested bandwidth of an application having real-time characteristics. A communication control device for controlling the communication between the wireless communication device and the communication destination by supplementing the insufficient bandwidth with another wireless communication path,
Based on a packet received via each wireless communication path, calculating means for calculating the ideal reception timing and allowable limit timing of the received packet in the wireless communication path;
Storage means for receiving and storing fluctuation absorption time of a received packet in the application transmitted from the wireless communication device;
In order that the fluctuation request time calculated based on the ideal reception timing and the allowable limit timing in each wireless communication path calculated by the calculation means does not exceed the fluctuation absorption time in the application stored in the storage means Selecting means for selecting different wireless communication paths;
It is characterized by having.

The invention according to claim 2 is the communication control apparatus according to claim 1,
The selecting means is configured to select, as the one wireless communication path, a wireless communication path with the latest allowable limit timing calculated by the calculating means from among the plurality of different wireless communication paths to be selected. is there.

The invention according to claim 3 is the communication control apparatus according to claim 1 or 2,
The selection unit subtracts the earliest ideal reception timing from the latest allowable limit timing calculated by the calculation unit among the plurality of different wireless communication paths to be selected, and obtains the fluctuation request time. Is.

Furthermore, in the invention according to claim 4 that achieves the above object, a plurality of different wireless communication paths are selected with the communication control device, and a single wireless is used for a requested bandwidth of an application having real-time characteristics. A wireless communication device that complements a bandwidth that is insufficient in a communication path with another wireless communication route and wirelessly communicates with a communication destination via the communication control device,
Based on a packet received via each wireless communication path, calculating means for calculating an ideal reception timing and an allowable limit timing of a received packet in the wireless communication path;
Storage means for storing fluctuation absorption time of received packets in the application;
The fluctuation request time calculated based on the ideal reception timing and the allowable limit timing in each wireless communication path calculated by the calculation means so as not to exceed the fluctuation absorption time in the application stored in the storage means. Selecting means for selecting the plurality of different wireless communication paths;
It is characterized by having.

The invention according to claim 5 is the wireless communication apparatus according to claim 4,
The selecting means is configured to select, as the one wireless communication path, a wireless communication path with the latest allowable limit timing calculated by the calculating means from among the plurality of different wireless communication paths to be selected. is there.

The invention according to claim 6 is the wireless communication device according to claim 4 or 5,
The selection unit subtracts the earliest ideal reception timing from the latest allowable limit timing calculated by the calculation unit among the plurality of different wireless communication paths to be selected, and obtains the fluctuation request time. Is.

Furthermore, the invention according to claim 7 that achieves the above object is to select a plurality of different wireless communication paths with a wireless communication apparatus and use one wireless for a required bandwidth of an application having real-time characteristics. A communication control method for controlling a communication between the wireless communication apparatus and a communication destination by supplementing a bandwidth that is insufficient in a communication path with another wireless communication path,
Calculating ideal reception timing and allowable limit timing of a received packet in the wireless communication path based on a packet received via each wireless communication path;
Receiving and storing fluctuation absorption time of a received packet in the application transmitted from the wireless communication device; and
The plurality of different wireless communication paths are set such that the fluctuation request time calculated based on the ideal reception timing and allowable limit timing in each of the calculated wireless communication paths does not exceed the fluctuation absorption time in the stored application. A step to choose;
It is characterized by having.

Furthermore, the invention according to claim 8 that achieves the above object is to select a plurality of different wireless communication paths with the communication control apparatus and use one wireless for a requested bandwidth of an application having real-time characteristics. A wireless communication method for supplementing a bandwidth that is insufficient in a communication path with another wireless communication path and performing wireless communication with a communication destination via the communication control device,
Calculating ideal reception timing and allowable limit timing of a received packet in the wireless communication path based on a packet received via each wireless communication path;
The plurality of different wireless communication paths so that the fluctuation request time calculated based on the ideal reception timing and allowable limit timing in each of the calculated wireless communication paths does not exceed the fluctuation absorption time in the preset application. A step of selecting
It is characterized by having.

  According to the present invention, a plurality of different wireless communication paths that are optimal for an application are selected so that the fluctuation request time calculated based on the allowable limit timing in each wireless communication path does not exceed the fluctuation absorption time in the application. , A communication control device, a wireless communication device, a communication control method, and a communication control device capable of performing communication by complementing a shortage of bandwidth in one wireless communication route with another wireless communication route, and efficiently reducing discard of packets that have arrived A wireless communication method can be provided.

  Next, an embodiment of the present invention will be described with reference to the drawings.

(Overall schematic configuration of communication system)
FIG. 1 is an overall schematic configuration diagram of a communication system 1 according to the present embodiment. As shown in FIG. 1, the communication system 1 includes a wireless IP network 10A and a wireless IP network 10B as a plurality of different wireless communication paths that can be used by a mobile node 300 (hereinafter abbreviated as MN 300) that is a wireless communication device. Is included. The wireless IP network 10A is an IP network that can transmit IP packets. In the wireless IP network 10A, the care-of IP address A1 is dynamically assigned to the MN 300 according to the position of the MN 300. The wireless IP network 10A is a mobile phone network that uses, for example, CDMA (specifically, HRPD, which is a 3GPP2 standard) as a wireless communication system.

