KR100420556B1 - Real time voice packet encryption/decryption system and encrypting/decrypting method using the same - Google Patents

Abstract

The present invention relates to a real-time voice packet encryption / decryption system and a method for encrypting / decoding the same. The present invention relates to a technique associated with a conventional technology of encrypting / decoding a voice packet of a Voice over Internet Protocol (VoIP) using a block cipher method. The present invention relates to a voice packet encryption / decryption system and a method of encryption / decryption thereof capable of minimizing difficulty and economic loss.

To this end, the present invention instead of the H.235 recommendations for encrypting all the information of the voice packet that is the output of the conventional vocoder compression / restoration unit, formant to convey the meaning of the voice signal using the characteristics of the vocoder compression / restoration unit Provided are a voice packet encryption / decryption system and a method of encryption / decryption, wherein only a component is extracted from a voice packet and the extracted formant component is encrypted / decrypted using a permutation ciper method.

Accordingly, the present invention eliminates the need for an expensive processor, which is essentially required for a voice packet encryption / decryption system using a conventional block cipher system, thereby minimizing the technical difficulties and economical losses associated with the prior art, thereby enabling high speed encryption of voice packets. Can be implemented.

Description

Real time voice packet encryption / decryption system and encrypting / decrypting method using the same}

The present invention relates to a real-time voice packet encryption / decryption system and a method for encrypting / decoding the same. The present invention relates to a technique associated with a conventional technology of encrypting / decoding a voice packet of a Voice over Internet Protocol (VoIP) using a block cipher method. The present invention relates to a voice packet encryption / decryption system and a method of encryption / decryption thereof capable of minimizing difficulty and economic loss.

Voice over Internet Protocol (VoIP) is a technology that supports simultaneous transmission of data and voice using IP (Internet Protocol), a network standard protocol. H.323, Session Initiation Protocol (SIP), and Media Gateway Control Protocol (MGCP) Standardization is underway through several protocols, such as).

While the ultimate goal of VoIP is to achieve the same voice quality and reliability as the Public Switched Telephone Network (PSTN), the voice quality and reliability are related to voice codes, compression, transmission and recovery technologies. Is not easy. Accordingly, in recent years, issues such as quality of service (QoS) and open call control have become a big issue in terms of voice quality and reliability, but security studies are still insignificant.

The only H.235 recommendation among security protocols recommends overall security measures in H.323-based systems, and standardization is still underway. The H.235 Recommendation proposes various security forms such as authentication and signaling security, media stream security, and protection from hacking, but in the present invention, the media stream security, particularly in the encryption / decryption of voice packets among media, Propose a new way.

Hereinafter, a method of encrypting / decoding a general VoIP voice packet proposed in the H.235 Recommendation will be described with reference to the conventional VoIP voice packet encryption / decoding system shown in FIG. 1.

Referring to FIG. 1, the VoIP voice packet encryption / decryption system according to the related art is composed of a vocoder compression / recovery unit 10, a padding unit 20, and a block encryption / decryption unit 30. Network 40 exchanges encrypted data with multiple voice packet encryption / decryption systems.

The vocoder compression / recovery unit 10 compresses the 'voice input' (a digital signal sampled at a constant period, for example, 8 KHz) into a voice packet and transmits the compressed voice to the padding unit 20, or the padding unit 20 Restores the voice packet transmitted from the original voice signal and outputs a 'voice output'.

In general, compression protocols of the vocoder compression / recovery unit 10 proposed in the H.235 recommendation include G.711, G.722, G.723, G.728, and G.729. The G.723.1 protocol with 5.3Kbps and 6.3Kbps data rates, and the G.729A protocol with compression rate of 1/8 or more are most widely used because they have a 8Kbps data rate for 10ms per frame.

The padding unit 20 pads the voice packet generated by the vocoder compression / recovery unit 10 to an input size of the block encryption / decryption unit 30 and outputs it to the block encryption / decryption unit 30. This is because the feature of the block cipher has a property of encrypting a block of a certain size, so that the vocoder compression / restoration unit 10 outputs a voice packet that is output in different sizes in one size.

