JP2007142786A - Handover server, and mobile communication terminal communcable thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology capable of smoothly switching communication networks in the case of switching two or more of the communication networks to continue a speech in a telephone service needing real time performance. <P>SOLUTION: The handover server (100) providing a handover function is installed between two terminals (104, 105) and the handover server introduces a contrivance of a speech session between the two terminals. The handover server executes seamless handover processing by two sessions of the same speech data at the same time between the handover server and the terminals so as to smoothly realize consecution of the speech service. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a technique for handover processing when communication is performed between a plurality of mobile communication terminals (for example, cellular phones) via a plurality of communication networks.

  When a mobile communication terminal such as a mobile phone is connected to another terminal or server and receives a communication service, a so-called handover process is performed when the communication service is continued by dynamically switching the communication network. In this handover process, for example, session information of a communication service is held in a relay device or an agent in a communication network (hereinafter also referred to as a communication path), and the session information is transmitted in the communication service in the switched communication network. Is used to continue communication services.

  Here, the session information includes, for example, an IP address and port number of a server / client terminal, a communication protocol, a session start time, a service elapsed time, a service name, and an application name used.

  As a conventional technique for smoothly switching the communication network, for example, one disclosed in Patent Document 1 is known. This discloses a method of temporarily establishing two sessions on a communication path before switching and two sessions on a communication path after switching with a mobile terminal at the same time.

JP 2004-356922 A

  The thing of the said patent document 1 can be applied only to the service which distributes the accumulated content, and considers smooth switching of the said communication network in communication between several different mobile communication terminals. It has not been.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a technique for smoothly switching communication networks in communication between a plurality of different mobile communication terminals.

  In order to achieve the above object, according to the present invention, when switching a communication network, two sessions are temporarily established simultaneously on each communication network before and after the switching. The same call data is transmitted and received in both sessions, and the communication network (session) is switched based on the time information of the call data received from the two communication networks before and after switching. It is.

  When the switching of the communication network fails, either a call through the communication network before switching or the retry of the switching process is executed. Further, when transmitting call data from one mobile communication terminal to another mobile communication terminal, using a certain communication network, when transmitting call data from the other mobile communication terminal to another mobile communication terminal, The use of a communication network is also a feature of the present invention.

  According to the present invention, it is possible to smoothly switch communication networks in communication between a plurality of different mobile communication terminals.

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

  A telephone handover system according to Embodiment 1 of the present invention will be described with reference to FIGS.

  FIG. 1 is a diagram illustrating a configuration of a handover server 100 and a handover terminal 105 that is a mobile communication terminal in the telephone handover system according to the first embodiment. In the figure, a system configuration in the case where the telephone terminal 104 and the handover terminal 105 make a call by relaying the handover server 100 is shown. Here, the telephone terminal 104 is a telephone terminal having a general IP telephone function that does not have the handover function disclosed in the present invention.

  In FIG. 1, the handover server 100 includes a handover processing unit 101, a call processing unit 102, and a communication network interface unit 103. The handover processing unit 101 performs a switching process between a call session via the communication network # 2 (108) and a call session via the communication network # 3 (107), and a call session between the telephone terminal 104 and the handover server 100; The contents of the call in the call session between the handover server 100 and the handover terminal 105 are reflected on each other. The call processing unit 102 controls message transmission / reception and call data transmission / reception in a call session with another terminal. The communication network interface unit 103 is a network interface for accessing the communication network # 1.

  The handover terminal 105 includes a communication network interface unit 109, a call processing unit 110, a handover processing unit 111, and a voice processing unit 112. The communication network interface unit 109 is a network interface for accessing the communication network # 2 (108) and the communication network # 3 (107). The handover terminal can be connected to the communication network # 1 (106) via the communication network # 2 (108) and the communication network # 3 (107). The call processing unit 110 controls transmission / reception of messages and transmission / reception of call data in a call session with another terminal. The handover processing unit 111 performs a switching process between a call session via the communication network # 2 (108) and a call session via the communication network # 3 (107). The voice processing unit 112 encodes a call sound input from an input device such as a microphone and passes it to the handover processing unit 111 (D103a), and also decodes the received call data (D103b) passed from the handover processing unit to generate a speaker or the like. The call sound is output to the output device.

  Next, FIG. 2 shows a flow when the handover of the IP telephone service is performed between the handover terminal 105 and the general telephone terminal 104 via the handover server 100. In the figure, an example is shown in which SIP (Session Initiation Protocol: IETF RFC3261) is used as a call control protocol for performing a call for an IP telephone service. However, the handover process of the present invention is performed in accordance with ITU-T H.264. The same applies to other call control protocols such as H.323.

  In the figure, the handover terminal 105, the handover server 100, and the telephone terminal 104 directly exchange SIP messages. However, calls between the handover terminal 105 and the handover server 100, and between the handover server 100 and the telephone terminal 104 are shown. The handover process of the present invention can be similarly applied even to a configuration in which one or more call control servers such as a SIP proxy are passed for establishment.

  The handover terminal 105 includes a mode flag indicating whether the call processing unit 110 operates with a normal IP telephone function or via a handover server. In the case of the handover server via mode, the call processing unit 110 transmits an INVITE (session establishment request) message (M200) to the telephone terminal 105 to the handover server 100 via the communication network # 2 (108). When the call processing unit 102 in the handover server 100 receives the INVITE message (M200), the call processing unit 102 returns a Ringing (calling) message (M201) to the handover terminal 105 and also includes a destination URI included in the “To” header in the INVITE message. (In the example of the figure, the URI of the telephone terminal 104) is analyzed, and an INVITE message is transmitted to the telephone terminal 104 (M202). When the call processing unit 110 in the handover terminal 105 receives the Ringing message (M201), the call processing unit 110 sounds a ringing tone at the handover terminal 105.

  When the telephone terminal 104 receives the INVITE message (M202), the telephone terminal 104 returns a Ringing message (M203), rings a ringing tone, and waits until the user of the telephone terminal 104 raises the handset to allow the call. When the call is permitted, an OK message (M204) is transmitted to the handover server 100. When the call processing unit 102 in the handover server 100 receives the OK message (M204), it transmits an OK message (M205) to the handover terminal 105. When the call processing unit 110 in the handover terminal 105 receives the OK message (M205), it returns an ACK (acknowledgement) message (M206) to the handover server 100. Upon receiving the ACK message (M206), the call processing unit 102 in the handover server 100 transmits an ACK message (M207) to the telephone terminal 104. When the telephone terminal 104 receives the ACK message (M207), a call session between the handover terminal 105 and the telephone terminal 104 is established through the handover server 100 via the communication network # 2 (108).