  The wireless IP network 10B can transmit IP packets in the same manner as the wireless IP network 10A. In the wireless IP network 10B, the care-of IP address A2 is assigned to the MN 300. The wireless IP network 10B is, for example, a mobile WiMAX that conforms to the specification of IEEE 802.16e as a wireless communication method.

  The care-of IP address A1 is given from the wireless IP network 10A when the MN 300 is connected to the wireless IP network 10A. Similarly, the care-of IP address A2 is given from the wireless IP network 10B when the MN 300 is connected to the wireless IP network 10B. The care-of IP address A1 and the care-of IP address A2 are associated with the home IP address AH (virtual address).

  In FIG. 1, for simplification of the drawing, two wireless IP networks 10A and 10B are shown, but different wireless communication paths that can be used by the MN 300 are the two wireless IP networks 10A and 10A. Not only the wireless IP network 10B but also more wireless IP networks may be available.

  A plurality of wireless IP networks including the wireless IP network 10 </ b> A and the wireless IP network 10 </ b> B are connected to the Internet 20. The Internet 20 is connected to a switching server 100 that constitutes a communication control device that controls a wireless communication path with the MN 300.

  The switching server 100 has a VPN router function for performing routing processing of IP packets, and by establishing a tunnel by VPN (IPSec) between the MN 300 and the switching server 100, virtualization of the OSI third layer To secure the IP mobility of the MN 300.

  That is, in this embodiment, unlike mobile IP (for example, RFC2002), the MN 300 selects a plurality of wireless IP networks, for example, the wireless IP network 10A and the wireless IP network 10B, according to the application, and the selected One of the plurality of wireless IP networks, for example, the wireless IP network 10A is used as a master route, and the remaining wireless IP network 10B is used as a slave route. Communication with the communication destination (specifically, the IP telephone terminal 40) is executed while complementing the bandwidth that is deficient in the master route with respect to the requested bandwidth with the slave route.

  The switching server 100 is connected to the IP telephone terminal 40 via a communication network 10 </ b> C connected to the Internet 20. The IP telephone terminal 40 mutually converts voice signals and VoIP packets, and transmits / receives IP packets.

  Specifically, the switching server 100 (communication control device) uses a plurality of different wireless IP networks selected by the MN 300 (wireless communication device) toward the IP telephone terminal 40 (communication destination) for a predetermined period ( The IP packet (VoIP packet) transmitted while complementing at 20 ms) is received and relayed to the IP telephone terminal 40, and the IP packet (VoIP) transmitted by the IP telephone terminal 40 to the MN 300 at a predetermined cycle (20 ms). Packet) and relay to MN 300 while complementing using a plurality of different selected wireless IP networks.

  In this embodiment, when a session is established from the MN 300 side to the switching server 100, the switching server 100 is controlled via all the wireless IP networks available from the MN 300 prior to the actual application execution. Send a message. Note that a control message sent to at least one of the available wireless IP networks, for example, a wireless IP network set in advance corresponding to an application, includes a jitter buffer allowable time (a fluctuation absorption time of received packets in the application). Insert application information including Tapr).

  In addition, the switching server 100 receives the control message transmitted from the MN 300, stores the allowable time (Tapr) included in the application information, and receives the received packet in each wireless IP network based on the received packet of the control message. The ideal reception timing and allowable limit timing are calculated. Based on the calculated ideal reception timing and allowable limit timing and allowable time (Tapr) in each wireless IP network, a fluctuation request time, that is, a jitter buffer accumulation request that does not discard any packet arriving from any wireless IP network. A plurality of wireless IP networks that are optimal for the application whose time (Tjit) does not exceed the allowable time (Tapr) are selected. The information on the selected wireless IP network is transmitted to the MN 300 as a transmission route selection message (Msg) together with the allowable bandwidth, whereby the switching server 100 and the MN 300 supplement the bandwidth by using the plurality of selected wireless IP networks simultaneously. While running the app.

  On the other hand, when a session is established from the switching server 100 side to the MN 300, similarly, prior to execution of the actual application, a control message is sent to the MN 300 via all the wireless IP networks available from the switching server 100. Is sent out.

  The MN 300 stores an allowable time (Tapr) for each application in advance, and receives the control message transmitted from the switching server 100, thereby reading the allowable time (Tapr) corresponding to the application to be executed. Therefore, in this case, it is not necessary to include the allowable time (Tapr) in the application information of the control message transmitted from the switching server 100. In addition, the MN 300 calculates the ideal reception timing and allowable limit timing of the received packet in each wireless IP network based on the control message packet received via each wireless IP network, and the calculated ideal in each wireless IP network. Based on the reception timing and allowable limit timing, and the read application allowable time (Tapr), the jitter buffer accumulation request time (Tjit) that is the fluctuation request time is optimal for applications that do not exceed the allowable time (Tapr) Select multiple wireless IP networks. The information of the selected wireless IP network is transmitted to the switching server 100 as a transmission route selection message (Msg) together with the allowable bandwidth, whereby the MN 300 and the switching server 100 simultaneously use the plurality of selected wireless IP networks. Run the app while complementing.

