JP2005252665A - Voice packet transferring method and terminal used for the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a voice packet transferring method that maximumly relaxes the load of communication equipment within the range of no troubles in voice communication in a normal network state and guarantees voice communication most preferentially when network quality deteriorates, and to provide a terminal used for the voice packet transferring method. <P>SOLUTION: A VoIP (voice over Internet protocol) terminal receives a voice packet, calculates an interpolation rate, determines its value (ST31 to ST35) and further controls so as to dynamically change or maintain a packeting interval in accordance with the value of the calculated interpolation rate (ST36 to ST42). When the packeting interval is changed, a receiving interval stored in a packeting interval database is updated to the changed value (ST43), and a call connection request change signal is transmitted (step ST44). <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a voice packet transfer method and a terminal used therefor, and particularly in an IP telephony network connected to an IP network represented by the Internet / intranet, a soft phone, an IP phone, an IP-PBX, a VoIP (Voice over Internet). The present invention relates to a voice packet transfer method using a VoIP terminal that performs signaling conversion and voice packet encoding, such as a terminal adapter / gateway for (Protocol), and a terminal used therefor.

  In an IP telephony network connected to an IP (Internet Protocol) network represented by a conventional Internet / intranet, a soft phone, an IP phone, an IP-PBX, a terminal adapter / gateway for VoIP (Voice over Internet Protocol), etc. In a voice packet transfer system using a VoIP terminal that performs signaling conversion and voice packet coding, a voice band is secured on a line with a slow line speed and a narrow transmission band, and voice quality is maintained by QoS (Quality of Service). It was. Therefore, it has been difficult to dynamically change the bandwidth per call in terms of network design.

  However, with the rapid spread of broadband communication, it has become an effective environment to reduce the processing load on communication devices and transfer a large amount of data traffic rather than guaranteeing excessive bandwidth. In order to reduce the processing load of communication equipment and transfer a large amount of data traffic, it is effective to monitor the interpolation rate, discard rate, and fluctuation of voice packets and change the packetization interval and packet size of voice packets.

  As a conventional method of changing the packet interval, the transmission side notifies the packet interval to the reception side, and the reception side receives this notification, detects congestion, notifies the transmission side of this congestion state, and transmits. A packet transfer method for changing the packet interval is known on the side (see, for example, Patent Document 1). Also, a packet transfer method in which a packet interval is notified from the transmission side to the reception side, the reception side receives this notification, detects congestion, notifies the transmission side of this congestion state, and the transmission side changes the packet interval. Conventionally known (see, for example, Patent Document 2). On the other hand, as a conventional method for changing the packet size, a packet transfer method is known in which the reception side calculates the packet loss rate and delay fluctuation, sends this data to the transmission side, and the transmission side changes the transmission packet size. (For example, see Patent Document 3).

JP-A-7-303117 Japanese Patent No. 3240240 Japanese Patent Laid-Open No. 11-177623

  However, each of the conventional packet transfer methods described in Patent Documents 1 and 2 described above is configured to change the packet transmission interval when congestion occurs or to change the packet size to lower the bit rate, which reduces the throughput. For this purpose, there is a problem that real-time characteristics are not considered.

  In addition, the conventional packet transfer method described in Patent Document 3 changes the transmission interval and the packet size according to the network status, but determines the packet transmission interval and the packet size only from the network status. However, there is a question as to whether or not there is real-time performance in terms of assuming data retransmission.

  The present invention has been made in view of the above points, and a voice packet transfer method capable of performing voice packet transfer without impairing real-time performance by simultaneously changing the transmission interval and the packet size while maintaining the characteristics of real-time traffic. And it aims at providing the terminal used for it.

  Another object of the present invention is to reduce the load on the communication device to the maximum extent that does not interfere with the voice call when the network is normal, and to give the highest priority to the voice call when the network quality deteriorates. An object of the present invention is to provide a voice packet transfer method for guaranteeing and a terminal used therefor.