The block cipher encryption / decryption unit 30 receives an output of the padding unit 20 and a key, for example, an encryption key and a decryption key, and uses the encryption key to determine the padding unit 20. The output is encrypted to be transmitted to the network 40 or the encrypted data transmitted from the network 40 is input, and the padding unit 30 is decrypted by decrypting the encrypted data transmitted from the network 40 using the decryption key. Will output Here, the key is determined during the H.245 capability exchange proposed by the H.235 protocol.

In general, the block cipher encryption / decryption algorithms proposed in the H.235 Recommendations include 56-bit Data Encryption Standard (DES), 56-bit RC2 compatible, and 168-bit Triple DES, depending on the level of security and the amount of computation. In the case of triple DES, it is possible to select among Electronic CodeBook (ECB), Cipher Block Chaining (CBC), Cipher FeedBack (CFB), Output FeedBack (OFB), and Counter mode.

Meanwhile, the encrypted data output from the block cipher encryption / decryption unit 30 is transmitted to the block cipher encryption / decryption unit 30 of another voice packet encryption / decryption system through the network 40, and then decrypted. Is sent). Then, the decoded data is transmitted to the vocoder compression / restoration unit 10 through the padding unit 20, and is restored to the original voice signal through the vocoder compression / restoration unit 10 and output as 'voice output'. do.

As described above, the conventional voice packet encryption / decoding system employs a block encryption method for encrypting / decoding a voice signal. The block encryption method generally maintains a high degree of complexity while delaying due to a large amount of calculation. This is because it is mainly used for encryption / decryption of data that is not sensitive to delay. However, since voice signals have delay-sensitive real-time propagation characteristics unlike data, many technical difficulties and economic losses are required in encrypting all information of a voice packet using a block cipher method every frame.

Accordingly, the present invention has been made to solve the above problems, and the voice packet is not viewed as simple data, but is viewed as a single media stream having the characteristics of the voice signal. An object of the present invention is to provide a VoIP voice packet encryption / decryption system.

To this end, the present invention instead of the H.235 recommendations for encrypting all the information of the voice packet that is the output of the conventional vocoder compression / restoration unit, formant to convey the meaning of the voice signal using the characteristics of the vocoder compression / restoration unit A method of extracting only components from a voice packet and extracting formant components using a permutation ciper method is presented.

1 is a block diagram of a VoIP voice packet encryption / decryption system according to the prior art.

FIG. 2 is a diagram illustrating a Code Excited Linear Prediction (CLEP) coding method among compression methods of a vocoder compression / restore unit.

3A and 3B are configuration diagrams of a VoIP voice packet encryption / decryption system according to an embodiment of the present invention.

4A and 4B are flowcharts illustrating an encryption / decryption method of the VoIP voice packet encryption / decryption system shown in FIG. 3.

<Explanation of symbols for main parts of drawing>

10: Vocoder Compression / Restoration 20: Padding

30: block cipher encryption / decryption unit 40: network

50: formant detector 60: residual component detector

110: vocoder compression unit 120: formant extraction unit

130: substitution password encryption unit 140: substitution table generation unit

150: packet combiner 210: vocoder recovery unit

220: formant synthesis unit 230: substitution code decoding unit

240: substitution table generation unit 250: packet separation unit

The present invention provides a vocoder for compressing a voice signal and outputs it in the form of a voice packet, a formant extractor for separating and extracting a formant component and a residual component of the voice packet component output from the vocoder, and the satiety. A substitution table generation unit for generating a substitution table used to encrypt the component, an encryption unit for encrypting the formant component using the substitution table, and the encrypted formant component and the formant extraction unit It provides a voice packet encryption system including a packet combiner for combining the remaining components of the transmitted voice packet.

The present invention also provides a packet separator for separating the encrypted formant component and the residual component from a voice packet in which an encrypted formant component and a residual component are combined, and used to decrypt the encrypted formant component. A substitution table generation unit for generating a substitution table, a decryption unit for decrypting the encrypted formant component using the substitution table, and the decrypted formant component and the residual components transmitted from the packet separation unit And a vocoder for restoring the decoded formant component and the residual component synthesized through the formant synthesis unit and outputting a speech signal.

In addition, the present invention comprises the steps of compressing a voice signal in the form of a voice packet, extracting the formant component by separating the formant component and the residual component in the voice packet, and using the substitution table for the formant component And encrypting and outputting a combination of the remaining components of the voice packet and the encrypted formant component.