  Next, the functional block configuration of the communication system 1 will be described. Specifically, functional block configurations of the switching server 100 and the MN 300 included in the communication system 1 will be described. Hereinafter, portions related to the present invention will be mainly described. Therefore, it should be noted that the switching server 100 and the MN 300 may include a logic block (a power supply unit or the like) that is not shown in the drawing or that is omitted in description in order to realize the function as the device.

(Switching server 100)
FIG. 2 is a functional block diagram illustrating a configuration of a main part of the switching server 100. The switching server 100 includes a communication interface (I / F) 101 with the MN 300, a receiving unit 103, a route selection unit 105, a storage unit 107, a transmission route selection Msg (message) generation unit 109, a packet transmission route control unit 111, and a transmission route. A selection Msg (message) analysis unit 113, a transmission unit 115, and a communication I / F 117 with the IP telephone terminal 40 are included.

  The communication I / F 101 constitutes a communication I / F corresponding to a plurality of available wireless IP networks (wireless communication paths). For example, the communication I / F 101 is configured by 1000BASE-T defined by IEEE 802.3ab and connected to the Internet 20. Connected and connected to different wireless communication paths.

  In the present embodiment, as described above, since VPN based on IPSec is set, packets transmitted / received by the communication I / F 101, specifically, VoIP packets transmitted / received between the switching server 100 and the MN 300 (specifically, The VoIP packet transmitted by the MN 300 has the configuration shown in FIG. That is, a VoIP packet encapsulates a home IP header (home IP address AH), a TCP / UDP header, and a payload, and a care-of IP address is added.

  The access control packet transmitted / received between the switching server 100 and the MN 300 has the configuration shown in FIG. That is, the access control packet includes a data link layer header, a care-of IP address, a TCP header, and a control code.

  The communication I / F 117 constitutes a communication I / F corresponding to the communication network 10C, and is connected to the Internet 20 and used to execute communication with the IP telephone terminal 40.

  The switching server 100 illustrated in FIG. 2 can use six different wireless communication paths with the MN 300 via the communication I / F 101. For this reason, the receiving unit 103 has six receiving units 131 corresponding to six wireless communication paths, and the transmitting unit 115 also has six transmitting units 141 corresponding to six wireless communication paths.

  In the receiving unit 103, each receiving unit 131 includes a received packet monitoring unit 133, an ideal reception timing generating unit 135, and an allowable range calculating unit 137, and packets transmitted from the MN 300 via the corresponding wireless communication path. Is received by the received packet monitoring means 133 via the communication I / F 101.

  The received packet monitoring unit 133 has a jitter buffer that absorbs jitter of the packet received by the communication I / F 101. The packet received by the received packet monitoring unit 133 is distributed according to the type of packet. That is, when the received packet is a control message sent by the session start from the MN 300, the received packet is supplied to the ideal reception timing generating unit 135 and the allowable range calculating unit 137, and at least one wireless communication path is set. The application information included in the control message received via the application is supplied to the storage unit 107, and the storage unit 107 stores the allowable time (Tapr) for the application to be executed. If the received packet is a transmission route selection message (Msg) sent from the MN 300 when the session from the switching server 100 is started, the transmission route selection message (Msg) is supplied to the transmission route selection Msg analysis means 113. . Further, when the received packet is a packet after application execution, the received packet is transmitted to the IP telephone terminal 40 via the communication I / F 119.

  The ideal reception timing generation unit 135 generates an ideal reception timing in the wireless communication path of the packet transmitted from the MN 300 based on the received packet from the reception packet monitoring unit 133, and calculates the generated ideal reception timing as an allowable range. This is supplied to the means 137 and supplied to the route selection means 105. A method for generating the ideal reception timing will be described later.

  Based on the received packet from the received packet monitoring unit 133 and the ideal reception timing from the ideal reception timing generation unit 135, the allowable range calculation unit 137 calculates the allowable range from the ideal reception timing of the packet in the wireless communication path. Then, the calculated allowable range is supplied to the route selection means 105. A method for calculating the allowable range will be described later.

  The route selection unit 105 calculates the allowable limit timing based on the ideal reception timing and the allowable range obtained from the receiving unit 131 corresponding to each wireless communication path, the calculated allowable limit timing of each wireless communication path, and the storage unit Based on the allowable time (Tapr) of the application to be executed stored in 107, the fluctuation request time, that is, the jitter buffer accumulation request time (Tjit) in which packets arriving from any wireless communication path are not discarded is determined as the allowable time (Tapr). ) Is selected and the result is supplied to the transmission route selection Msg generation means 109 and the packet transmission route control means 111.

  Therefore, in the present embodiment, the ideal reception timing generation unit 135, the allowable range calculation unit 137, and the route selection unit 105 of each wireless communication path set the ideal reception timing and allowable limit timing of the received packet in each wireless communication path. The calculation means for calculating is configured, and the path selection means 105 is configured to select a plurality of wireless communication paths so that the jitter buffer accumulation request time (Tjit) does not exceed the allowable time (Tapr) of the application. ing. A specific route selection operation including the operation of the route selection means 105 will be described later.