  After this point, the handover terminal 105 transmits and receives call data (D101) to and from the handover server 100. In addition, the handover server 100 transmits and receives call data (D100) with the telephone terminal 104. The handover processing unit 101 in the handover server 100 stores call data (D101b) sent from the handover terminal 105 in the internal buffer area 401 in order to absorb fluctuations in communication delay such as jitter, and a certain amount of buffer is stored. The call data is transferred to the telephone terminal 104 from the oldest when the call reaches the telephone terminal 104, and the call data (D100a) sent from the telephone terminal 104 is transferred to the handover terminal 105. The handover over terminal 105 stores the received call data (D101a) in the buffer 301 in order to absorb fluctuations in communication delay such as jitter, and the call data (D101a) in chronological order when the buffer reaches a certain amount. Is transferred to the voice processing unit 112 (D103). The voice processing unit 112 performs packet analysis of the call data (D103a), voice data separation, and voice decoding, and outputs voice from the output device. Also, voice is input from the input device, voice encoding and packet generation are performed, and the data is sent to the call processing unit 110 as call data (D103b), which is sent to the handover server 100 as call data (D101b). Through the processing as described above, a call between the handover terminal 105 and the telephone terminal 104 is realized. In SIP, the above call data is transmitted and received in a voice packet format according to RTP (Realtime Transfer Protocol: IETF RFC3550). However, the present invention is not necessarily limited to RTP.

  Note that FIG. 2 shows an example in which a call is made from the handover terminal 105 to the telephone terminal 104. However, when a call is made from the telephone terminal 104 to the handover terminal 105, session establishment via the handover server 100 is similarly performed. Done. In this case, the handover terminal 105 needs to perform processing such as registering in advance in the SIP server that manages the terminal 105 so as to transfer the incoming call to the handover server 100 (for example, a SIP redirect function). Etc.). The telephone terminal 104 needs to know the URI of the handover server 100 as a transfer destination of the handover terminal 105.

  Next, in a state where the handover terminal 105 is making a call with the telephone terminal 104 via the handover server 100 via the communication network # 2 (108) as described above, the handover terminal 105 somehow communicates with the communication network # 2. 2 is a process for switching a call via the communication network # 3 (107) when a throughput drop in (108) is detected and a handover is determined (201). However, the handover decision (201) is not necessarily performed only by detecting a decrease in throughput. For example, if the handover terminal 105 is set to preferentially switch to the communication network # 3 (107) when the communication network # 3 (107) is available, the communication network # 2 (108) The handover is determined even if the throughput of () is not reduced. Alternatively, the handover may be determined by performing an operation on the terminal so that the user of the handover terminal 105 forcibly switches to the communication network # 3 (107). In addition, when the handover terminal 105 can acquire location information of the terminal itself by some means such as having a function such as GPS (Global Positioning System), it is related to the throughput of the communication network # 2 and the communication network # 3. Instead, handover may be determined according to the location information. The configuration of a terminal that determines handover according to location information will be described later as an eighth embodiment.

  As a method of detecting the above-described decrease in throughput, there is a method of exchanging packet transmission / reception information by RTCP (RTP Control Protocol: IETF RFC3550). In addition, a method for detecting an increase in packet loss from the sequence number of the received RTP packet, a method for detecting a decrease in the remaining buffer capacity, a method for performing dummy communication for throughput measurement, or a radio wave condition disclosed by the wireless device of the terminal It is also possible to apply a method of referring to the above.

  When the handover terminal 105 detects a decrease in throughput of the communication network # 2 (108) and decides to execute handover (201), the handover processing unit 110 sends an INVITE message (M208) via the communication network # 3 (107). Transmit to the handover server 100. Upon receiving the INVITE message (M208), the handover processing unit 102 in the handover server 100 receives the incoming caller identifier (using the tag value of the From header, etc.) and the incoming call of the established session via the communication network # 2 (108). It is verified whether the information with the destination address (To header) matches the information in the INVITE message (M208) (202). When it is confirmed by the above collation that the incoming call source and the incoming call destination of communication network # 2 (session established) and communication network # 3 (session establishment requested) match, the call processing unit in handover server 100 102 returns an OK message (M209) to the handover terminal 105. When the call processing unit 110 in the handover terminal 105 receives the OK message (M209), it returns an ACK message (M210) to the handover server 100. At this time, since the session with the handover server 100 via the communication network # 3 (107) is established, the communication network # 2 (108) and the communication network # 3 (107) are simultaneously connected on the two communication networks. Two sessions are established.

  Thereafter, the handover server 100 transmits / receives call data to / from the handover terminal 105 (D102). At this time, the handover processing unit 101 in the handover server 100 transfers the same data as the call data already transmitted to the handover terminal 105 to the handover terminal 105 via the communication network # 3 (107) (D102a). The handover terminal 105 continues to store the received call data (D102a) in the buffer 302. When the remaining capacity of the buffer 320 is exhausted, the buffer 320 is deleted in the oldest order. Further, the handover processing unit 101 receives call data (D102b) sent from the handover terminal 105 via the communication network # 3 (107) and stores it in the internal buffer area 402. When the buffer capacity runs out, the old call data is deleted.

  At this point, the call data (D100) sent from the telephone terminal 104 from the handover server 100 is simultaneously sent from both the communication networks # 2 (108) and # 3 (107) to the call data D101a and D102a, respectively. To the handover terminal 105. Call data D101b and D102b are sent from the handover terminal 105 to the handover server via the communication network # 2 (108) and the communication network # 3 (107). Then, the call data (D101) of the communication network # 2 (108) is transferred to the telephone terminal 104 as call data (D100).

  In this state, the handover processing unit 111 of the handover terminal 105 starts executing the handover process 203, and the handover processing unit 101 of the handover server 100 starts executing the handover process 204.

  Here, the handover process 203 performed in the handover terminal 105 will be described. When two call sessions are established as described above, the handover processing unit 111 in the handover terminal 105 receives the call data (D101a) received via the communication network # 2 (108) and the communication network # 3. Call data (D102a) received via (107) is stored in buffers 301 and 302, respectively. In the handover process 203, the time information of the call data stored in each buffer is managed, and the time at which the call data in both buffers is not interrupted in time is determined as the handover time. The subsequent call data (D103a) is switched from the call data (D101a) in the buffer 301 to the call data (D102a) in the buffer 302. However, the difference in time is reduced as much as possible when data that overlaps in time is not stored in both buffers due to a large delay difference between the communication network # 2 and the communication network # 3. Such a time is determined as the handover time, and the call data is switched as described above.

  The handover process 204 performed in the handover server 100 is the same as the handover process 203 (see FIG. 4). That is, at the time when two call sessions are established as described above, the handover processing unit 101 in the handover server 100 receives the call data (D101b) received via the communication network # 2 (108) and the communication network #. Call data (D102b) received via 3 (107) is stored in buffers 401 and 402, respectively. In the handover process 204, the time information of the call data stored in each buffer is managed, and the time at which the call data in both buffers is not interrupted in time is determined as the handover time. The subsequent call data (D100b) is switched from the call data (D101b) in the buffer 401 to the call data (D102b) in the buffer 402. However, the difference in time is reduced as much as possible when data that overlaps in time is not stored in both buffers due to a large delay difference between the communication network # 2 and the communication network # 3. Such a time is determined as the handover time, and the call data is switched as described above.

