KR20020085931A - Voice multicasting method using logical ring - Google Patents

Abstract

The multi-way voice communication method in the Internet of the logical ring method of the present invention is a multi-way voice communication method in the Internet by a logical ring method comprising a plurality of voice communication participants including a group manager, and rotates along the logical ring. A first step of delivering the mixed voice frames after mixing the voices of the respective voice communication participants, and once again according to the logic ring, each voice communication participant plays the voice except the own voice among the mixed voice frames It consists of a second step.

Description

VOICE MULTICASTING METHOD USING LOGICAL RING}

The present invention relates to a multi-party voice communication method through the Internet, and in particular, by performing a voice communication through the Internet using a logic ring method, the quality of communication voice is improved and the logic ring method can minimize the voice processing time of each participant. The present invention relates to a multi-way voice communication method on the Internet.

Since the creation of the ARPANET in the United States for the purpose of exchanging information between universities, it has evolved into an internet that enables the exchange of information globally by accessing numerous servers. The Internet was originally created for the purpose of exchanging academic information among scholars, but now it is in the limelight as a means of data communication with numerous computers connected around the world. With the development of computer technology and the development of network technology, the Internet has become an e-mail service, remote login (Telnet), file transfer protocol, information retrieval, usenet news, video conferencing. , A number of services such as voice communication services are being made.

In particular, voice communication services are being actively researched in terms of cost reduction since real-time services are possible according to the development of a high-speed network. A representative service of these voice communication services is Voice over Internet Protocol (VoIP) service. The VoIP service provides a telephone service provided by an existing analog telephone network through the Internet. The VoIP service modulates analog voice into digital data and transmits the data in real time.

As the first voice communication network, the telephone network that has been widely used up to now uses a circuit switched network that occupies a line between the two subscribers on the phone, making it difficult and efficient to use the line, as well as various computers such as computers. There was a problem that it was difficult to add a function. On the other hand, VoIP services using packet-switched networks have many advantages over circuit-switched networks because they do not guarantee the sound quality of existing telephone networks due to delay, jitter, packet loss, and narrow bandwidth, but they provide inexpensive and diverse services. . In addition, with the development of the Internet infrastructure, data compression technology, and computer performance, the technical limitations of the Internet network have been overcome to provide one-to-one and multi-party voice communication based on a high level of QoS.

Initially, VoIP service provided 1: 1 voice (Unicast) voice call, which is a conventional telephone service, and is gradually expanded to N: N (Multicast) multi-party voice communication.

There are various methods for implementing multi-party voice communication, including a method using a router that supports multicasting. By the way, many applications have been developed to support this, but since there are not many routers that support the multicasting function, there is a disadvantage that the generality is poor. In addition, there is a method for implementing multi-party voice communication without using a multicasting router. The multi-party voice communication can be largely divided into a centralized, decentralized, and mixed method. The centralized method has a separate independent server for multi-party voice communication. The server manages groups to synthesize and distribute voice data of each participant. However, this method has a burden of separately constructing an expensive server, and the voice data of each group participant is concentrated on the server, and as the number of users increases, the server load increases so that the voice quality cannot be uniformly guaranteed. have. The distributed method allows voice communication between participants in each group without an independent server, and the management is one of the participants. The hybrid method is a method performed by mixing the centralized method and the distributed method.

One of the existing methods of implementing the distributed method is that one participant sends a voice frame to all the other participants. This method has a disadvantage in that network traffic and the amount of voice data to be processed by each participant rapidly increase as the number of users increases, making it difficult to efficiently use bandwidth and guaranteeing voice quality.

SUMMARY OF THE INVENTION The present invention has been made in view of the above, and a multi-way voice communication method in the Internet of the logical ring method that can improve the call quality even if the number of voice communication participants increases by minimizing network traffic and voice data to be processed by each participant. It aims to provide.

Another object of the present invention is to provide a multi-way voice communication method in the Internet of the logical ring method that can minimize packet delay by configuring the shortest distance logical ring when a new voice communication participant joins the logical ring.