  The transmission route selection Msg generation unit 109 creates a transmission route selection Msg including the allowable bandwidth information of each selected wireless communication route based on the information on the wireless communication route selected by the route selection unit 105. Note that the allowable bandwidth of each wireless communication path is calculated based on the number of received packets within a unit time at the ideal reception timing of the wireless communication path, for example. This transmission route selection Msg is supplied to the transmission unit 115 and communicated from, for example, a predetermined transmission unit 141 or an arbitrary transmission unit 141 set according to the application under the control of the packet transmission path control unit 111. The data is transmitted to the MN 300 via the I / F 101. The transmission route selection Msg is transmitted using the access control packet shown in FIG.

  Further, the transmission route selection Msg analysis unit 113 starts the session from the switching server 100, and based on the transmission route selection Msg from the MN 300 supplied from the received packet monitoring unit 133, the wireless communication route selected by the MN 300 and its The allowable bandwidth of each wireless communication path is analyzed, and the information is supplied to the packet transmission path control unit 111.

  When establishing a session from the switching server 100 to the MN 300, the packet transmission path control unit 111 sends all available transmission units 115 from the switching server 100 to the MN 300 prior to execution of the actual application. Control is performed so that a control message is transmitted via the transmission means 141.

  In actual application execution, the packet transmission path control unit 111 is analyzed based on information on a plurality of wireless communication paths selected by the path selection unit 105 or by the transmission path selection Msg analysis unit 113. Based on the information of the plurality of wireless communication paths selected by the MN 300, the corresponding transmission unit 141 of the transmission unit 115 is controlled, and the VoIP packet from the IP telephone terminal 40 received via the communication I / F 117 is selected. While allocating (complementing) according to the allowable bandwidth of each wireless communication path using a plurality of wireless communication paths, the data is transmitted to the MN 300 via the communication I / F 101. Each transmission unit 141 of the transmission unit 115 includes a jitter buffer that absorbs jitter of a packet received by the communication I / F 117.

  Specifically, each selected transmission means 141 adds the corresponding care-of IP address to the IP packet including the home IP address AH received from the IP telephone terminal 40, and the IP to which the care-of IP address is added. The packet is transmitted from the communication I / F 101 to the corresponding wireless IP network.

  In addition to the above functions, the switching server 100 according to the present embodiment converts the order of IP packets transmitted and received between the MN 300 and the IP telephone terminal 40 via each wireless communication path into VoIP packets. It also has a function of checking using the included RTP (real-time transport protocol) sequence number (SN). The switching server 100 also has a function of acquiring statistical information (for example, packet loss, throughput, jitter buffer underrun count and overrun count) of the relayed IP packet and transmitting the acquired information to the MN 300. Yes.

  Further, the switching server 100 includes a home IP address AH included in the IP packet transmitted by the IP telephone terminal 40 and a home IP address registered in a home agent (not shown) accessible via the Internet 20. It has a function of collating, so that it is possible to determine which home IP address AH is the home IP address assigned to the MN 300 by which carrier.

(MN300)
FIG. 4 is a functional block diagram illustrating a configuration of a main part of the MN 300. Similar to the switching server 100, the MN 300 can execute communication using a plurality of (here, a maximum of six) wireless communication paths simultaneously. Hereinafter, description of functional blocks similar to those of the switching server 100 described above will be omitted as appropriate.

  As illustrated in FIG. 4, the MN 300 includes a communication I / F 301, a reception unit 303, a route selection unit 305, a transmission route selection Msg generation unit 309, a packet transmission route control unit 311, a transmission route selection Msg analysis unit 313, and a transmission unit 315. Application processing means 317 is provided.

  The communication I / F 301 transmits and receives IP packets based on the corresponding care-of IP addresses assigned to the MN 300 in a plurality of wireless communication paths that can be used by the MN 300.

  The reception unit 303 includes six reception units 331 corresponding to six wireless communication paths, and the transmission unit 315 also includes six transmission units 341 corresponding to six wireless communication paths. In FIG. 4, in order to simplify the drawing, six receiving units 331 are provided in the receiving unit 303 and six transmitting units 341 are also provided in the transmitting unit 315, but reception corresponding to each wireless communication path is provided. The means 331 and the transmission means 341 can also be configured with a detachable wireless communication card.

  In the reception unit 303, each reception unit 331 includes a reception packet monitoring unit 333, an ideal reception timing generation unit 335, and an allowable range calculation unit 337, and is transmitted from the switching server 100 via a corresponding wireless communication path. The received packet monitoring unit 333 receives the packet via the communication I / F 301.

  The received packet monitoring unit 333 has a jitter buffer that absorbs jitter of the packet received by the communication I / F 301, and the packet received by the received packet monitoring unit 333 is distributed according to the type of packet. That is, when the received packet is a control message sent out by starting a session from the switching server 100, the received packet is supplied to the ideal reception timing generation unit 335 and the allowable range calculation unit 337. If the received packet is a transmission route selection message (Msg) sent from the switching server 100 when the session from the MN 300 is started, the transmission route selection message (Msg) is supplied to the transmission route selection Msg analysis means 313. . Further, when the received packet is a packet after application execution, the received packet is supplied to the application processing means 317.