  When the handover process 204 in the handover server 100 ends, the handover processing unit 101 of the handover server 100 transmits a handover completion message (M211) to the handover terminal 105. The handover terminal 105 knows the end of the handover process 204 of the handover server 100 by receiving the handover completion message (M211). When the handover process 203 and the handover server 204 are completed, the handover terminal 105 transmits a BYE message (M212) to the handover server 100 via the communication network # 2, and the handover terminal 105 transmits to the handover server via the communication network # 2. The transmission of the call data (D101b) to be transmitted is terminated (205). When the handover server 100 receives the BYE message (M212), the handover server 100 stops transmission of the call data (D101a) to the handover terminal 105 via the communication network # 2 (206), and sends the OK message (M213) to the handover terminal. It transmits to 105. As described above, when the call session of the communication network # 2 ends, seamless switching of the call session from the communication network # 2 to the communication network # 3 can be realized. The above example is a case where the handover terminal 105 and the handover server 100 stop the transmission of call data in both after the handover process is completed, and the call session of the communication network # 2 is terminated.

  On the other hand, the handover terminal 105 and the handover server 100 can also operate to stop transmission of call data to the other when the other handover process is completed without waiting for the completion of the handover process of the handover terminal 105 and the handover server 100. Processing in this case will be described with reference to FIG.

  In FIG. 7, the flow until the handover processes 203 and 204 are executed in both the handover terminal 105 and the handover server 100 is the same as described above. Here, the handover terminal 105 transmits a handover completion message (M700) to the handover server 100 when the handover process 203 is completed. When the handover server 100 receives this completion message (M700), it stops transmitting call data to the communication network # 2 (206). Then, the handover server 100 transmits a handover completion message (M701) to the handover terminal 105 when the handover process 204 is completed. When the handover terminal 105 receives the completion message (M701), the handover terminal 105 knows that the handover process has been completed in both the handover terminal and the handover server, and transmits a BYE message (M212) to the handover server 100 via the communication network # 2. . Then, the handover terminal 105 stops transmitting call data to the handover server 100 (205). Thereafter, the handover server 100 returns an OK message (M213) for the BYE message (M212) to the handover terminal 105. As described above, when the communication session of the communication network # 2 ends, seamless switching of the communication session from the communication network # 2 to the communication network # 3 can be realized.

  As described above, in the telephone handover system according to Embodiment 1 of the present invention, a handover server that provides a handover function is installed between two terminals, and a mechanism is introduced in which the handover server relays a call session between the two terminals. The same call data session can be simultaneously established between the handover server and the handover terminal. Each of the handover terminal and the handover server acquires the reproduction time of the two call data received in these sessions, determines the timing so as not to cause interruption as much as possible, and switches both data. Therefore, the seamless handover function that can be realized only by the conventional one-way distribution service can be realized also by a bidirectional communication service such as an IP telephone.

  In a second embodiment of the present invention, as a modification of the first embodiment, when the handover processing unit 111 of the handover terminal 105 sends the call data D103b passed from the voice processing unit 112 to the handover server, A configuration is adopted in which the call data D103b is sent to the handover server 100 after reaching a certain buffer amount. Similarly, when the handover processing unit 101 of the handover server 100 sends the call data D100a sent from the telephone terminal 104 to the handover terminal 105, it accumulates in the buffer and then sends it to the handover terminal 105 after reaching a certain buffer amount. Take the configuration. According to the second embodiment of the present invention, the handover terminal 105 and the handover server 100 both reduce the influence of rate fluctuation and jitter at the time of packet transmission to the communication network by providing a certain delay in the buffer at the time of transmission. Can do.

  The third embodiment of the present invention has a configuration in which the codec and the content bit rate are changed according to the monitored throughput as a modification of the second embodiment. According to Embodiment 3 of the present invention, packet loss can be reduced by adapting the codec and bit rate at the time of transmission to the throughput.

  The fourth embodiment of the present invention has a configuration including not only audio but also a video processing function as a modification of the first embodiment. A seamless handover function can also be realized for the IP-TV telephone service by adding the same configuration as the call data of the first embodiment to the video data.

  The fifth embodiment of the present invention operates according to a sequence including the process in the case where the handover process fails in the sequence in the first embodiment shown in FIGS. Here, the phenomenon in which the handover process fails is that the communication path is switched to the new communication network # 3, but the throughput in the communication network # 3 is insufficient, and sufficient call data is not received to continue the call. means.

  In FIG. 2, the handover terminal 105 and the handover server 100 do not stop the transmission of the call data until the handover process in both is completed. That is, even when the handover process fails, call data is transmitted bidirectionally in both the communication network # 2 and the communication network # 3. In the handover system of the present embodiment, when the handover process fails, bidirectional communication is performed either on the communication network before the handover execution (communication network # 2) or on the communication network after the handover execution (communication network # 3). Operates to return the path.

  The flow of processing for returning a bidirectional communication path to the handover communication network # 2 when the handover processing fails will be described with reference to FIG. In FIG. 5, the flow until the handover processes 203 and 204 are executed in both the handover terminal 105 and the handover server 100 is the same as described above. When the handover process 204 fails, the handover server 100 transmits a handover failure message (M500) to the handover terminal 105. When receiving the handover failure message (M500), the handover terminal 105 stops the handover process 203 (501).

  This handover cancellation process 501 is a voice message if the call data sent to the voice processing unit 112 by the handover process 203 is data received from the communication network # 2 when the handover failure message (M500) is received. The operation of switching the call data sent to the processing unit 112 via the communication network # 3 is stopped. If the call data sent to the voice processing unit 112 by the handover process 203 is data received from the communication network # 3, an operation of seamlessly switching the call data sent to the voice processing unit 112 via the communication network # 2. Execute.

  The handover terminal 105 transmits a BYE message (M212) to the handover terminal 100 via the communication network # 3 when the call data reception path becomes the communication network # 2 by the handover cancellation process 501. Then, the handover terminal 105 stops transmission of call data via the communication network # 3 to the handover server 100 (205).

  Thereafter, when the handover server 100 receives the BYE message (M212), the handover server 100 stops transmitting the call data via the communication network # 3 to the handover terminal 105 (206), and returns an OK message (M213) to the handover terminal 105. According to the flow described above, even when the handover process fails in the handover server, the bidirectional communication path can be returned to the communication network before the handover is executed. Such an operation is effective in a situation where the communication network # 2 is expected to be relatively stable compared to the communication network # 3.

  Conversely, in a situation where it can be expected that communication network # 3 is more stable than communication network # 2, when the handover process fails, the communication path is not returned to communication network # 2. In addition, it is more likely that the communication can be continued if the switching to the communication network # 3 is repeatedly performed.

  FIG. 6 shows the flow of processing when repeatedly trying to switch to the communication network # 3 when the handover processing fails. In FIG. 6, the flow until the handover processes 203 and 600 are executed in both the handover terminal 105 and the handover server 100 is the same as described above. Here, the handover process 600 receives the call data from the communication network # 3 without transmitting the handover failure notification message as shown in FIG. 5 even when the handover process 204 has failed. A function to repeatedly switch until it stabilizes is added. That is, the handover process is not completed until the handover process 600 is completed and the call data from the communication network # 3 is supplied to the voice processing unit 101. Until the handover process is completed, bidirectional call data is transmitted and received in both the communication network # 2 and the communication network # 3. The flow of processing after the handover processing 600 is completed is as described above.