  In order to achieve the above object, a first invention is a call control / management server in which a plurality of terminals having voice packet encoding processing, signaling conversion and switching functions perform call control signal relay processing and centralized management. A voice packet transfer method for a system that is bi-directionally connected via a communication network to a voice server that sends a hold tone and a guidance in response to a request from the terminal. A first step of periodically transmitting voice packets to a counterpart terminal at a transmission packetization interval obtained by referring to its own database, and calculating an interpolation rate of voice packets received via the communication network According to the second step, the third step of changing the packetization interval of voice packets to be transmitted according to the interpolation rate, and the changed packetization interval, From the fourth step of changing the packetization interval for transmission of its own database, the fifth step of transmitting a call connection change request of the changed packetization interval to the counterpart terminal, and the received call connection change request And a sixth step of extracting the changed packetization interval and updating the packetization interval for transmission in its own database to the extracted packetization interval.

  In general, when the packetization interval of voice packets is long, the voice packet length included in one packet is large, so the overhead of the IP packet is small and the load on the VoIP terminal or relay device is small. However, it is weak to the quality in the communication network (network), and a plurality of voice packets are discarded by discarding one packet. On the other hand, when the packetization interval of voice packets is short, the overhead as IP packets is large and the load on the apparatus is large. On the other hand, it is strong in the network quality.

  In view of the above points, the present invention obtains a result of dividing the number of packet interpolations that occur when packets cannot be absorbed by the jitter buffer due to packet discard or delay / fluctuation in the communication network by the total number of received voice packets. By dynamically changing the transmission packetization interval according to the interpolation rate, the voice packet packetization interval can be changed according to the quality of the quality in the network corresponding to the interpolation rate.

  In order to achieve the above object, in the second invention, the third step in the first invention is changed to a preset minimum value when the interpolation rate is not less than the first predetermined value. When the interpolation rate is less than the first predetermined value and greater than or equal to a second predetermined value that is smaller than the first predetermined value, the smaller the interpolation rate value, the smaller the packetization interval is, and the interpolation rate becomes the second value. When the packetization interval is less than the predetermined value or the packetization interval is the minimum value, the packetization interval is not changed, and the current value is maintained. In the present invention, when the interpolation rate is high, it is determined that there is a problem in the line quality, and the packetization interval is decreased by a certain value within a range equal to or larger than a preset minimum value. Voice packets can be transferred with the packet size.

  In order to achieve the above object, the third aspect of the invention relates to the third step of the first aspect of the invention, wherein when the interpolation rate is 0%, the packetization interval is not the maximum value set in advance. The interval is increased by a certain value, and when the packetization interval is the maximum value, the current value is maintained. In the present invention, even if the interpolation rate is the lowest and the line quality is good, the packetization interval can be increased by a certain value within a range not exceeding the preset maximum value.

  Furthermore, in order to achieve the above-mentioned object, the fourth aspect of the invention includes the first step in the first aspect of the invention, wherein the call connection request source is selected from the own database in which various parameters are set for each counterpart terminal. The voice packet is periodically transmitted to the counterpart terminal at the transmission packetization interval determined for the counterpart terminal. According to the present invention, an allowable maximum delay value, which is one of the data base parameters, can be set in advance according to the communication partner terminal.

  In order to achieve the above object, a terminal according to a fifth aspect of the present invention provides a communication network to a call control / management server that performs call control signal relay processing and centralized management, and a voice server that sends hold tone and guidance, respectively. Is a terminal that is bidirectionally connected to transmit signaling and voice packets for voice call connection, a voice interface indicating a voice-side physical interface, a call signal from the voice interface, and a communication network side Based on the incoming call signal, a voice packet is obtained by encoding or decoding a PCM signal transmitted / received between a call control / management unit that executes call processing and call management, such as call connection processing and call state management, and a voice interface. The voice packet processing unit that executes the conversion processing to be converted to the voice packet received from the communication network The voice packet is delivered to the voice packet processing unit periodically, and if the voice packet does not arrive, interpolation processing is performed, and a jitter processing unit that outputs statistical information such as the degree of voice packet fluctuation and the number of interpolations is received from the jitter processing unit. A statistical processing unit that calculates the statistical information obtained, determines an optimal packetization interval, stores it in the packetization interval database for each partner, a protocol processing unit that performs predetermined protocol processing, and a network interface that connects to the communication network The statistical processing unit calculates the interpolation rate of the received voice packet, updates the packetization interval database by changing the packetization interval of the voice packet to be transmitted according to the interpolation rate, and performs call control.・ The management unit sends a call connection request change signal to the partner terminal and Characterized in that to perform an instruction to change the socket of intervals.

  In the present invention, the number of packet interpolations that occur when packets cannot be absorbed by the jitter buffer due to packet discard or delay / fluctuation within the communication network is divided by the total number of received voice packets in the statistical processing unit. The transmission packetization interval can be dynamically changed according to the calculated interpolation rate.

  In order to achieve the above object, according to a sixth aspect, in the packetization interval database according to the fifth aspect, the packetization interval for transmission, the packetization interval for reception, the monitoring cycle, the allowable delay for each counterpart terminal The threshold of the interpolation rate is set, and the statistical processing unit calculates the interpolation rate based on the number of interpolations within the monitoring period, and when the interpolation rate is equal to or greater than the threshold of the interpolation rate, the transmission packetization interval is set in advance. When the interpolation rate is less than the interpolation rate threshold, the transmission packetization interval is reduced by a certain value, and when the interpolation rate is 0%, the transmission packetization interval is set to the preset maximum value. The packetization interval database is updated by making a change to increase the constant value within the range of the value or less.

  In the present invention, when the interpolation rate is high, it is determined that there is a problem in the line quality, and the packetization interval is decreased by a certain value within a range equal to or larger than a preset minimum value. Voice packets can be transferred with the packet size, and when the interpolation rate is 0%, if the packetization interval is not the preset maximum value, the packetization interval is increased by a certain value, and the packetization interval is the maximum value. In this case, since the current value is maintained, even when the interpolation rate is the lowest and the channel quality is good, the packetization interval of the voice packet to be transmitted can be prevented from exceeding the preset maximum value.

  According to the present invention, the interpolation rate obtained by dividing the number of packet interpolations that occur when packets cannot be absorbed by the jitter buffer due to packet discard or delay / fluctuation in the communication network by the total number of received voice packets. Accordingly, the packetization interval of the voice packet to be transmitted is dynamically changed. Therefore, when the interpolation rate is low and the quality in the network is good, it is preferable that the voice packetization interval is long within the allowable delay range. Normally, the interval is long, but if the quality in the network deteriorates, the interpolation rate increases, so the packetization interval can be automatically shortened to adjust the direction to increase the durability of the quality, and the interpolation rate When the network quality is improved and the quality in the network is improved, the processing load can be automatically reduced by automatically increasing the packetization interval.

  In addition, according to the present invention, even if the line quality is good, the packetization interval is increased by a certain value within a range that does not exceed the preset maximum value, so that the real-time property can be maintained, and real-time traffic can be reduced. While maintaining the characteristics, the voice packet transmission interval and the packet size can be changed at the same time so as not to impair the real-time property.

  Furthermore, according to the present invention, the maximum allowable delay value, which is one of the data base parameters, can be set in advance according to the partner communication terminal. Therefore, when the partner is one-way traffic, delay tolerance is allowed. The maximum value can be set to a large value. In this case, the music on hold and the guidance message output from the voice server do not care about the delay due to one-way traffic, so that the packetization interval is set to two-way communication. It can be made longer than that, and further efficiency can be expected.

  Furthermore, according to the present invention, since a parameter can be set for each destination in the packetization interval database, real-time characteristics are essential for a one-way traffic such as a guidance server and a partner that does not pursue real-time characteristics so much. The other party can be identified and operated in advance.

  FIG. 1 is a block diagram showing an example of a network to which a voice packet transfer method according to the present invention can be applied. In the figure, a network X constituted by an IP network typified by the Internet / intranet has WAN or LAN lines x1, x2, x3 connected to VoIP (Voice over Internet Protocol) terminals A, B, C for implementing the present invention. And a call control / management server S1 that performs relay processing and centralized management of call control signals transmitted from VoIP terminals A, B, and C; Guidance is sent out and connected to a voice server S2 that implements the present invention.

  FIG. 2 shows a block diagram of an embodiment of the VoIP terminals A, B and C of FIG. 2, the VoIP terminal 100 (A, B or C in FIG. 1) includes a call control / management unit 101, a voice packet processing unit 102, a jitter processing unit 103, a statistical processing unit 104, a packetization interval database 105, a layer 2 And a layer 3 protocol processing unit 106, a voice interface 107, a WAN and a LAN interface 108.

  The call control / management unit 101 executes call processing and call management, such as VoIP call connection processing and state management of the same call, based on a call signal from the voice interface 107 and an incoming call signal from the IP network side. The voice packet processing unit 102 performs a conversion process of encoding (decoding) or decoding (decoding) a 64 kPCM signal transmitted / received to / from the voice interface 107 into a voice packet.

  The jitter processing unit 103 absorbs fluctuation (jitter) of the voice packet received from the IP network, and delivers the voice packet to the voice packet processing unit 102 at a constant period. Further, the jitter processing unit 103 performs interpolation processing when the voice packet does not arrive, and further notifies the statistical processing unit 104 of statistical information such as the degree of fluctuation of the voice packet and the number of interpolations. The statistical processing unit 104 calculates statistical information received from the jitter processing unit 103, determines an optimal packetization interval, and creates a database for each destination. This database is the packetization interval database 105.

  The layer 2 and layer 3 protocol processing unit 106 performs layer 3 protocol processing such as TCP / IP (Transmission Control Protocol / Internet Protocol) and UDP / IP (User Datagram Protocol / Internet Protocol), Ethernet (registered trademark), frame Layer 2 protocol processing such as relay (Frame Relay) and PPP (Point-to-Point Protocol) is performed. The audio interface 107 indicates a physical interface on the audio side. The WAN and LAN interface 108 indicates a physical interface for connecting to the VoIP network.

  Next, the VoIP terminals A, B, and C that implement the present invention, the call control / management server S1, the network connection configuration, and the role of each configuration will be described with reference to FIGS. The WAN and LAN interface 108 in FIG. 2 of each of the VoIP terminals A, B, and C is connected to the IP network X via the WAN or LAN lines x1, x2, and x3.

  The VoIP terminals A, B, and C are normally connected to the IP network X and transmit signaling and voice packets for voice call connection. Signaling is relayed through the call control / management server S1 and is call-connected between the VoIP terminals A, B, and C. Voice packets are transmitted and received between VoIP terminals A, B, and C. The voice server S2 is used when the connection partner hears a hold sound or voice guidance during a call connection or during a call, and is activated by a request from the VoIP terminals A, B, and C.

  Next, the flow between functional blocks regarding the voice packetization interval determination method according to the network quality situation of the present invention will be described with reference to the block diagram of FIG.

  (1) When a calling side VoIP terminal receives a call request from the voice interface 107, the call control / management unit 101 refers to the packetization interval database 105 to determine the voice packetization interval that the terminal itself wants to receive. Then, the interval is set in the VoIP signaling, and a call connection request on VoIP is transmitted through the layer 2 and layer 3 protocol processing unit 106 and the WAN / LAN interface 108.

  In addition, when the call-side request is received by the call control / management unit 101 via the WAN / LAN interface 108 and the layer 2 and layer 3 protocol processing unit 106, the calling-side VoIP terminal receives the received call connection request response. The voice packetization interval from the signal to the called VoIP terminal is extracted and notified to the voice packet processing unit 102. When the call state transitions to the call phase or the call phase, the call control / management unit 101 notifies the voice packet processing unit 102 and instructs transmission / reception of voice packets.

  The voice packet is transmitted by generating a voice packet in the voice packet processing unit 102 based on the packetization interval notified to the voice packet processing unit 102 and passing through the layer 2 and layer 3 protocol processing unit 106 and the WAN / LAN interface 108. The voice packet is transmitted.

  The voice packet is received by the WAN / LAN interface 108, the layer 2 and layer 3 protocol processing unit 106, and the jitter processing unit 103. The jitter processing unit 103 determines the amount of fluctuation from a preset initial value and a predetermined number of packet reception intervals, and determines a waiting time. The amount of fluctuation (n) and the number of interpolations (m) for this waiting are periodically notified to the statistical processing unit 104 together with the transmission source IP address and the number of transmitted / received packets.

  The statistical processing unit 104 determines a voice packetization interval based on the obtained fluctuation amount (n), the number of interpolations (m), and a preset delay tolerance (p), and the value of the current packetization interval database 105 is determined. Compare with As a result of the comparison, when the voice packetization interval obtained by the statistical processing is different from the value of the packetization interval database 105, the value is reset in the packetization interval database 105, and the call control / The management unit 101 is notified, and the call connection request change signal is used to instruct the change of the packetization interval of the other party.

  (2) When receiving a call connection request from the WAN / LAN interface 108, the called VoIP terminal refers to the packetization interval database 105 in the call control / management unit 101, and converts the voice packet to be received by itself. Referring to the interval, the voice packetization interval is extracted to set the interval in VoIP signaling, and a call connection request response on VoIP is transmitted through the layer 2 and layer 3 protocol processing unit 106 and the WAN / LAN interface 108. When the call state transitions to the call phase or the call phase, the call control / management unit 101 notifies the voice packet processing unit 102 and instructs transmission / reception of voice packets. The called VoIP terminal refers to the transmission interval in the packetization interval database 105 and executes voice packet transmission / reception processing. The subsequent voice packet transmission / reception processing and packetization interval changing algorithm are the same as in (1) above.

  In the packetization interval database 105, the following parameters are set for each destination.

Transmission interval: SI Monitoring cycle: T
Reception interval: RI Interpolation rate 1: P1
Allowable delay: D Interpolation rate 2: P2
Next, each call connection processing of the originating VoIP terminal and the terminating VoIP terminal in the present embodiment will be described with reference to the flowchart of FIG. When receiving a call request from the voice interface 107 (step ST1), the calling VoIP terminal refers to the reception interval value set in the packetization interval database 105 by the call control / management unit 101 (step ST2). ) Determine the voice packetization interval that the terminal wants to receive, set the interval in VoIP signaling (step ST3), and call the VoIP call through the layer 2 and layer 3 protocol processing unit 106 and the WAN / LAN interface 108. A connection request is sent out (step ST4).

  When the called VoIP terminal receives a call connection request from the WAN / LAN interface 108 (step ST5), the call control / management unit 101 extracts the voice packet transmission interval from the call connection request signal to the calling VoIP terminal. Then, the packetization interval database 105 is updated (step ST6). Subsequently, the called VoIP terminal refers to the reception interval in the packetization interval database 105 (step ST7), extracts the value, sets it as a call connection request response signal, and transmits the call connection request response signal ( Steps ST8 and ST9). The called VoIP terminal refers to the transmission interval of the packetization interval database 105 (step ST10), and executes voice packet transmission processing and reception preparation (step ST11).

  When the calling side VoIP terminal receives the call connection request response by the call control / management unit 101 through the WAN / LAN interface 108 and the layer 2 and layer 3 protocol processing unit 106 (step ST12), the received call connection request response signal is received. After extracting the voice packetization interval from the VoIP terminal to the called VoIP terminal (step ST13) and updating the packetization interval database 105 (step ST14), the call control is performed when the call state transitions to the call phase or the call phase. The management unit 101 notifies the voice packet processing unit 102, and performs voice packet reception processing and transmission preparation (step ST15). When the process of step ST11 or ST15 is completed, the process proceeds to step ST21 of FIG. 4 or step ST31 of FIG.

  Next, an embodiment of the voice packet transmission process of the present invention will be described with reference to the flowchart of FIG. When the call connection phase ends (A) and the VoIP terminal enters the call phase or the call phase, the VoIP terminal refers to the transmission interval set in the packetization interval database 105 (step ST21), and performs voice according to the transmission interval. A packet is generated and transmission is started (step ST22).

  After sending the voice packet, the VoIP terminal monitors whether or not a call connection request change signal is received (step ST23). The VoIP terminal does not change the packetization interval until the end of the call unless it receives a call connection request change signal. However, when the call connection request change signal is received, the packetization interval is extracted from the received call connection request change signal (step ST24), and the packetization interval database 105 is updated to the extracted packetization interval (step ST25). Thereafter, until the end of call or call connection request change signal is received, generation and transmission of voice packets are continued at the updated packetization interval (step ST26).

  Next, an embodiment of voice packet reception processing of the present invention will be described with reference to the flowchart of FIG. When receiving the voice packet from the counterpart VoIP terminal by the jitter processing unit 103 shown in FIG. 2 (step ST31), the VoIP terminal counts the number of interpolations of the voice packet. Packet interpolation occurs when packets cannot be absorbed by the jitter buffer due to packet discard or delay / fluctuation in the network. The jitter processing unit 103 observes the presence / absence of interpolation for each preset monitoring period (steps ST32 and ST33). When interpolation occurs, the number of interpolations (m) is increased (step ST34).

  As described above, the jitter processing unit 103 determines a waiting time by determining a fluctuation amount from a preset initial value and a predetermined number of packet reception intervals. A fluctuation amount (n) for the waiting and the number of interpolations (m) are periodically notified to the above-described statistical processing unit 104 together with the transmission source IP address and the number of transmission / reception packets.

  The statistical processing unit 104 calculates an interpolation rate based on the obtained fluctuation amount (n), the number of interpolations (m), and a preset delay allowable amount (p), and determines the value of the calculated interpolation rate ( Step ST35). The interpolation rate is a value obtained by dividing the number of interpolations (m) by the total number of received packets. When the interpolation rate is 0%, that is, when there is no interpolation within the monitoring period, it is determined whether the current packetization interval is an allowable maximum value (step ST36). For example, it is increased by 10 ms (step ST37).

  When the interpolation rate is not less than the first threshold value P1 and less than the second threshold value P2 (> P1), it is determined whether or not the current packetization interval is the minimum value (step ST38). Then, the packetization interval is reduced by, for example, 10 ms (step ST39). When the interpolation rate is as large as the second threshold P2 or more, the packetization interval is unconditionally changed to the minimum value (step ST40). A high interpolation rate means that there is a problem with the line quality, so when there is a problem with the line quality, the packetization interval is narrowed and the transfer is performed with a small packet size, so the effects and obstacles of interpolation are encountered. Probability can be reduced.

  Further, when there is no interpolation within the monitoring period and the packetization interval is the maximum allowable value, or when the interpolation rate is less than P1, the current packetization interval is maintained (step ST41). . Also, the current packetization interval is maintained even when the interpolation rate is greater than or equal to P1 and less than P2 and the packetization interval is the minimum value (step ST42).

  When the packetization interval is changed in the above step ST37, ST39 or ST40, the reception interval stored in the packetization interval database 105 is updated to the changed value (step ST43), and a call connection request change signal is transmitted. Then, a request is made to change the voice packet transmission interval of the partner VoIP terminal (step ST44). When the process of step ST44 is completed, the process proceeds to step ST23 of FIG.

  Thus, in this embodiment, since the packetization interval is dynamically changed and controlled according to the interpolation rate, the load on the communication device is maximized within a range that does not hinder voice calls in a normal network state. When the network quality is reduced, voice calls can be guaranteed with the highest priority.

  Note that a fixed amount of audio data is periodically generated. For example, in the case of an 8k compressed voice packet, 1000 bytes of data are generated per second. In order to ensure real-time performance, this audio data must be transferred periodically (20 ms interval for transferring every 20 bytes, and 40 ms interval for transferring every 40 bytes. Need to be transferred in). In FIG. 5, the packetization interval is dynamically changed and controlled according to the interpolation rate. However, since the voice data is periodically generated in a fixed amount as described above, the packetization interval size is dynamically changed due to its real-time characteristics. The change is substantially the same as a dynamic change of the packet size of the voice packet.

  Further, in FIG. 5, the allowable maximum value of the packetization interval is referred to in step ST36 so that the packetization interval does not exceed the allowable maximum value, so the real-time property is not impaired. As described above, according to the present embodiment, it is possible to simultaneously change the voice packet transmission interval and the packet size while maintaining the real-time traffic characteristics so as not to impair the real-time property.

  Furthermore, in the present embodiment, as described above, since the allowable delay D is set for each destination in the packetization interval database 105, the above-described allowable delay D (delay of delay) is set when the destination is one-way traffic. The maximum allowable value can be set to a large value. In this case, music on hold and guidance messages output from the voice server do not care about delays due to one-way traffic, which makes the packetization interval bidirectional. It can be longer than a call.

  Furthermore, in the present embodiment, since the packetization interval database 105 can set parameters for each destination, real-time characteristics are essential for a one-way traffic such as a guidance server and a partner that does not pursue real-time characteristics so much. The other party can be identified and operated in advance.

It is a block diagram of an example of the network which can apply the voice packet transfer method of this invention. It is a block diagram of one Embodiment of the VoIP terminal in FIG. It is a flowchart for explaining each call connection process of the originating side VoIP terminal and the terminating side VoIP terminal in the present invention. It is a flowchart for description of one embodiment of a voice packet transmission process in the present invention. It is a flowchart for description of one embodiment of a voice packet reception process in the present invention.

Explanation of symbols

A, B, C, 100 VoIP terminal S1 Call control / management server S2 Voice server X network 101 Call control / management unit 102 Voice packet processing unit 103 Jitter processing unit 104 Statistical processing unit 105 Packetization interval database 106 Layer 2 and layer 3 Protocol processing unit 107 Voice interface 108 WAN and LAN interface

Claims (6)

Multiple terminals with voice packet encoding, signaling conversion and switching functions send call control signal relay processing and centralized management, call control / management server, and on-hold tone and guidance at the request of the terminal A voice packet transfer method for a system that is bi-directionally connected to a voice server via a communication network,
When the terminal receives a call connection request from the counterpart terminal, a first step of periodically transmitting voice packets to the counterpart terminal at a transmission packetization interval obtained by referring to its own database;
A second step of calculating an interpolation rate of voice packets received via the communication network;
A third step of changing a packetization interval of voice packets to be transmitted according to the interpolation rate;
A fourth step of changing a packetization interval for transmission of the own database according to the changed packetization interval;
A fifth step of transmitting a call connection change request at the changed packetization interval to the counterpart terminal;
And a sixth step of extracting the changed packetization interval from the received call connection change request and updating the packetization interval for transmission in its own database to the extracted packetization interval. Voice packet transfer method.   In the third step, when the interpolation rate is equal to or greater than a first predetermined value, the packetization interval is changed to a preset minimum value, and the first interpolation value is less than the first predetermined value. When the value is equal to or larger than a second predetermined value smaller than a predetermined value, the packetization interval is decreased by a certain value as the interpolation rate value decreases, and the interpolation rate is less than the second predetermined value or the packetization interval is minimized. 2. The voice packet transfer method according to claim 1, wherein when the value is a value, the current value is maintained without changing the packetization interval.   In the third step, when the interpolation rate is 0%, if the packetization interval is not the preset maximum value, the packetization interval is increased by a certain value, and when the packetization interval is the maximum value, 3. The voice packet transfer method according to claim 2, wherein the current value is maintained.   In the first step, voice packets are periodically transmitted from the own database in which various parameters are set for each partner terminal at a transmission packetization interval determined by the partner terminal of the call connection request source. 2. The voice packet transfer method according to claim 1, wherein the voice packet is transmitted to the destination terminal. Signaling and voice packets for voice call connection are connected bi-directionally via a communication network to a call control / management server that performs call control signal relay processing and centralized management, and a voice server that sends music on hold and guidance. A terminal that transmits
A voice interface showing the physical interface on the voice side;
A call control / management unit that performs call processing and call management, such as call connection processing and call status management, based on a call signal from the voice interface and an incoming call signal from the communication network side;
A voice packet processing unit for performing a conversion process for encoding or decoding a PCM signal transmitted to and received from the voice interface to form a voice packet;
Absorbs fluctuations in voice packets received from the communication network, delivers voice packets to the voice packet processing unit at regular intervals, performs interpolation processing when the voice packets do not arrive, and performs the degree of fluctuation or interpolation of the voice packets A jitter processing unit that outputs statistical information such as numbers,
Calculating the statistical information received from the jitter processing unit, determining the optimal packetization interval, and storing the statistical processing unit in the packetization interval database for each partner;
A protocol processing unit for performing predetermined protocol processing;
A network interface connected to the communication network, and the statistical processing unit calculates an interpolation rate of the received voice packet, changes a packetization interval of the voice packet to be transmitted according to the interpolation rate, and transmits the packet. Updating the communication interval database, and sending a call connection request change signal to the counterpart terminal by the call control / management unit to instruct the destination terminal to change the packetization interval of the transmission voice packet. Terminal. In the packetization interval database, a transmission packetization interval, a reception packetization interval, a monitoring cycle, an allowable delay, and an interpolation rate threshold are set for each counterpart terminal, and the statistical processing unit is configured to monitor the monitoring cycle. The interpolation rate is calculated based on the number of interpolations within the range, and when the interpolation rate is equal to or greater than the interpolation rate threshold, the transmission packetization interval is changed to a preset minimum value, and the interpolation rate is calculated based on the interpolation rate. When the transmission packetization interval is less than the threshold value, a change is made to decrease the transmission packetization interval by a constant value, and when the interpolation rate is 0%, the transmission packetization interval is changed to a constant value within a predetermined maximum value or less. 6. The terminal according to claim 5, wherein the terminal updates the packetization interval database.

Images