  The ideal reception timing generation unit 335 generates the ideal reception timing in the wireless communication path of the packet transmitted from the switching server 100 in the same manner as the ideal reception timing generation unit 135 of the switching server 100, and the generated ideal reception timing Is supplied to the allowable range calculation means 337 and also to the route selection means 305.

  The allowable range calculation unit 337 calculates the allowable range from the ideal reception timing of the packet in the wireless communication path in the same manner as the allowable range calculation unit 137 of the switching server 100, and sends the calculated allowable range to the route selection unit 305. Supply.

  The application processing unit 317 processes a received packet from the receiving unit 303 according to the application, generates a packet (for example, a VoIP packet) according to the application, and sends the packet to the transmitting unit 315. The application processing means 317 is provided with a storage means 307 for storing the home IP address AH of the MN 300 associated with the care-of IP address in each wireless communication path that can be used by the MN 300 and receiving packets corresponding to the application. The jitter buffer allowable time (Tapr), which is the fluctuation absorption time, is stored.

  The route selection unit 305 calculates the allowable limit timing based on the ideal reception timing and the allowable range obtained from the reception unit 331 corresponding to each wireless communication path, and the calculated allowable limit timing of each wireless communication path and the storage unit Based on the allowable time (Tapr) in the application to be executed stored in 307, the jitter buffer accumulation request time (Tjit) does not exceed the allowable time (Tapr) in the same manner as the path selection unit 105 of the switching server 100. A plurality of radio communication paths optimal for the application are selected, and the results are supplied to the transmission path selection Msg generation means 309 and the packet transmission path control means 311.

  Therefore, as in the case of the switching server 100 described above, in the present embodiment, the ideal reception timing generation unit 335, the allowable range calculation unit 337, and the route selection unit 305 of each wireless communication route The calculation means for calculating the ideal reception timing and the allowable limit timing of the received packet is configured, and the path selection means 305 includes a plurality of jitter buffer accumulation request times (Tjit) that do not exceed the application allowable time (Tapr). Selection means for selecting a wireless communication path is configured.

  In the transmission route selection Msg generation unit 309, in the same manner as the transmission route selection Msg generation unit 109 of the switching server 100, based on the information on the wireless communication route selected by the route selection unit 305, A transmission route selection message (Msg) including allowable bandwidth information is created, and communication is performed from the transmission unit 315 using the access control packet shown in FIG. 3B under the control of the packet transmission route control means 311. The data is transmitted to the switching server 100 via the I / F 301.

  The transmission route selection Msg analysis unit 313 is selected by the switching server 100 based on the transmission route selection message (Msg) from the switching server 100 supplied from the received packet monitoring unit 333 when the session from the MN 300 starts. The wireless communication path and the allowable bandwidth of each wireless communication path are analyzed, and the information is supplied to the packet transmission path control means 311.

  When establishing a session from the MN 300 to the switching server 100, the packet transmission path control unit 311 uses the transmission unit 315 from the application processing unit 317 to the switching server 100 prior to execution of the actual application. Control is performed so that a control message is transmitted via all possible transmission means 341. Note that the control message transmitted from at least one transmission unit 341 includes a jitter buffer allowable time (Tapr) that is a fluctuation absorption time of received packets in the application stored in the storage unit 309 of the application processing unit 317. Insert app information.

  In actual application execution, the packet transmission path control unit 311 is analyzed based on information on a plurality of wireless communication paths selected by the path selection unit 305 or by the transmission path selection Msg analysis unit 313. Based on the information of the plurality of wireless communication paths selected by the switching server 100, the corresponding transmission unit 341 of the transmission unit 315 is controlled, and a packet (for example, a VoIP packet) from the application processing unit 317 is selected. The wireless communication path is allocated (complemented) according to the allowable bandwidth of each wireless communication path and transmitted to the switching server 100 via the communication I / F 301.

  Specifically, each selected transmission unit 341 adds the corresponding care-of IP address to the IP packet including the home IP address AH from the application processing unit 317, and the IP packet to which the care-of IP address is added. Is transmitted from the communication I / F 301 to the corresponding wireless IP network.

  Note that the MN 300 according to the present embodiment checks the order of IP packets transmitted / received to / from the IP telephone terminal 40 using the RTP sequence number (SN) included in the VoIP packet, similarly to the switching server 100. It also has a function to do.

(Ideal reception timing generation method and tolerance calculation method)
Next, an ideal reception timing generation method and an allowable range calculation method on the switching server 100 side and the MN 300 side described above will be described. Note that the ideal reception timing generation method and the allowable range calculation method are the same on the switching server 100 side and the MN 300 side, and therefore, here, a case where it is performed on the switching server 100 side will be described as a representative.

  In generating the ideal reception timing, first, as shown in FIG. 5, the straight line I of the inclination angle α indicating the relationship (t−SN) between the transmission time (t) of the packet transmitted by the MN 300 and the sequence number (SN). Ask for. For example, when the packet transmission interval is 20 ms, a straight line I with α = 50 (SN / s) is obtained. Next, assuming that the received packet is located in the direction of delay with respect to the straight line I, when the received packet is plotted on the SN-t drawing, the straight line I is moved from left to right, A plot of the received packet does not exist on the left side of the position, and a straight line I ′ at the position moved to the rightmost side is generated as an ideal reception timing.