  FIG. 7 shows a handover process (success example) different from FIG. The difference between FIG. 7 and FIG. 2 is that the handover terminal 105 and the handover server 100 stop transmitting the call data to the other when the other handover process is completed without waiting for the completion of the other handover process. Only. The rest is the same as that of FIG.

  The process of the handover terminal when the handover process message is completed will be described with reference to FIG. Here, the operation when the handover process fails in the case shown in FIG. 7 will be described.

  In FIG. 8, in the communication network # 2, the throughput in the uplink (terminal-> server) direction is better than that in the downlink (server-> terminal), and in the communication network # 3, the throughput in the downlink direction is increased. The situation is assumed to be better than that. In FIG. 8, handover processing 203 and 204 are executed in both the handover terminal 105 and the handover server 100, and the handover processing 203 is completed. Thereafter, a completion message (M700) is notified to the handover server 100. Then, the call server 100 stops the call data transmission to the communication network # 2 (206). The flow of processing up to this point is the same as the flow of FIG.

  In this state, two-way call data is sent in communication network # 3, and call data in the upward (terminal-> server) direction is sent in communication network # 2 (D800). When the handover process 204 fails in the switching process to the communication network # 3 in the handover process 204, the handover server 100 transmits a handover failure notification message (M 701) to the handover terminal 105. Even if the handover terminal 105 receives the message (M701), it continues the handover process 203 and stops transmitting the call data to the communication network # 3 (205). In this state, the communication network # 2 is used as the communication path for the upstream communication, and the communication network # 3 is used for the downstream communication.

  If each communication network is stable in this state, it is possible to continue in this communication mode. If the uplink throughput of the communication network # 3 is recovered from this state, the handover server 100 transmits a call data retransmission request message via the communication network # 3 to the handover terminal 105 ( M800).

  When receiving the message (M800), the handover terminal 105 transmits the transmission data to the handover server 100 via the communication network # 3. In this state, two-way call data is sent in communication network # 3, and call data in the upward (terminal-> server) direction is sent in communication network # 2 (D802). During this state, the handover server 100 re-executes the handover process 204 and attempts to switch to the communication network # 3. Thereafter, the processing when the handover processing 204 is successful is the same as that in FIG.

  In the above description, the case where the handover process has failed in the handover server 100 has been described. However, the case where the handover process has failed in the handover terminal 105 is also applicable.

  As shown in FIG. 9, Embodiment 6 of the present invention has a NAT (Network Address Translator) traversal function in a handover system in which a handover terminal belongs to a private communication network and a handover server belongs to a global communication network. is there. For example, in FIG. 1, the communication network # 1 is the global communication network 903, and this corresponds to the case where at least one of the communication network # 2 and the communication network # 3 is the private communication network 904.

  As specific examples of the sixth embodiment, the communication network # 1 is the Internet, the communication network # 2 is a private wireless LAN communication network in a home or office, and the communication network # 3 is a global communication network connected to a mobile phone network. A configuration is typical. A router exists between the private communication network and the global communication network, and the router generally has a NAT function. NAT allows a single global IP address to be shared by multiple computers in the private communication network by transparently converting the IP address of the private communication network and the IP address of the global communication network. Technology. In the present invention, unless otherwise specified, NAT refers to NAPT (Network Address / Port Translation) for converting an IP address and a port number.

  In NAT, since the port number is changed, the computer on the global communication network cannot access the computer on the private communication network. For this reason, in order to transmit an SIP message or voice data from the handover server to the handover terminal, it is necessary to implement some method for exceeding NAT.

  There are several NAT traversal techniques. Here, an embodiment using a technique called UDP hole punching will be described with reference to FIGS. However, in these figures, only NAT traversal communication between the handover terminal and the handover server will be described. Communication with telephone terminals ahead of the handover server is omitted in these figures, but is the same as in the first embodiment.

  With reference to FIG. 10, a NAT message crossing method for SIP message exchange will be described. Both the handover terminal 902 and the handover server 900 have a function for performing NAT traversal. The handover terminal 902 creates a SIP socket (1001), and transmits a global address request message (M1000) to the handover server 900 using the socket (that is, the port number for SIP). Here, the socket is a descriptor that is created every time the communication program API (Application Program Interface) designates an IP address, port number, protocol, etc. to perform communication. Implements communication using an API. The global address means a pair of a global IP address and a port number assigned to a packet transmitted from the private communication network to the global communication network by the NAT.

  Message M1000 is transmitted as a UDP packet. Hereinafter, unless otherwise specified, it is assumed that all the messages shown in FIGS. 10 to 17 are transmitted as UDP packets, but the same applies to TCP packets. The port number of the transmission destination of the message M1000 is a port number dedicated for NAT traversal processing prepared by the handover server 900. This port number is set in advance in the handover server, and is a value known to the handover terminal. That is, in the handover server 900, it is assumed that a socket for NAT traversal processing is created in advance (not shown in FIG. 10).

  The handover server 900 acquires a global address (port number on NAT and global IP address) mapped to the message packet M1000, and notifies the handover terminal 902 of the global address by a message M1001.

  As described above, the handover server 900 can transmit the UDP packet storing the SIP message to the acquired global address, thereby transmitting the SIP message by UDP to the handover terminal 902 over the NAT. However, if there is no communication using this mapping for a certain period of time, the mapping information is deleted in the router. For this reason, the handover terminal 902 maintains mapping information by periodically transmitting dummy UDP packets to the global address with the SIP port number as the source (D1002). The handover server 900 simply discards the received dummy data.

  Next, a flow up to a call with the handover server by a call from the handover terminal will be described with reference to FIG. The handover terminal 902 generates a socket for RTP and RTCP when starting a call by executing a call action such as making a call by the terminal user (1101). The RTP socket is a socket for transmitting / receiving voice call data (RTP packet format), and the RTCP socket is a socket for transmitting / receiving statistical information (RTCP packet) regarding the transmission / reception result.

  Then, a global address request message (M1100) is sent to the handover server 900 using the RTP socket and the RTCP socket (that is, the RTP port number and the RTCP port number are the sources), respectively. Send. Here, M1100 means both a global address request message for RTP and a global address request message for RTCP.

  The destination port number is a port number dedicated to NAT traversal processing (same as in FIG. 10). The handover terminal 900 checks the global address mapped to this message M1100 and notifies the handover terminal 902 of this global address by the message M1101.

  Here, M1100 means both a global address return message for RTP and a global address return message for RTCP. The handover terminal 902 acquires the port number in the received global address, and generates an SDP (Session Description Protocol) message including this (1102). SDP is a message protocol defined in RFC2327 for describing session information.