In order to achieve the above object, the voice communication method according to the present invention is a multi-way voice communication method in the Internet by a logical ring method consisting of a plurality of voice communication participants including a group manager, the group manager has his voice Delivering the recorded voice frame to a next voice communication participant; Mixing each voice participant with a voice frame transmitted from a previous voice communication participant and delivering the same to a next voice communication participant; A voice communication participant located at the last node of the logical ring, mixing his voice with the voice frame received from the previous voice communication participant, and transmitting the voice to the group manager to play the voice frame simultaneously; Playing voice frames excluding their own voice among voice frames delivered from previous voice communication participants by a voice communication participant including a group manager; And playing a voice frame excluding itself in a voice frame from which a voice communication participant located before the last node of the logical ring is delivered.

In addition, when a new voice communication participant joins, the joining is performed by the group manager allowing the group participation from the voice communication group participant, and the participant sends a 'PING' message to each group participant to make a round trip time. Detecting a voice communication participant at the closest distance by inserting the participant between the detected closest voice participant and the voice communication participant at the next distance. In this case, the logical ring can always form the shortest distance.

Figure 1 (a) is a diagram briefly showing the structure of a voice communication system according to the present invention.

Figure 1 (b) is a diagram showing the transmission of the voice packet in Figure 1 (a).

2 is a simplified diagram showing the structure of a voice communication apparatus according to the present invention;

3 is a block diagram illustrating a structure of a voice processing unit of FIG. 3.

4 is a diagram illustrating an operation mechanism of the voice communication system according to the present invention;

FIG. 5 is a diagram illustrating recording time and playback time of each participant in the present invention on a time axis.

6 is a flowchart illustrating a multi-party voice communication method in the Internet using a logic ring method according to the present invention.

7 is a view showing a data structure in the form of a linked list included in each participant in the present invention.

8 is a view showing a message exchanged between a group manager and other group participants in the logical ring of the present invention.

9 is a diagram illustrating a message flow diagram for a process in which a specific participant leaves a group in the logical ring of the present invention.

FIG. 10 is a view illustrating a recovery process of a logical ring when a logical participant is damaged due to abnormal termination of a specific participant in the present invention. FIG.

11 is a message flow diagram for logical ring recovery when the logical ring is corrupted in the present invention.

Explanation of symbols on the main parts of the drawings

1: Internet 3: Group Manager

5,7,9: Voice communication participant 11: Microphone

13 speaker 20 voice communication device

22: group management unit 24: voice processing unit

23: recording unit 25: playback unit

27: receiver 29: transmitter

32: mixing section 34: echo processing section

In general, multi-way voice communication using a logical ring connects call participants to a virtual ring for each group so that the voices spoken by each participant are rotated in sequence along the logical ring to other participants. In addition, since voice frames are played back at regular intervals, call quality is guaranteed even when multiple people talk at the same time, and the unique voice mixing technique is applied so that network traffic does not increase even if the number of participants increases. By not only minimizing the speech processing time, but also minimizing the packet delay with the logical ring configuration mechanism of the shortest distance.

1 is a view showing a multi-party voice communication system using a logical ring in the Internet according to the present invention, Figure 1 (a) is a view showing the operation during the first rotation of the voice packet and Figure 1 (b) of the voice packet It is a figure which shows the operation | movement at the time of 2nd rotation. As shown in the figure, the group manager 3 and the plurality of voice communication participants 3, 5, 7 are connected to each other via the Internet 1, respectively. In the drawing, reference numeral M denotes a voice packet of a group manager, and A, B, and C denote voice packets of each participant. Such a system is a fully distributed system for exchanging voices among participants of each group without a separate independent server. As shown in the drawing, a voice packet recorded by a group manager rotates twice along a logical ring.

As shown in Fig. 1 (a) and 1 (b), the voice of each participant is collected in the first rotation of the voice packet, and the voice collected in the first rotation is reproduced in the second rotation.

The voice data spoken by the participants is mixed with the voices of the other participants through the logic ring (actually, the voice represented by the binary code is added together), and the number of bits of the voice data mixed with two or more voices is the original number. The same size as the voice data, the network traffic does not increase even if the number of participants increases. That is, as shown in Fig. 1 (a), the PC of the group manager 3 converts the input voice signal into digital data, stores the converted voice packet M, and then stores the converted voice packet M to the next participant 5. send. The participant 5 mixes his voice packet A with the voice packet M of the group manager 3 that is transmitted, and then mixes the mixed voice packet M + A to the next participant 7. Send again. The participant 7 also mixes its own voice packet B to the next participant 9 by mixing the voice packet M + A, which is also transmitted, and the participant 9 also sends its own voice packet C. After mixing, the mixed voice packet (M + A + B + C) is transmitted to the group manager 3.

On the other hand, as shown in Figure 1 (b), when the sum of the voice packets (M + A + B + C) is added to the group manager (3), the group manager (3) is his voice packet (M) in duplicate After restoring and outputting the remaining voice packets A + B + C except for), the voice packets M + A + B + C are again transmitted to the participant 5. The participant 5 also restores and outputs the voice packet M + B + C except for the voice A of the voice packet M + A + B + C and then returns to the next participant 7. The subsequent participants 7 and 9 also restore the voice packets except their voices B and C.

In addition, in the present invention, since the above-mentioned speech packet is processed by the mute processing method, the speech processing time can be minimized by omitting the speech processing process.

The PC is equipped with software for multilateral voice communication. The multi-party voice communication software is divided into group manager software and general participant software, which are responsible for group management according to their functions. The separation of roles is determined at the time of group creation. In addition, the software installed on each participant's PC has a group participation, withdrawal, and recovery functions related to group management, and the group manager software performs the overall group management function.

The voice communication device 20 as shown in FIG. 2 is installed in the terminal PC of the group manager 3 and the voice communication participants 5, 7, and 9. As shown in the figure, the voice communication device 20 converts the voice signals spoken by the participants in the voice communication from the input device 11 such as a microphone mounted in the terminal, converts the voice signals into digital signals, and mixes them with the received voice. The mixed voice data is transmitted to the Internet 1 along a logical ring path, and the voice data is removed from the received voice data and reproduced by the output device 13 such as a speaker. And a sound processor 24.

The group manager 22 is a means for managing the logical ring, and performs management functions such as creation, participation, and withdrawal of the group. The group manager 22 configures each participant in a virtual ring while performing group participation and withdrawal functions, and in particular, allows a logical ring of the shortest distance to be configured during group participation. In addition, the group manager 22 performs a function of restoring the logical ring to a group composed of the remaining participants except the terminated participant when the participant's PC is abnormally terminated and the logical ring is damaged.

The voice processing unit 24 is responsible for recording, mixing, echo processing, playback, reception, transmission, and the like. As shown in FIG. 3, the mixing unit 32, the echo processing unit 34, and the recording unit 23. ), A playback section 25, a reception section 27, and a transmission section 29.

The mixing unit 32 performs a function of mixing its own voice data and voice data of other participants, and the echo processing unit 34 performs a task of removing its own voice from the mixed data before playing the received voice. do.

In addition, the recording unit 23 receives analog voice through the input device 2 such as a microphone and converts it into digital data, and the reproduction unit 25 delivers the echo-processed digital data by the echo processing unit 34. And outputs voice through an output device 13 such as a speaker. The receiver 27 receives the voice data coming in through the ring path, and the transmitter 29 transmits the mixed voice data to the designated group participant.

4 is a view showing an operation mechanism of the speech processing system of the present invention. In the figure, group participants are assumed to be N, and (N, N ') represents the count field value at the first rotation and the count field value at the second rotation, respectively. As shown in the figure, when there are N group participants, the voice processing apparatus 24 receives voice data of another participant through the receiving unit 27 therein and examines the count field of the voice data. The count field value starts with the voice generated by the group manager and increases by 1 each time the participant passes through the ring until the value reaches 2N-2. The speech processing device 24 processes the received speech packet differently according to the count field value of the packet. The operation of each terminal node according to the count field value is shown in Table 1 below.

Count field value of received packet Operation at each terminal node Mixing loop One Deliver recording data 2 to (N-1) Forward by mixing received data and recorded data N Play forwarding received data by mixing received data and recorded data Loop 1 to (2N-2) Receives the received data as it is, removes and plays its own voice from the received data 2N-1 Remove and play your own voice from received data

As shown in the figure, in the first rotation, when the count field value is 1, the voice is recorded and transmitted to the next participant, and received at each terminal node until the count field value is 2 to N-1. Mix your own recorded voice into the voice frame and deliver it to the next participant. In addition, if the count field value is N, the last value is N, the recorded voice is mixed with the received voice frame and transmitted to the next participant, and at the same time, the received voice frame without the added voice is played.

In the second rotation, when the count field value is 1 to (2N-2), the received voice frame is transmitted to the next participant and the voice frame is removed from the received frame and played. In the last field 2N-1 of the second rotation, only the own voice is removed from the received voice frame without transmission of the voice frame and played.

In general, in the logical ring, when there are N participants, voice frames of each node are generally stored from the first rotation to the Nth node, and the stored voice frames are transmitted to the first node, and then N nodes from the first node in the second rotation. Play the voice frame from which the voice has been removed up to the first node. However, in the present invention, in the present invention, in the multi-party voice communication group having N participants, the start of playback of the mixed voice frames of all participants is performed at the Nth node, not the first node, and the end of the final playback. Is performed at N-1 node instead of N node, which reduces the ring rotation time.

FIG. 5 is a diagram showing the recording time and playback time of each participant when processing the voice on the time axis when the above method is applied, and shows that each participant reproduces a voice frame at a predetermined time interval.

In addition, the speech processing apparatus of the present invention applies the silent processing technique to perform the speech processing more efficiently. That is, when mixing and transmitting its own recording voice to the voice frame received in the first rotation of the voice packet, it is first determined whether the recorded voice is silent. In this case, when the sound is muted, the received voice frame is transmitted to the next participant without mixing. When the audio frame is removed from the voice frame received in the second rotation, the voice frame is muted in the first rotation. If it is determined that the mixing process is omitted, the process of removing the own voice is skipped and reproduced.

6 is a flowchart showing a multi-party voice communication method in the Internet using the above-described logic ring method of the present invention. Referring to the voice communication method of the present invention based on the drawings as follows. Here, it is assumed that the total number of participants in the multi-party voice communication is N people.

First, as shown in the figure, when a voice packet is received, the received voice packet is taken (S101), and the count field value of the received frame is increased by one (102). Subsequently, the count field value of the received voice frame is examined (S103). If the value is 1 to N, the voice frame is retrieved from the recording storage queue after recognizing that it is the first rotation of voice mixing (S104).

Then, it is determined whether the recorded voice frame is silent (S105), and if not, the received voice frame and the recorded voice frame are added (mixed) and stored in the transmission data queue for transmission to the next participant (S106). On the contrary, when the recorded voice frame is muted, the received voice frame is stored in the transmission data queue without mixing (S107).

Thereafter, it is checked whether the count field value is N (the last step of the first rotation of the voice frame) (S108). That is, it is determined whether the voices of all participants are mixed. If the count field value is N, the voices of all participants are mixed, and thus, the reception frame in which the voice of the participant (Nth) is not mixed is immediately stored in the play queue, and if the count field value is not N, the skip field is omitted. (S109). Subsequently, the recording data is stored in its previous data queue to be echoed in the next rotation of the voice frame (S110).

On the other hand, when the count field value is 1 to 2N-1 in step S103, it corresponds to the second rotation in which audio reproduction is performed. First, it is checked whether the count field value is 2N-1, and if the count field value is not 2N-1, the received frame is stored in the transmission data queue for transmission to the next participant (S112), and if the count field value is 2N-1, This is the last step of the second rotation, so omit it. Subsequently, the voice recorded by the user in the first rotation is taken from the previous recording data queue for echo processing before the received voice frame is reproduced (S113). Then, it is checked whether the recorded voice is silent data (S114). If it is not the silent data, the previous recorded voice is removed from the received data and stored in the reproduction data queue (S115). The data is stored in the data queue (S116).

The group manager 22 basically creates a group and handles group participation and withdrawal of the participants. It is responsible for establishing and updating a logical ring type connection between group participants, and recovering a logical ring due to an abnormal situation. Although not shown in the figure, when the group management unit 22 is divided into functions, it is divided into a group generation unit, a group participation unit, a group withdrawal unit, and a logical ring recovery unit. The configuration of the group management unit can be easily configured on the basis of the functions described below by a person in the technical field to which the present invention belongs, and the drawings will be omitted and only the functions will be mainly described.

The group generator performs a function of creating a specific group. At this time, the group creator becomes a group manager and waits for group participation of other users, and performs information management of the participants or the entire group. Participants' information is managed in a data structure in the form of a linked list as shown in FIG. Information about one participant is represented in one node of the linked list, as shown in the figure, the participant's information includes the participant's name, the participant's IP address, the participant's next participant's IP address, and the participant's It consists of the previous participant's IP address, 'voice socket' and 'control socket'.