A specific example of calculating the ideal reception timing in this case will be described below. First, a reference straight line (SN-tstd) corresponding to the straight line I in FIG.
SN = αstd · tstd, tstd = SN / αstd
Here, αstd is a slope determined by the packet transmission interval. For example, when the packet transmission interval is 20 ms, it is 1 SN / 20 ms (50 SN / s).

Next, the time difference (Tstdsub (s)) between the reference line and the received packet is calculated by the following equation.
Tstdsub (s) = t (s) -tstd (s)
For example, if the sequence number of the sth received packet is SN (s),
Tstdsub (s) = t (s) − {SN (s) / αstd}
It becomes. Here, the minimum value (Tsubmin) of the time difference (Tstdsub) in the received packet (s) is obtained.

If a straight line based on Tsubmin is the ideal reception timing (SN-tass),
tass = (SN / αstd) + Tsubmin
It becomes. That is, a straight line (corresponding to I ′ in FIG. 5) based on a packet having the smallest (reception time−transmission time) among reception packets is set as an ideal reception timing.

  Further, for example, the allowable range can be obtained by uniquely calculating the allowable limit timing from the ideal reception timing according to the communication method in the corresponding wireless communication path or according to the modulation method or transmission band to be used. However, in the present embodiment, based on the generated ideal reception timing and the actual packet reception timing, for example, the allowable range is calculated as follows.

That is, when the ideal reception timing is generated, the difference time Tsub between the reception time trec of the received packet and the reception time tide at the ideal reception timing is calculated in a predetermined period, for example, by the following equation.
Tsub (s) = trec (s) -tide (s)
Next, the allowable range Tp is calculated based on the average value of the difference time Tsub, that is, the average value of the spread of the delay time of the received packet (average reception timing) by the following equation. In the following equation, s is a reception number assigned to a packet received during a predetermined period, β is a coefficient, and mean is an averaging process.

  In this way, when the ideal reception timing and the allowable range are obtained, the allowable limit timing of the wireless communication path can be calculated by adding the allowable range to the ideal reception timing.

  FIG. 6 shows an example of an ideal reception timing generation result and an allowable range calculation result. The horizontal axis represents time (t) and the vertical axis represents a sequence number (SN). Here, the generation result of the ideal reception timing and the calculation result of the allowable range in the two wireless communication routes of route A and route B are shown.

  In FIG. 6, a straight line I indicates a transmission timing, and a straight line R indicates an assumed reproduction timing. Tpa indicates the allowable range (period) of the route A, and Tpb indicates the allowable range (period) of the route B. Here, the allowable limit timing of the route A is the lower limit of the allowable range Tpa, and the allowable limit timing of the route B is the lower limit of the allowable range Tpb. Tab indicates a delay difference between the path A and the path B, and Tjit indicates a jitter buffer accumulation request time for the path A and the path B, respectively.

  In calculating the jitter buffer accumulation request time Tjit, first, the latest allowable limit timing among the allowable limit timings of each wireless communication path is obtained as the assumed reproduction timing R. In FIG. 6, the assumed reproduction timing R matches the allowable limit timing of the path B. Next, among the ideal reception timings of the wireless communication paths, a value obtained by subtracting the earliest ideal reception timing (the smallest delay with respect to the transmission timing) from the assumed reproduction timing R is calculated as the accumulation request time Tjit. In FIG. 6, the accumulation request time Tjit overlaps the allowable range Tpa of the route A and the allowable range Tpb of the route B.

  Here, if the calculated accumulation request time Tjit is less than or equal to the allowable time Tapr of the jitter buffer in the application, assuming that the reproduction by the application is performed within the assumed reproduction timing R, and within the allowable range Tpa of the path A and the path All packets respectively received within the allowable range Tpb of B are used for application playback without being discarded.

  FIG. 7 shows an example of simulation results of ideal reception timing and allowable range. In this example, the allowable limit timing of the route A and the ideal reception timing of the route B substantially coincide. Further, the accumulation request time Tjit for the route A and the route B is about 218.6 [ms] (actual measurement from the drawing). Therefore, in this case, if the allowable time Tapr of the jitter buffer by the application is 218.6 [ms] or more, by performing communication while complementing using these two paths A and B, the allowable path A can be obtained. Packets received within the area Tpa and within the allowable area Tpb of the path B are reproduced without being discarded.

(Route selection operation)
Next, with reference to the flowchart shown in FIG. 8, the operation of selecting a wireless communication path according to this embodiment will be described. Since this route selection operation is the same on the switching server 100 side and the MN 300 side, here, a case where it is performed on the switching server 100 side will be described as a representative.

  First, when the session starts from the MN 300, the allowable time (Tapr) of the application included in the control message transmitted from the MN 300 is set in the storage unit 107 (step S1). Next, for each of all available wireless communication paths (L), ideal reception timing generation means 135 calculates ideal reception timing (step S2), and allowable range calculation means 137 calculates an allowable range (step S2). S3) After that, the route selection unit 105 adds an allowable range to the ideal reception timing and calculates an allowable limit timing (step S4).