  An example of this SDP message is shown in FIG. In the SDP message shown in the figure, the line “m =” is essential for describing the media type and transport information. The second value (= 49000) in the “m =” row represents the port number for receiving this medium (ie, voice). The value of this port number is set as the port number notified by the message M1101. Other lines are omitted because they are not directly related to the contents of the present invention, but refer to RFC2327 for details.

  The handover terminal 902 transmits an INVITE message M1102 to the handover server 900 using the SIP port number already described in FIG. Subsequently, in FIGS. 11 to 17, the SIP port numbers described in FIG. 10 are used for all the SIP messages transmitted from the handover terminal 902 to the handover server 900. The INVITE message (M1102) includes the SDP message (1102) described above, and the handover server 900 acquires the port number in the SDP message. After that, the handover server 900 creates an RTP / RTCP socket (1103), creates an SDP message including its own RTP / RTCP port number, and transmits a RINGING (M1103) after preparation for reception is completed.

  The transmission destination of RINGING is a global address for SIP message that has been acquired in the processing described in FIG. Thereafter, in FIGS. 11 to 17, the destination of the SIP message transmitted from the handover server 900 to the handover terminal 902 is the SIP message previously notified from the handover server by the process described in FIG. 10 or FIG. 13 (described later). Global address for Thereafter, when the handover server 900 completes preparation for a response to the handover terminal 902, OK (M1104) is transmitted.

  The handover terminal 902 then returns an ACK (M1105) to the handover server 900. As described above, a call session is established, and transmission / reception of call data is started (D1101). At this time, the transmission destination IP address of the call data transmitted from the handover server 900 to the handover terminal 902 is a global IP address in the global address notified in M1101.

  The destination port number of the call data (RTP packet) transmitted from the handover server 900 to the handover terminal 902 is the port number indicated by the SDP message in INVITE (M1102). However, normally, the destination port number of the RTCP packet corresponding to the call data (RTP packet) is a value obtained by adding 1 to the RTP port number. As described above, NAT traversal of a call when a call is made from a handover terminal can be realized.

  FIG. 12 shows a call sequence when the handover server makes a call. Compared to FIG. 11, since the call origination starts from the handover server 900, the order of RTP / RTCP socket generation (1101, 1103) and SDP message creation (1102, 1104) is different, and the direction of the SIP message is different. Other than that, the same applies.

  In FIG. 11 and FIG. 12, message exchange (M1100, M1101) for global address acquisition was performed after the call from the terminal side and the start of the call from the server side, respectively. However, in order to realize seamless handover, it is necessary to establish a call on the communication network after switching and start a voice call while the data buffer via the communication network before switching is not exhausted. Therefore, it is desirable that the time required from the call origination to the start of the call is short. Especially when the delay of the communication network is large, it is desirable that the message exchange after the call is as small as possible. An example of the flow of processing for reducing message exchange will be described with reference to FIG.

  In FIG. 17, after the NAT traversal of SIP communication in FIG. 10 is completed, both the RTP and RTCP sockets are generated immediately without waiting for the handover server and the handover terminal to make and receive calls. Immediately thereafter, the handover terminal 902 transmits a global address request message (M1100) for RTP and RTCP communication, and the handover server 900 returns a global address (M1101).

  Thereafter, the handover terminal 902 maintains a mapping in the NAT 901 by periodically transmitting a dummy packet for hole punching to the handover server 900 using the RTP and RTCP ports as the source port numbers (D1003). As described above, by acquiring the global address before the actual call starts, it is not necessary to acquire the global address in FIGS. 11 and 12, and thus the time from the call to the start of the call can be shortened. .

  Since the handover terminal 902 normally has one party to call (handover server 900), it is easy to generate sockets for RTP and RTCP in advance. However, the handover terminal 900 makes a call with an unspecified number of handover terminals. For this reason, the handover terminal 900 requires management such as assuming a plurality of handover terminals in advance, preparing N sets of RTP and RTCP sockets, and preparing additional N sets if they become insufficient.

  In the example of NAT traversal described so far, in FIG. 10, the NAT processing port number and the SIP port number in the handover server 900 are different. Similarly, the NAT processing port number and the RTP and RTCP port numbers in FIGS. 11 and 12 are different. That is, in FIGS. 10 to 13, when transmitting a packet from the handover server 900 to the NAT 901, it is assumed that the destination port number of the handover terminal 902 converted by the NAT 901 has the same value regardless of the source port number. It was.

  However, depending on the type of NAT, the above assumption may not hold. Examples of this include a symmetrical NAT and a port-limited cone NAT. In these NATs, restrictions are imposed so that only an external host having an IP address and a port address transmitted by a private network internal host to the global network can communicate with the internal host.

  That is, in the example of FIG. 12, the source port number of the message M1101 transmitted from the handover server 900 to the NAT 901 is the NAT processing port number (# P1), and the destination port number (# P2) of M1101 is The NAT 901 converts it to the RTP port number (# P3) of the handover terminal 902.

  At this time, in the case of exceeding NAT in FIG. 12, the handover server 900 uses the port number for RTP (# P4) different from the port number (# P2) as the transmission source port number and transmits the packet to the NAT 901. Conversion to (# P3). However, in the symmetric NAT, in order to convert to the RTP port number, the source port number of the packet sent from the handover server 900 needs to be the port number (# P2).

  Therefore, in the case of FIG. 12, in order to perform NAT traversal assuming the symmetric NAT, the following is necessary. That is, the destination port number when requesting the global address when receiving the RTP packet from the handover terminal 902 needs to be the source port number (# P4) when the handover server 900 actually transmits the RTP packet. is there.

  Hereinafter, the flow of NAT traversal processing assuming a symmetric NAT will be described with reference to FIGS.

  With reference to FIG. 13, processing for NAT traversal of SIP communication will be described. The difference from FIG. 10 is that the destination port number of the global address request message (M1000) transmitted from the handover terminal 902 is a SIP processing port number predetermined in the handover server 900. In the handover server 900, a SIP socket designated with this port number is created in advance (1002), and the global address is returned using this SIP socket (that is, using the SIP processing port number as the source port number). A message (M1001) is transmitted.

  As a result, in the symmetric NAT (1301), M1001 is correctly converted into the SIP port number of the handover terminal 902. Subsequent hole punching packet transmission (D1002) is the same as in FIG. Therefore, in the configuration of FIG. 13, in the handover server 900, the data received through the SIP socket includes not only the SIP message but also the global address request message (M1100) prior to the SIP message. Therefore, the handover server 900 needs to distinguish the above two types of messages.

  With reference to FIG. 14, a description will be given of the symmetric NAT traversal process when the handover terminal 902 is called. 11 is different from FIG. 11 in that, for example, the RTP / RTCP port number in the handover server 900 is set in advance prior to the handover terminal 902 requesting a global address (M1100) and is known to the handover terminal 902. And that an RTP transmission socket and an RTCP transmission socket in which the port number is designated are created in advance (1103). The other is that the destination port number of the global address request message M1100 transmitted by the handover terminal 902 and the source port number of the global address return message M1101 transmitted by the handover server 900 are for RTP and RTCP transmission in the handover server 900. It is a port number.