'Participant name' is information for informing the name of the participant to the user, and 'participant IP address' is the IP address of the computer used by the participant. The next participant IP address of the participant and the previous participant IP address of the participant are information representing the IP address of the next participant and the previous participant of the participant in the logical ring structure. It is used to refer to the information of the next and previous participants of the terminated participant. In addition, the 'voice transmission and reception socket' is a socket for voice communication with the participant, 'control message transmission and reception socket' is a socket used to exchange control messages with the participant.

Group Participation performs tasks required for general users to join a specific group. That is, it sends a participation request message (JOIN) to the group manager of the group to join, receives the response (JOIN_ACK) and performs the function of joining the group while maintaining the logical ring structure of the participating group. FIG. 8 is a diagram illustrating a message exchanged between a group manager and other group participants (participant A and participant B) by a participant C in order to join a group in a situation where a logical ring is formed as a group manager → participant A → participant B. FIG.

As shown in FIG. 8, when Participant C carries his information on the message 'JOIN' and transmits it to the Group Manager, the Group Manager adds the Participant C's information to the group participant information list and joins the message 'JOIN_ACK'. 'Send the message back to Participant C. At this time, the 'JOIN_ACK' message includes the IP addresses of all participants in the current group.

Participant C receiving the 'JOIN_ACK' message sends a 'PING' message to all participants in the group using the IP addresses of all participants in the group to measure the round trip time (RTT). Assuming that Participant A has the shortest RTT, Participant C sends Participant A a message of 'NEXT_MEMBER' to inform Participant A that it is the next Participant of Participant A, and a 'NEXT_MEMBER_ACK' response message from Participant A. Receive. In addition, Participant C sends a 'PREV_MEMBER' message to Participant B, which is the original next participant of Participant A, indicating that it is a previous participant of Participant B, and receives a 'PREV_MEMBER_ACK' message. By the above process, the participation process of the participant C is terminated, and the participant C notifies the group manager by sending a 'UPDATE_RING' message that the update of the logical ring is completed. Meanwhile, the group manager receives the 'UPDATE_RING' message, updates the group information, and responds with a 'UPDATE_RING_ACK' message, thereby ending the group participation process. Eventually, a logical ring is formed in the order of the group manager-participant A-participant B.

The leaving group handles the work required for one participant to leave the group. FIG. 9 is a diagram illustrating a message flow diagram for a process in which participant C leaves a group in a group configured with logical rings (group manager-participant A-participant C-participant B-group manager).

As shown in the figure, when Participant C sends an 'EXIT' message to the Group Manager that he or she wants to leave the Group, the Group Manager processes the task of leaving Participant C from the group, and then sends a 'EXIT_ACK' message to the Participant C. Signals that the group leave operation has ended normally.

In order to withdraw Participant C from the group, the Group Manager first sends Participant C the information of Participant C in the message 'EXITED_MEMBER' to inform Participant C to leave the group. Participant B deletes Participant C from his / her participant information data structure and responds to the Group Manager with an 'EXITED_MEMBER_ACK' message. The group manager performs the same process to Participant A to inform Participant A that Participant C will leave the group. That is, the other participants (participant A and Participant B) of the group except Participant C are sent with the information of Participant C in the message 'EXITED_MEMBER' to inform Participant C that they will leave the group.

Thereafter, the group manager sends the information of the participant A (the previous participant of the participant C) to the participant B (next participant of the participant C) in a message 'PREV_MEMBER'. As the 'PREV-MEMBER' message is received, Participant B modifies its previous Participant to Participant A and responds to the Group Manager with a 'PREV_MEMBER_ACK' message. In addition, the group manager sends the information of the participant B (the next participant of the participant C) to the participant A (the previous participant of the participant C) in a 'NEXT_MEMBER' message. As the 'NEXT_MEMBER' message is received, Participant A modifies its next Participant to Participant B, and responds to the Group Manager with a 'NEXT_MEMBER_ACK' message. By the above process, the group manager configures logical ringing as group manager-participant A-participant B to perform participant C withdrawal from the group.

If any participant in the group is abnormally terminated or abnormally operated due to a network problem, and the logical ring structure is damaged, the group recovery unit forcibly removes the participant and restores the logical ring. FIG. 10 is a diagram illustrating a logic ring recovery process when a participant C is abnormally terminated and a logical ring is damaged. FIG. 11 is a diagram illustrating a message flow diagram for logical ring recovery.