  When the calculation of the permissible limit timing in each wireless communication path is completed, the path selection unit 105 next assumes and reproduces the latest permissible limit timing of each wireless communication path, as described with reference to FIG. The jitter buffer accumulation request time (Tjit) is calculated by subtracting the earliest ideal reception timing from the assumed reproduction timing (step S5). A specific method of calculating the accumulation request time (Tjit) will be described later.

  Next, the calculated accumulation request time (Tjit) is compared with the allowable time (Tapr) of the application set in step S1 (step S6). If Tjit ≦ Tapr, the accumulation request time (Tjit) is calculated. All of the used wireless communication paths are selected as usable, and the wireless communication paths with higher priority are set in the order of late tolerance limit timing, that is, in order of increasing delay (step S7).

  On the other hand, if Tjit> Tapr in step S6, the accumulation request time (Tjit (Llst)) is calculated from the set of wireless communication paths excluding the wireless communication path (Llst) with the slowest allowable limit timing (step S8). Further, the accumulation request time (Tjit (Lfst)) is calculated from the set of wireless communication paths excluding the wireless communication path (Lfst) with the earliest ideal reception timing (step S9), and Tjit (Llst) and Tjit ( Lfst) is compared (step S10).

  As a result, in the case of Tjit (Llst) ≦ Tjit (Lfst), the accumulation request time Tjit (Llst) calculated by excluding the wireless communication path (Llst) with the slowest allowable limit timing from the set of wireless communication paths is calculated as Tjit = Tjit (Llst) is set (step S11). In other cases, the accumulation request time (Tjit (Lfst)) calculated by excluding the wireless communication path (Lfst) with the earliest ideal reception timing from the set of wireless communication paths is defined as Tjit. = Tjit (Lfst) (step S12), the process proceeds to step S6, and the accumulation time (Tjit) is compared with the allowable time (Tapr) of the application again.

  As described above, in step S6, by selecting a set of wireless communication paths that satisfy the accumulation request time (Tjit) that is equal to or less than the allowable time (Tapr) of the application, the number of selected wireless communication paths is increased and complementary communication is performed. Therefore, it is possible to reduce the discarded packets more efficiently.

  FIG. 9 is a flowchart showing a specific method for calculating the accumulation request time (Tjit) calculated in step S5 of FIG. First, for the set of target wireless communication paths, the wireless communication paths are sorted in the order of late allowable limit timing (step S21), and the latest allowable limit timing is set as an assumed reproduction timing in the application (step S22).

  Next, the routes are sorted in order of the ideal reception timing with respect to the set of target wireless communication routes (step S23), and the value obtained by subtracting the earliest ideal reception timing from the assumed reproduction timing is used as the jitter buffer accumulation request. Calculated as time (Tjit) (step S24).

  When a plurality of wireless communication paths are selected as described above, information on the selected wireless communication paths is transmitted to the MN 300 as a transmission path selection message (Msg) together with information on the allowable bandwidth. Thereby, in the switching server 100 and the MN 300, a plurality of selected wireless communication paths are used at the same time, and the wireless communication path having the largest allowable limit timing delay (the highest priority) is set as the main path (one wireless communication path). As described above, the application is executed by allocating the requested bandwidth of the application according to the allowable bandwidth so that the bandwidth shortage in the main route is supplemented by another wireless communication route.

  According to the present embodiment, the allowable limit timing of the received packet in each wireless communication path is calculated based on the packet of the control message received via a plurality of available wireless communication paths, and each calculated wireless communication Based on the allowable limit timing in the path and the jitter buffer allowable time (Tapr) that is the fluctuation absorption time of the received packet in the application, the jitter buffer that is the fluctuation request time is selected from a plurality of available wireless communication paths. Select multiple wireless communication paths whose accumulation request time (Tjit) does not exceed the allowable time (Tapr), so select the optimal wireless communication paths for the application considering the radio propagation environment at the start of communication. Thus, complementary communication can be executed, and discarding of arrived packets can be efficiently reduced.

  In addition, this invention is not limited to the said embodiment, Many deformation | transformation or a change is possible. For example, in the above embodiment, the path selection operation is executed on the partner side by the control message sent to the partner side when the session is established, and the result is received. It is also possible to execute a route selection operation on the own device side according to the control message received from the mobile station and transmit the result to the other side.

  The operation of selecting a wireless communication path is not limited to when a session is established, but can also be performed while an application is being executed. In this case, for the wireless communication path that is selected, the ideal reception timing and allowable range are calculated based on the received packet of the application to calculate the allowable limit timing, and for the wireless communication path that is not selected, it is necessary. Accordingly, by transmitting / receiving the control message at an appropriate cycle and calculating the ideal reception timing and allowable range based on the received packet and calculating the allowable limit timing, it is possible to select a plurality of wireless communication paths optimal for the application. Dynamic selection is performed, and complementary communication is executed according to the allowable bandwidth at that time.

  In this way, even when the propagation environment changes during communication and the allowable bandwidth of the selected wireless communication path fluctuates, it can be dealt with almost in real time, thereby further improving communication reliability. be able to. In addition, in the case of dynamically selecting a plurality of wireless communication paths in this way, a session can be disconnected for a non-selected wireless communication path, but preferably, while maintaining the session, Control is made to stop packet transmission until the next selection operation. In this way, it is possible to quickly execute the next selection operation.