  As a result, the packets transmitted from the RTP and RTCP transmission sockets are converted into the RTP reception port number and the RTCP reception port number of the handover terminal 902 by the symmetric NAT 1301, respectively, and NAT traversal is realized. . The subsequent processing is the same as the processing described in FIG. The handover server 900 can process the data received by the RTP socket and the RTCP socket separately from the RTP packet and the RTCP packet as well as the global address request message (M1100). There is a need.

  FIG. 15 is a modification of the symmetric NAT traversal process at the time of the terminal side call described in FIG. In FIG. 15, unlike FIG. 14, it is assumed that the RTP and RTCP transmission port numbers set in advance in the handover server 900 are not informed to the handover terminal 902 or are dynamically set. . In this case, in FIG. 15, the handover terminal 902 performs a process of requesting the RTP and RTCP port numbers from the handover server 900 (M1500), and returning these RTP and RTCP port numbers (M1501). The destination port number of the message M1500 and the source port number of the message M1501 are port numbers determined in advance between the handover server and the terminal for NAT processing. Thereafter, as in FIG. 14, a global address request message (M1100) is transmitted to each of the RTP and RTCP port numbers. The subsequent processing is the same as the processing described in FIG.

  FIG. 16 is a diagram showing a processing example of symmetric NAT traversal at the time of terminal-side incoming call. The difference from FIG. 12 is that the destination port number of the global address request message (M1100) transmitted from the handover terminal 902 and the source port number of the global address return message (M1101) transmitted from the handover server 900 are different from each other. This is a port number for RTP and RTCP in the server 900.

  As a result, the destination port numbers of the RTP and RTCP packets transmitted from the handover server 900 are converted into the RTP port number and the RTCP port number of the handover terminal 902 in the symmetric NAT 1301, respectively, and transmitted to the handover terminal 902. The In FIG. 16, it is assumed that the RTP and RTCP port numbers in the handover server 900 are known in the handover terminal 902. If these are not known by the handover terminal 902, message exchange (M1500, M1501) for acquiring the RTP and RTCP port numbers in the handover server 900 is performed in advance as in the method described in FIG. Accordingly, the handover terminal 902 can acquire the RTP and RTCP port numbers.

  As described above, the normal (non-restricted) NAT and the NAT traversal processing of the symmetric NAT have been described using the UDP hole punching as an example, but the present invention does not necessarily use the UDP traversal method as the NAT traversal technique. It is not limited to hole punching. For example, if the router between the private communication network and the global communication network is a router that supports UPnP (Universal Plug and Play), UPnP is used, and the handover terminal sets port mapping settings for NAT to NAT. Do it. As a result, it is possible to achieve NAT traversal in the same manner as described above by acquiring a global address mapped by NAT and transmitting this global address to the handover server.

  Embodiment 7 of the present invention is a handover system for performing auditory voice quality degradation correction processing (hereinafter also referred to as concealment processing) when a seamless handover cannot be realized in a handover terminal in the telephone handover system of Embodiment 1. It is.

  FIG. 18 shows the configuration of a handover terminal according to the seventh embodiment of the present invention. An error correction unit (a concealment unit) 1802 in the figure is a part newly added to the first embodiment, and the other parts are the same as those of the first embodiment. The decoder 1801 acquires call data D103a (voice frame data) according to the voice data rate, decodes it, and generates sample data (voice waveform data). Therefore, even if there is a temporal discontinuity (difference) in the call data D103a passed to the decoder 1801 due to packet loss at the time of communication, the error concealment unit 1802 is continuously transmitted regardless of the reproduction time. Passed as D1810. The D1810 includes reproduction time information of the sample data in addition to the audio sample data. The error concealment unit 1802 performs processing on the sample data output from the decoder 1801 and passes the processed data to the reproduction buffer as sample data D1811.

  When the seamless handover fails, when the call data D103a passed to the decoder 1801 is switched between the buffer 301 and the buffer 302 at the time of handover by the handover control unit 1805, there is a discontinuous part in the playback time. Occurs at certain times.

  Processing in the error concealment unit 1802 when discontinuity occurs during handover will be described with reference to FIGS. 19 and 20.

FIG. 19 shows processing when the call data playback time in the buffer 301 is later than that in the buffer 302 when handing over from the buffer 301 to the buffer 302. In this case, as shown in the figure, the speech waveform corresponding to each speech frame has a time gap. As processing in the error concealment unit 1802 in this case, for example, one of the following processing is performed.
(1) Packing discontinuous data: Although there is a problem that the playback time is skipped, data switching is completed easily and instantaneously.
(2) Silence state: Although there is an advantage that it is simple and the reproduction time is not skipped, the auditory deterioration due to the silent part is large.
(3) Replacing with noise or another sound: Compared with (2), auditory deterioration can be reduced by inserting a noise sound into the gap. Examples of the noise include white noise noise having the same amplitude as the previous frame data. In addition, when the discontinuous time is long, it is possible to reduce the user's discomfort by replacing the gap portion with sound such as a melody sound or a message instead of noise.
(4) Replace with front and rear voices: The entire gap is covered by replacing the waveforms before and after the missing part. Further, it is possible to take an average value for the boundary portion of the waveform from both ends. If the discontinuous time is relatively short, auditory deterioration may be avoided by simply replacing the sound with the preceding and following sounds.
(5) Stretching the sound before and after: Covering the entire gap by stretching the waveform before and after the missing part. For example, the pitch period is extended so as to cover the missing part from both ends of the missing part, and the average value is taken for the part where the waveform from both ends collides (for details of this, see “H. A. Stenger, KBYounes, and B. Girod. See New Technique for Audio Packet Loss Concealment, IEEE Global Internet Symposium, London, December 1996.).

FIG. 20 shows processing when the call data playback time in the buffer 301 is ahead of that in the buffer 302 when handing over from the buffer 301 to the buffer 302. In this case, as shown in the figure, the speech waveform corresponding to each speech frame is a time-reversed one. As processing in the error concealment unit 1802 in this case, for example, any of the following processing can be performed.
(1) Packing discontinuous data: Although there is a problem that data that has been played back in the past is played back, the data switching ends easily and instantaneously.
(2) Silence state until data at the next time is received: There is an advantage that it is simple and the reproduction time is not skipped, but auditory deterioration due to the silent portion is large.
(3) Replace with noise or another sound until the next time data is received: Compared with (2), by inserting the noise sound into the gap, auditory deterioration can be reduced. Examples of the noise include white noise noise having the same amplitude as the previous frame data. In addition, when the discontinuous time is long, it is possible to reduce the user's discomfort by replacing the gap portion with sound such as a melody sound or a message instead of noise.
(4) Replacing with the previous voice until the next time data arrives: The entire gap is covered by repeatedly replacing the waveform before the missing part as it is. Further, it is possible to take an average value for the boundary portion of the waveform from both ends. In the case where the discontinuous time is relatively short, auditory deterioration may be avoided by simply replacing the sound with the front and back sounds.
(5) Extend the previous sound until the next time data comes: Cover the entire gap by stretching the waveform before and after the missing part. For example, the pitch period is extended so as to cover the missing portion at a predetermined rate from before and after the missing portion. Since the next data arrives late, if the missing part cannot be covered, it may be silenced or gently faded out.

  Embodiment 8 of the present invention is a handover system in which a handover terminal in the telephone handover system of Embodiment 1 determines a handover by acquiring terminal location information from the outside.

  FIG. 22 shows the configuration of the handover terminal 105 according to Embodiment 8 of the present invention. In the eighth embodiment, it is assumed that a location server 2201 that provides location information to a handover terminal is connected to the communication network # 3. However, the communication network to which the location server 2201 is connected is not necessarily the communication network # 3 as long as it can be connected from the handover terminal 105 via some communication network. Hereinafter, a flow in which the handover terminal 105 acquires location information and determines a handover will be described with reference to FIG.

The location information acquisition process is performed by the location information acquisition unit 2202 in the handover processing unit 111 in the handover terminal 105. The position information acquisition unit 2202 periodically exchanges messages with the location server 2201 (M2210, M2211), and finally acquires position information.
A method for exchanging messages and a positioning method between the location server and the position information acquisition unit 2202 are not particularly limited in the present invention. The position information may be passed from the location server, or may be calculated on the terminal side based on the information from the location server. The positioning method is also a method using GPS satellites, a method for specifying position information by regarding the base station as a satellite, a method for performing positioning based on information such as transmission speed, delay, and radio wave intensity between multiple base stations. But you can.

  The location information acquisition unit 2202 acquires the location information of the terminal by some method, and then passes a message M2212 including this location information to the mapping information management unit 2203. Upon receiving the position information in the message M2212, the mapping information management unit 2203 determines whether to use the communication network # 2 or the communication network # 3 at the position indicated by the position information. Assume that the location information and the mapping information of communication network selection for this determination are set in advance in the terminal by some method.

  The mapping information management unit 2203 passes the message M2213 including the determination result (that is, which communication network the terminal uses) to the location information acquisition unit 2202. This message M2213 is passed from the position information acquisition unit 2202 to the handover control unit 1805. The handover control unit 2213 always keeps track of which communication is currently being performed in the communication network # 2 or the communication network # 3. In the handover control unit 2213, when the currently communicating communication network is different from the communication network sent by the message M2213, the handover is determined (201 in FIG. 2). The processing after the handover decision is the same as the flow shown in FIG.

  As described above, according to the telephone handover system according to the present embodiment, the video and the sound are temporarily interrupted even in the middle of communication or in a bidirectional communication service such as a telephone service including video. It is prevented. Therefore, it is possible to switch the communication network without such interruption, and the user can continue the communication service without suffering the disadvantage of interruption.

  In this embodiment, since the codec and the bit rate can be changed in each session, a call with a bit rate corresponding to the current throughput in each communication network can be realized in a timely manner.

  Further, in this embodiment, it is possible to continue a call between a private communication network via a router and a global communication network.

  Further, according to the present invention, even when a call before and after switching is discontinuous due to a delay in the communication network or the like, the subjective deterioration of the voice at the time of switching the call can be reduced by performing the concealment process.

  As described above, according to the present invention, even in a plurality of communication networks having different characteristics and a communication network in which the communication speed fluctuates greatly, a bidirectional communication service having a high real-time requirement can be obtained without temporary interruption. It can be provided continuously. Therefore, it is possible to provide a highly convenient service to the user by using a portable terminal that is widely spread now.

The block diagram which shows the structural example of the telephone handover system by Embodiment 1 of this invention The figure which shows the example of the processing flow of the telephone handover system by Embodiment 1 of this invention. The figure which shows the structural example of the hand-over process part in the hand-over terminal of the telephone hand-over system by Embodiment 1 of this invention. The figure which shows the structural example of the hand-over process part in the hand-over server of the telephone hand-over system by Embodiment 1 of this invention. The figure which shows the 1st example of the processing flow at the time of the handover failure of the telephone handover system by Embodiment 5 of this invention The figure which shows the 2nd example of the processing flow at the time of the handover failure of the telephone handover system by Embodiment 5 of this invention. The figure which shows the 3rd example of the processing flow at the time of the handover failure of the telephone handover system by Embodiment 5 of this invention. The figure which shows the 4th example of the processing flow at the time of the handover failure of the telephone handover system by Embodiment 5 of this invention. The figure which shows the structural example of the telephone handover system by Embodiment 6 of this invention. The figure which shows the example of the process sequence which performs NAT traversal of SIP message of the telephone handover system by Embodiment 6 of this invention. The figure which shows the example of the process sequence which performs NAT crossing of the voice call in the call of the hand-over terminal side of the telephone handover system by Embodiment 6 of this invention. The figure which shows the example of the process sequence which performs NAT traversal of the voice call in the call of the telephone hand-over system by Embodiment 6 of this invention at the hand-over terminal side call. The figure which shows the example of the process sequence which performs the symmetric NAT traversal of a SIP message of the telephone handover system by Embodiment 6 of this invention. The figure which shows the 1st example of the process sequence which performs the symmetric NAT crossing of the voice call in the call of the hand-over terminal side of the telephone handover system by Embodiment 6 of this invention. The figure which shows the 2nd example of the process sequence which performs the symmetric NAT crossing of the voice call in the call of the hand-over terminal side of the telephone handover system by Embodiment 6 of this invention. The figure which shows the example of the process sequence which performs the symmetric NAT crossing of the voice call in the telephone hand-over system by Embodiment 6 of this invention at the hand-over terminal side call. The figure which shows the example of the process sequence which acquires the global address for voice calls beforehand of the telephone handover system by Embodiment 6 of this invention. The figure which shows the structural example of the handover terminal by Embodiment 7 of this invention. The figure which shows the 1st example of operation | movement in the error concealment process in the handover terminal by Embodiment 7 of this invention. The figure which shows the 2nd example of operation | movement in the error concealment process in the handover terminal by Embodiment 7 of this invention. The figure which shows the example of the SDP message in embodiment of this invention Configuration example for position information acquisition processing according to embodiment 8 of the present invention

Explanation of symbols

100, 900: Telephone handover server 101: Handover processing unit (server)
102 ... Call processing unit (server)
103 ... Communication network interface unit (server)
104 ... Telephone terminals 105, 902 ... Handover terminals 106, 107, 108 ... Communication network 109 ... Communication network interface unit (terminal)
110: Communication processing unit (terminal)
111 ... Handover processing unit (terminal)
112 ... Voice processing unit (terminal)
901 ... NAT
1301 ... Symmetric NAT
2201 Location server

Claims (22)

In a handover server for handing over a call between a first terminal and a second terminal,
A call processing unit for establishing a call with the first terminal in at least one of the first communication network and the second communication network;
Call data transmitted from the second terminal is simultaneously transferred to the first terminal via the first and second communication networks, and is received via the first and second communication networks from the first terminal. A handover processing unit for storing the received call data in a buffer and transferring the call data to the above-mentioned two terminals;
The handover processing unit transfers the call data to the second terminal based on time information of the call data from the first terminal received from both the first and second communication networks. Switch the communication network used for
A handover server characterized in that when the switching fails, a process of either a telephone call through the communication network before the switching or the switching retry is executed. In a handover server for handing over a call between a first terminal and a second terminal,
A call processing unit for establishing a call with the first terminal in at least one of the first communication network and the second communication network;
Call data transmitted from the second terminal is simultaneously transferred to the first terminal via the first and second communication networks, and is received via the first and second communication networks from the first terminal. A handover processing unit for storing the received call data in a buffer and transferring the call data to the above-mentioned two terminals;
The handover processing unit performs switching between the first and second communication networks based on time information of call data from the first terminal received from both the first and second communication networks, Transferring the call data to the second terminal via the switched communication network;
When the call data is transmitted from the first terminal to the second terminal, the first communication network is used. When the call data from the second terminal to the first terminal is transmitted, the second communication network is used. Feature handover server.   The handover server according to claim 2, wherein the first communication network is a public communication network, and the first communication network is a wireless LAN. 2. The handover server according to claim 1, wherein the call processing unit is at least one of a port number for NAT traversal processing, a port number for call establishment, a port number for call data, and a port number for call control. A receiver for receiving the sent global address request message;
A transmission unit that acquires a global address of a sender from a message received by the reception unit, and transmits a message including the global address to the global address;
A handover server comprising:   5. The handover server according to claim 4, wherein the call processing unit sends a message including a call data port number of the handover server or a call control port number in response to a request from an external mobile communication terminal. A handover server that transmits to a requesting mobile communication terminal.   5. The handover server according to claim 4, wherein the call processing unit reserves at least one global port number in advance before starting a call and an incoming call, and the global address having the port number as a destination. A handover server characterized by receiving a request message.   5. The handover server according to claim 4, wherein the call processing unit receives a dummy packet having the same source port number as the message requesting the global address from an external mobile communication terminal and discards the dummy packet. A handover server. In a mobile communication terminal that communicates with a handover server,
A communication network interface capable of communicating via the handover server and at least one of the first communication network and the second communication network;
A call processing unit that establishes at least two calls in parallel with the handover server via the first and second communication networks;
A handover processing unit for transmitting transmission voice data in parallel to the first and second communication networks, and storing call data received from the first and second communication networks in a buffer;
A voice processing unit that decodes and reproduces the call data stored in the buffer of the handover processing unit;
The handover processing unit switches communication networks used when receiving the call data based on time information of the call data from the handover server received from both the first and second communication networks. And
A mobile communication terminal characterized in that, when the switching fails, a process of either a telephone call via the communication network before the switching or the switching retry is executed. In a mobile communication terminal that communicates with a handover server,
A communication network interface capable of communicating via the handover server and at least one of the first communication network and the second communication network;
A call processing unit that establishes at least two calls in parallel with the handover server via the first and second communication networks;
A handover processing unit for transmitting transmission voice data in parallel to the first and second communication networks, and storing call data received from the first and second communication networks in a buffer;
A voice processing unit that decodes and reproduces the call data stored in the buffer of the handover processing unit;
The handover processing unit switches communication networks used when receiving the call data based on time information of the call data from the handover server received from both the first and second communication networks. And
A mobile communication terminal using the first communication network when transmitting communication data to the handover server and using the second communication network when receiving communication data from the handover server. 9. The mobile communication terminal according to claim 8, wherein the call processing unit uses a known NAT traversal port number in the handover server as a destination, a call establishment global address, a call data global address. , Request and get one of the global addresses for call control,
A mobile communication terminal characterized in that a global port number of the corresponding mobile communication terminal is acquired from the acquired global address and notified to the handover server. 9. The mobile communication terminal according to claim 8, wherein the call processing unit includes a known call establishment port number, call data port number, call control port number or call data port number in the handover server. Request and obtain a global address corresponding to the destination,
A mobile communication terminal characterized in that a global port number of the mobile communication terminal is acquired from the acquired global address and notified to the handover server. The mobile communication terminal according to claim 8, wherein the call processing unit requests and acquires a port number for call control of the handover server from the handover server,
Using the acquired call control port number as the destination, requesting and acquiring the call control global address of the mobile communication terminal,
A mobile communication terminal, wherein the call control global port number of the mobile communication terminal is acquired from the acquired call control global address and notified to the handover server.   The mobile communication terminal according to claim 10, wherein the call processing unit acquires the global port number in advance before outgoing calls and incoming calls are started.   11. The mobile communication terminal according to claim 10, wherein the call processing unit periodically transmits a dummy packet having the same source port number as the message requesting the global address to the same destination port number as the message. A mobile communication terminal characterized by continuing to perform. The mobile communication terminal according to claim 8, wherein
The handover processing unit includes a handover control unit that switches and outputs the call data stored in the buffer according to the reproduction time of the call data received from the first and second communication networks,
The voice processing unit includes a decoder that decodes the call data from the handover control unit, and an error correction unit that receives the call data decoded by the decoder,
The mobile communication terminal, wherein the error correction unit corrects a difference in reproduction time before and after switching of the decoded call data.   16. The mobile communication terminal according to claim 15, wherein when the reproduction time of the call data after switching is ahead of that before switching, the error correcting unit converts the call data after switching to the call data before switching. The mobile communication terminal is modified so as to be played back at a time immediately after.   16. The mobile communication terminal according to claim 15, wherein, when the call data playback time after switching is ahead of that before switching, the error correction unit starts from the call data playback time before switching after the switching. A mobile communication terminal characterized by interpolating with silence or other voices until the call data playback time.   16. The mobile communication terminal according to claim 15, wherein the other voice includes sound data prepared in advance, sound data created based on received call data, and one of the calls before or after the switching. A telephone terminal characterized in that the telephone terminal is generated based on data or sound data obtained by expanding the call data in terms of time.   16. The mobile communication terminal according to claim 15, wherein when the reproduction time of the call data after switching is later than that before switching, the error correcting unit converts the call data after switching to the call data before switching. The mobile communication terminal is modified so as to be played back at a time immediately after.   16. The mobile communication terminal according to claim 15, wherein the error correction unit, when the reproduction time of the call data after switching is later than that before the switching, calls data after the reproduction time of the call data before switching. Until a signal is received from the decoder, a mobile communication terminal characterized by interpolating with silence or other voices and using the received call data as the first call data after switching. The mobile communication terminal according to claim 20,
The other voices are obtained by temporally expanding sound data prepared in advance, sound data created based on received call data, call data before switching, or call data before and after switching. A mobile communication terminal generated based on the sound data. 9. The mobile communication terminal according to claim 8, wherein the handover processing unit exchanges information with a location server via one of the first and second communication networks to thereby obtain position information of the mobile communication terminal. A position information acquisition unit for acquiring
The location information acquisition unit determines a communication path of the telephone terminal in the location information using preset mapping information and the location information,
A mobile communication terminal, wherein switching between the first and second communication networks is determined based on the determination result.

Images