As shown in FIG. 10, when participant C is abnormally terminated, voice data is not delivered to participant B, the next participant of participant C. Accordingly, Participant B sends an 'IS_ALIVE' message to Participant C, and if there is no response for a certain time, Participant C recognizes that Participant C is abnormally terminated and performs the operation to remove Participant C from the logical ring. Participant B informs the group manager that Participant C is abnormally terminated by sending the Participant C's IP address to the 'PREV_MEMBER_DOWN' message. The group manager who receives the message performs the task of forcibly disengaging Participant C. First, the Participant C is informed that the Participant C is abnormally terminated by sending the Participant C information to the 'DOWNED_MEMBER' message.

Subsequently, the group manager obtains the information of the previous and next participants of participant C by referring to his group list, and then sends the following messages to the two participants (the previous participant of participant C and the next participant of participant C) and the message 'PREX_MEMBER' respectively. Send a message to update the logical ring information. The ring recovery operation is then terminated by removing participant C from the group list.

In the present invention, after the logical ring is formed, the last participant of the logical ring records the voice of another participant at the same time as recording the voice of the other participant and plays the voice of other participants. Therefore, in the second rotation that reproduces the voices of the participants other than their own voices, only the N-1 th participants (when there are N participants) need not be rotated until the entire participants. In addition, in the present invention, when the logical ring is damaged due to abnormality of the network or abnormal termination of the participant, the logical ring can be smoothly returned by forcibly withdrawing the participant.

As described above, the present invention not only improves voice quality during multi-party voice communication on the Internet, but also minimizes network traffic and voice processing time of each participant and configures the shortest distance logical ring to minimize packet delay.

Claims (9)

In the multi-way voice communication method on the Internet by a logical ring method consisting of a plurality of voice communication participants including a group manager, Transmitting, by the group manager, a voice frame in which his voice is recorded to a next voice communication participant; Mixing each voice participant with a voice frame transmitted from the previous voice communication participant and delivering the same to the next voice communication participant; A voice communication participant located at the last node of the logical ring, mixing his voice with the voice frame received from the previous voice communication participant, and transmitting the voice to the group manager to play the voice frame simultaneously; Playing voice frames excluding their own voice among voice frames delivered from previous voice communication participants by a voice communication participant including a group manager; And And reproducing a voice frame excluding itself in a voice frame from which a voice communication participant located before the last node of the logical ring is delivered. The method of claim 1, wherein the mixed voice data is processed by adding the data of each voice communication participant to be delivered in the same size as one voice data. The method of claim 1, wherein the group manager generates a logical ring and performs management of information or a group of voice communication participants. The voice communication participant information includes a participant's name, a participant's IP address, a participant's next participant's IP address, a participant's previous participant's IP address, a voice transmission and reception socket, and a control message transmission and reception socket. The voice communication participant information is managed in the form of a linked list. The method of claim 1, Allowing a group manager to join a group from a voice communication group participant; A participant sending a 'PING' message to each group participant to measure a round trip time (RTT) to detect a voice communication participant at the closest distance; And And inserting the participant between the detected closest distance voice communication participant and the next distance voice communication participant to join the participant as a new voice communication participant. The method of claim 1, Transmitting a signal to the previous voice communication participant when a voice frame is not transmitted from the previous voice communication participant; Transmitting a message indicating that the previous voice communication participant was abnormally terminated to a group manager when there is no response from the previous voice communication participant for a predetermined time; And And if the group manager receives the message indicating the abnormal termination, forcibly withdrawing the corresponding voice communication participant from the logical ring to restore the logical ring. In the multi-way voice communication method on the Internet by a logical ring method consisting of a plurality of voice communication participants including a group manager, A first step of mixing the voices of the participants in the voice communication by one rotation along the logic ring and delivering the mixed voice frames; And And a second step in which each voice communication participant reproduces a voice other than its own voice among the mixed voice frames by rotating once again along the logic ring. 8. The method of claim 7, wherein the first step comprises the step of recording the voice of the voice communication participant located at the last node of the logical ring to the group manager and playing the received voice frame at the same time. Method comprising a. 8. The method according to claim 7, wherein the reproduction of the voice frame in the second step is terminated at the voice communication participant located before the last node of the logical ring.