1 is an overall schematic configuration diagram of a communication system according to an embodiment of the present invention. It is a functional block block diagram of the switching server shown in FIG. It is a block diagram of the IP packet which concerns on embodiment shown in FIG. It is a functional block block diagram of MN shown in FIG. It is a figure for demonstrating the production | generation method of an ideal reception timing. It is a figure which shows an example of the production | generation result of an ideal reception timing, and the calculation result of an allowable range. It is a figure which shows an example of the simulation result of an ideal receiving timing and a tolerance | permissible_range. It is a flowchart which shows the selection operation | movement of the radio | wireless communication path | route by embodiment shown in FIG. It is a flowchart which shows the specific calculation method of the accumulation | storage request | requirement time (Tjit) calculated by step S5 of FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Communication system 10A, 10B Wireless IP network 10C Communication network 20 Internet 40 IP telephone terminal
100 switching server 101, 301 communication I / F
103,303 Receiving unit 105,305 Route selection unit 107,307 Storage unit 109,309 Transmission route selection Msg generation unit 111,311 Packet transmission path control unit 113,313 Transmission route selection Msg analysis unit 115,315 Transmission unit 117 Communication I / F
131,331 receiving means 133,333 received packet monitoring means 135,335 ideal reception timing generating means 137,337 allowable range calculating means 141,341 transmitting means 317 application processing means

Claims (8)

Select a plurality of different wireless communication paths with the wireless communication device, and supplement the bandwidth that is lacking in one wireless communication path with the other wireless communication paths with respect to the requested bandwidth of the application that has real-time characteristics. A communication control device for controlling communication between the wireless communication device and a communication destination,
Based on a packet received via each wireless communication path, calculating means for calculating an ideal reception timing and an allowable limit timing of a received packet in the wireless communication path;
Storage means for receiving and storing fluctuation absorption time of a received packet in the application transmitted from the wireless communication device;
In order that the fluctuation request time calculated based on the ideal reception timing and the allowable limit timing in each wireless communication path calculated by the calculation means does not exceed the fluctuation absorption time in the application stored in the storage means Selecting means for selecting different wireless communication paths;
A communication control device comprising:   The selection means selects a wireless communication path with the latest allowable limit timing calculated by the calculation means as the one wireless communication path from among the plurality of different wireless communication paths to be selected. The communication control apparatus according to 1.   The selection unit subtracts the earliest ideal reception timing from the latest allowable limit timing calculated by the calculation unit among the plurality of different wireless communication paths to be selected, and obtains the fluctuation request time. The communication control apparatus according to claim 1 or 2. Multiple different wireless communication paths are selected with the communication control device, and the bandwidth that is insufficient in one wireless communication path is supplemented by other wireless communication paths with respect to the required bandwidth of the application having real-time characteristics to be used. A wireless communication device that wirelessly communicates with a communication destination via the communication control device,
Based on a packet received via each wireless communication path, calculating means for calculating an ideal reception timing and an allowable limit timing of a received packet in the wireless communication path;
Storage means for storing fluctuation absorption time of received packets in the application;
The fluctuation request time calculated based on the ideal reception timing and the allowable limit timing in each wireless communication path calculated by the calculation means so as not to exceed the fluctuation absorption time in the application stored in the storage means. Selecting means for selecting the plurality of different wireless communication paths;
A wireless communication apparatus comprising:   The selection means selects a wireless communication path with the latest allowable limit timing calculated by the calculation means as the one wireless communication path from among the plurality of different wireless communication paths to be selected. 4. A wireless communication device according to 4.   The selection unit subtracts the earliest ideal reception timing from the latest allowable limit timing calculated by the calculation unit among the plurality of different wireless communication paths to be selected, and obtains the fluctuation request time. The wireless communication apparatus according to claim 4 or 5. Select a plurality of different wireless communication paths with the wireless communication device, and supplement the bandwidth that is lacking in one wireless communication path with the other wireless communication paths with respect to the requested bandwidth of the application that has real-time characteristics. A communication control method for controlling communication between the wireless communication device and a communication destination,
Calculating ideal reception timing and allowable limit timing of a received packet in the wireless communication path based on a packet received via each wireless communication path;
Receiving and storing fluctuation absorption time of a received packet in the application transmitted from the wireless communication device; and
The plurality of different wireless communication paths are set such that the fluctuation request time calculated based on the ideal reception timing and allowable limit timing in each of the calculated wireless communication paths does not exceed the fluctuation absorption time in the stored application. A step to choose;
A communication control method characterized by comprising: Multiple different wireless communication paths are selected with the communication control device, and the bandwidth that is insufficient in one wireless communication path is supplemented by other wireless communication paths with respect to the required bandwidth of the application having real-time characteristics to be used. A wireless communication method for wirelessly communicating with a communication destination via the communication control device,
Calculating ideal reception timing and allowable limit timing of a received packet in the wireless communication path based on a packet received via each wireless communication path;
The plurality of different wireless communication paths so that the fluctuation request time calculated based on the ideal reception timing and allowable limit timing in each of the calculated wireless communication paths does not exceed the fluctuation absorption time in the preset application. A step of selecting
A wireless communication method comprising: