KR100545655B1 - Internet phone system for selecting voice compression type by internet protocol network peculiarity and thereof method - Google Patents

Abstract

The present invention provides an Internet phone system for selecting a voice compression method according to an improved IP network characteristic to automatically select an optimal voice compression method and a packet processing priority in a situation in which packet delay time changes along a path such as a public IP network. And a method of selecting the same.

The present invention is to construct a ping client and a ping server to respond to the ping function used in Unix / Linux, etc. in the call setup module of the configuration of a typical Internet phone system when the sender enters the receiver's telephone number Before connecting the call, the IP packet exchange between the ping client and the ping server through the IP network records and judges the number, delay time, packet loss rate, etc. via the IP address of the IP network. By determining the priority of the method, it is possible to automatically select the optimal voice compression method and packet processing method according to the given network condition to ensure the best call quality and to always ensure the optimal even if the network condition changes over time. It is to enable call quality.

Description

INTERNET PHONE SYSTEM FOR SELECTING VOICE COMPRESSION TYPE BY INTERNET PROTOCOL NETWORK PECULIARITY AND THEREOF METHOD}

1 is an exemplary block diagram of a Jongno Internet phone system,

2 is an exemplary configuration block diagram of an Internet phone system according to the present invention;

3 is an exemplary configuration block diagram of a network state measurement module used in the present invention;

Figure 4 is a flow chart showing a control flow for determining the priority of the speech compression method according to the present invention.

♧ description of the symbols for the main parts of the drawing ♧

100,200 call setup module 100 ', 200' ... network status measurement module

110 .... Ping Client 210 .... Ping Server

120 .... Voice Coder 140,240 .... Protocol Processor

160 .... Mic / ADC 220 .... Voice Decoder

260 .... DAC / Speaker

The present invention relates to an Internet phone system using an IP network, and more particularly, to improve an automatic voice compression method and a packet processing priority automatically in a situation in which packet delay time varies along a path such as a public IP network. The present invention relates to an internet phone system for selecting a voice compression method according to network characteristics and a method of selecting the same.

Currently, the number of Internet users is increasing rapidly every year, and accordingly, various services using the Internet network are provided.

In particular, the Internet phone system, which enables voice communication between PCs and PCs, PCs and telephones, telephones and telephones using VoIP technology, has reached a considerable level of technology. In addition, e-mail and fax services using the Internet network Is being developed.

In the Internet phone system, since a voice signal has to be transmitted, there is a problem that data throughput and transmission time increase rapidly. This is a voice compression technology that compresses and encodes a voice signal and then decodes it after transmitting the encoded coded data. This is being solved by commercialization.

In the conventional Internet phone system using an IP network, for example, as shown in FIG. 1, when a sender inputs a phone number of a receiver, the call setup module 10 finds an IP address and a port address of the receiver's Internet phone, and receives a call setup module ( 20) send and receive information necessary for call setup and control the voice coder 12 and the protocol processor 14 to connect the call, when the sender generates a voice microphone / ADC (ANALOG TO DIGITAL CONVERTER) (16) In voice, the voice signal is converted into an electrical signal, that is, an analog voice signal, into a digital voice signal, and the digital voice signal is compressed by an algorithm such as international standard G.723.1 or G.729.a through the voice coder 12. The compressed voice data is converted into a packet through the protocol processor 14 so that the compressed voice data can be transmitted through the IP network and then transmitted through the Internet network (IP network).

Meanwhile, the receiver side extracts only the compressed voice data by analyzing the protocol header / trailer from the received packet by the receiver protocol processor 24. After the compressed state is restored through the voice decoder 22, the compressed voice data is restored. Only the digital voice signal is extracted, and then the digital voice signal is converted into voice through DAC (DIGITAL TO ANALOG CONVERTER) / speaker 26 and the receiver listens to the voice.

However, when the call is made through the above-described process, if the path characteristics on the IP network are changed, the voice quality is deteriorated.

For example, if the packet forwarding delay time on the path of the IP network becomes long, the response of the other party arrives late, making the conversation impossible, or if the bandwidth on the path of the IP network decreases, packet loss occurs and voice quality drops sharply. The phenomenon can be mentioned.

In addition, since voice compression and packet processing priority are always fixed, there is a structural disadvantage in that voice transmission through a better method is fundamentally impossible even if the IP network is deteriorated or improved.

The present invention was created in view of the above-described problems of the prior art, and solved this problem. The use of an Internet phone over an IP network provides an optimal voice compression scheme and priority in packet processing according to a given IP network state. It is possible to automatically select the best call quality and to measure the status of the IP network whenever an individual call is initiated, so that the optimal voice quality can be ensured at all times. Its purpose is to provide an Internet phone system and a method of selecting the same.

In order to achieve the above technical problem, the present invention communicates with a receiver through an IP network according to an input telephone number, and a caller call setting module for connecting a call to each other, and a microphone for converting a sender's voice into a digital voice signal. / ADC, a voice coder for compressing the converted digital voice signal, a protocol processor for converting the compressed voice data into a packet form, an IP network for transmitting calls and packets, and a packet received from an IP network. A receiver call setup module for extracting voice data, a voice decoder for extracting the extracted compressed voice data into a digital voice signal, and a DAC / speaker for converting the extracted digital voice signal into an analog voice signal and converting it into an audible form. In the Internet phone system using the IP network configured to; Internet phone which selects voice compression method according to IP network characteristics by adding network status measurement module that records path, delay time and packet loss rate of IP network by pinging sender and receiver call setup module through IP network. It is characteristic of providing a system.

In addition, the present invention, in order to achieve the above technical problem, when the sender enters the receiver's telephone number, the network state measurement module finds the IP address and port address on the receiver's Internet phone system to find the IP packet between the two points. A first process of exchanging; If the number of IP network paths recorded in the first step is greater than or equal to a certain number, the coding time is prioritized to give priority to voice compression in the shortest order, and then the delay time is checked. A second process of checking a delay time after assigning; A third step of, if the delay time recorded after the second process is equal to or greater than the set time, gives priority to voice compression in order of high compression rate, and then measures packet loss rate, and if less, immediately measures packet loss rate; If the packet loss rate measured after the third process is greater than or equal to the set%, the Diffserv Code Point value is determined to be '42', and if the packet loss rate is less than '4', the fourth is configured to connect the call. There is also a feature to provide a voice compression method selection method of the Internet phone system to select the voice compression method according to the IP network characteristics including the process.

Hereinafter, with reference to the accompanying drawings will be described a preferred embodiment of the present invention.

2 is a block diagram of the Internet phone system according to the present invention, Figure 3 is a block diagram showing an example of the network state measurement module of FIG.

As shown in FIG. 2, the present invention provides an Internet phone system by adding network state measurement modules 100 'and 200' capable of measuring respective network states to call setup modules 100 and 200 on both a sender side and a receiver side. .

The network state measurement module 100 ', 200' may be configured as shown in FIG. 3, which is the same as or similar to the ping function used in the Unix / Linux system.

For example, the ping client 110 is mounted on the sender side network state measurement module 100 ', and a ping daemon, that is, the ping server 210, is mounted on the receiver side network state measurement module 200'. When the ping client 110 periodically sends a fixed size IP packet, the ping server 210 changes the header information of the received IP packet to echo to measure the network state between both modules through the IP network. .

At this time, the user can specify the transmission period and the size of the IP packet transmitted by the pink client 110, the contents of the packet is a form that repeats the number between 0 and 255, and the pink client 110 is echoed It measures the reception time of the IP packet and displays the difference from the transmission time, that is, the delay time of the IP network. If the packet is lost in the middle and does not echo after a certain time, the loss rate is calculated and displayed.

A method of selecting a voice compression method of the Internet phone system of the present invention having such a configuration will be described.

As shown in FIG. 4, when the sender first inputs the receiver's telephone number, the sender's network status measurement module 100 'finds an IP address and a port address on the receiver's Internet phone system and uses the ping function to find a connection between the two points. The IP network path, delay time and packet loss rate are measured (S10 and S11).

That is, as described above, when the ping client 110 loaded in the sender-side network state measurement module 100 'periodically transmits an IP packet having a predetermined size to the receiver-side ping server 210 through the IP network, the ping server 210 ) Receives this and echoes only the header information of the received IP packets.

After that, the recorded IP network path is checked and if there are more than N IP addresses (= IP routers) in between, the priority of the compression method to be used by the voice coder is set as follows. Priority is given to performing compression by the compression method (S12, S13, S14).

In this case, N, that is, the number of IP routers can be obtained by dividing the response delay time (typically 100ms) that a user can detect in a voice call by r, where r is the average packet delay time of the IP router. It is desirable to allow the user to set arbitrarily.

For example, if N is set to 3, if the recorded IP network has more than 3 paths, the coding time is shorter and the priority is given to voice compression such as ①G.711 ②G.726 ③G.729 ④G.723.1. If it is less than 3, priority is given to performing voice compression in G.711 manner without determining the priority.

After that, if the delay time is checked and the value is more than D seconds, the priority of the compression scheme to be used by the voice coder is set as follows. If the condition is not satisfied, the priority determined in the above-described IP network path determination process is set. In this state, the packet loss rate is measured (S15, S16, S17).

In this case, the D value can be arbitrarily determined by the user like the N value. For example, if the general call quality is desired, D = 100ms, and if the bandwidth is saved while sacrificing the call quality, D = 200ms is determined. .

In other words, if D = 100ms, the delay time is 100ms or more, and the general call quality is desired, and the compression rate is high.The priority is to perform voice compression in the order of ①G.723.1 ②G.729 ③G.726 ④G.711. The packet loss rate should be measured immediately if it is less than 100ms, and the packet loss rate is measured according to the priority with the priority set in the process of determining the upper level, that is, the number of IP network paths.

In this way, when the priority required for voice compression is determined, the loss rate of the recorded packet is measured (S18).

In this process, if the loss rate is more than L%, the priority of IP packet processing is set to Diffserv code point value = 42. Otherwise, the Diffserv code point value is set to 56. In this case, the L value may be arbitrarily determined by the user (S19). , S20, S21).

Here, '42' is an IP packet processing priority value that minimizes packet loss rate as medium priority and low drop precedence, and '56' is an IP packet processing priority value that safely maximizes transmission speed as expedited forwarding.

In addition, L% refers to the threshold of packet loss rate, L = 0.2% for general call quality, and L = 2% for bandwidth saving at the expense of call quality, where 0.2% is 50 It is the ratio of something like noise heard momentarily once a second.

When the voice compression method is determined through this process, the call setup module sends and receives the remaining necessary information to connect the call (S22).

When the call is connected, as shown in FIG. 2, the transmitter's voice is converted into a digital voice signal through the microphone / ADC 160, and the converted digital voice signal is compressed and compressed by the selected compression method through the voice coder 120. The voice data is converted.

Subsequently, the transmitter protocol processor 140 converts the compressed voice data into an Internet packet, and then transmits the voice packet of the transmitter to the receiver through the Internet (IP network).

The receiving side receives the voice packet and the compressed voice data is extracted by the receiver protocol processor 240.

The extracted compressed voice data is extracted as a digital voice signal through the voice decoder 220, and the extracted digital voice signal is converted into an analog voice signal through the DAC / speaker 260 and then listened to the receiver.

As described in detail above, the present invention provides the following effects.

First, it is possible to realize the most optimal call quality by automatically selecting the optimal voice compression method and priority for packet processing according to the given IP network status.

Second, it is possible to secure the best call quality even if the IP network state changes over time by measuring the IP network state each time the individual call is started and performing the optimal setting.

Third, currency delays are minimized and the problem of poor currency improves.

Claims (5)

Sender call setup module that communicates with the receiving party through the IP network according to the input telephone number and connects mutual calls, a microphone / ADC that converts the sender's voice into a digital voice signal, and compresses the converted digital voice signal. A voice coder, a protocol processor for converting the compressed voice data into a packet form, an IP network for transmitting calls and packets, a receiver call setting module for extracting compressed voice data from packets received through the IP network, and extraction An internet phone system using an IP network including a voice decoder for extracting a digital voice signal from compressed compressed data and a DAC / speaker for converting the extracted digital voice signal into an analog voice signal and converting the voice signal into an audible form; Perform a ping function on the sender and receiver call setup module through an IP network to record the path, delay time, and packet loss rate of the IP network, and determine the number of transit IP router addresses, the delay time, and the packet in the path of the IP network. An internet phone system for selecting a voice compression method according to the characteristics of an IP network, comprising adding a network state measurement module for setting a voice compression method of the voice coder according to a loss ratio . The method according to claim 1, The network state measurement module includes: a ping client loaded on the call setup module of the transmitting side; The voice compression method is selected according to the characteristics of the IP network, which comprises a ping server loaded in the call setup module to receive the IP packet transmitted through the IP network from the ping client and change only the header information. Internet phone system. A first step of the network state measuring module exchanging an IP packet between two points by searching for an IP address and a port address of the receiver's Internet phone system when the sender inputs the telephone number of the receiver; If the number of IP network paths recorded in the first step is greater than or equal to a certain number, the coding time is prioritized to give priority to voice compression in the shortest order, and then the delay time is checked. A second process of checking a delay time after assigning; A third step of, if the delay time recorded after the second process is equal to or greater than the set time, gives priority to voice compression in order of high compression rate, and then measures packet loss rate, and if less, immediately measures packet loss rate; If the packet loss rate measured after the third process is greater than or equal to the set%, the Diffserv Code Point value is determined to be '42', and if the packet loss rate is less than '4', the fourth is configured to connect the call. The voice compression method selection method of the Internet phone system to select the voice compression method according to the characteristics of the IP network, characterized in that it comprises a process. The method according to claim 3, The priority of the short time that the coding takes in the second process is as follows: ①G.711 ②G.726 ③G.729 ④G.723.1 Voice of the Internet phone system to select the voice compression method according to the characteristics of the IP network Compression method selection method. The method according to claim 3, The voice compression method of the Internet phone system that selects the voice compression method according to the characteristics of the IP network, characterized in that the priority of the compression ratio in the third process is in the order of ①G.723.1 ②G.729 ③G.726④G.711. Way.

Images