In another aspect, the present invention is the step of separating the encrypted formant component and the residual component from the voice packet in which the encrypted formant component and the residual component are combined, and decrypting the encrypted formant component using a substitution table. And synthesizing the residual component and the decoded formant component of the speech packet, and outputting a speech signal by restoring the synthesized residual component and the decoded formant component. .

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

FIG. 2 is a diagram illustrating a Code Excited Linear Prediction (CLEP) encoding method which is generally used among compression methods of a vocoder compression / recovery unit applied to implement preferred embodiments of the present invention.

First, in order to compress the voice, the formant component corresponding to the 'voice input', for example, the meaning of the voice signal, is first extracted through the formant detector 50, and the remaining components are detected by the residual component detector 60 after extracting the formant. Model residual components. When a person speaks, the air discharged through the lungs passes through the saints, the oral cavity, and the shape of the lips. The formant modeling the saints, the oral cavity and the shape of the lips with one filter is a formant. The remaining residual components are pitches modeling the tremor of the vocal cords separating voiced and unvoiced voices, and codebooks, which are other residual components.

For example, when a person speaks a word, it consists of three parts, a formant component, a pitch component, and a residual component, on the time axis 50a of the voice signal, and on the frequency axis 50b, the entire outline of the envelope is represented. It consists of a formant component, a pitch component showing a dense part, and a residual component. Referring to the drawing showing the time and frequency axes 60a and 60b after the formant component extraction, only the pitch component and the residual component remain on the time axis 60a, and the formant corresponding to the entire contour on the frequency axis 60b. It can be seen that the component has been removed. Subsequently, the residual component detector 60 models the remaining components except for the formant component. The pitch component and the residual component modeled after this become an important factor that determines the quality of the voice signal.

As such, when a person wants to deliver a voice, the most important element corresponding to the meaning of the voice to be delivered is a formant component, and the other components are models for improving sound quality. Therefore, if only the formant component corresponding to the meaning is encrypted rather than encrypting all the packets, it is possible to perform the encryption having the desired degree of content with a small amount of computation.

In general, in G.723.1, which is widely used as the compression protocol of the vocoder compression / restore described in H.235 Recommendation, the formant component of 24 bits (for example, 800bps) is extracted in 20ms, both 5.3Kbps and 6.3Kbps. The 729A extracts 18 bits (for example, 1.2 Kbps) of formant components in 10 ms, and all other components are used to model pitch components and residual components.

Hereinafter will be described a preferred embodiment of the present invention implemented by applying the operating characteristics of the vocoder compression / restoration described above.

3A is a schematic diagram of a VoIP voice packet encryption / decryption system according to a preferred embodiment of the present invention, which applies a substitution encryption scheme as an encryption scheme for encrypting a formant component.

Referring to FIG. 3A, a VoIP voice packet encryption system according to a preferred embodiment of the present invention includes a vocoder compression unit 110, a formant extraction unit 120, a substitution code encryption unit 130, a substitution table generation unit 140, and a packet. Combining unit 150, and outputs the encrypted data to a plurality of voice packet decryption system via a network.

The vocoder compression unit 110 uses a vocoder compression unit having the characteristics shown in FIG. 2, and compresses the 'voice input' (a digital signal sampled at a constant period, for example, 8 KHz) into a voice packet to form a form. Transmission to the extraction unit 120. For example, the vocoder compression unit 110 performs voice compression using G.723, G.729, and the like.

The formant extractor 120 separates the formant component and the residual components from the components of the voice packet generated through the vocoder compression unit 110, extracts only the double formant component, and transmits the extracted formant component to the substitution coder 130. The remaining components other than the formant component are transmitted to the packet combiner 150.

The substitution code encryption unit 130 performs a substitution operation for exchanging each bit of the formant component transmitted from the formant extraction unit 120 based on the substitution table transmitted from the substitution table generation unit 140 and outputs the packet. Transmission to the coupling unit 150.

The substitution table generation unit 140 generates a substitution table based on the shared key and transmits the substitution table to the encryption encryption unit 130. The substitution table generator 140 may generate a weak key according to the position and the key of the formant. In order to prevent this, an appropriate constraint must be applied, and the secret key is seeded for each frame. To generate another substitution table. In addition, for normal operation, proper synchronization between the transmitter and receiver must be premised.

The packet combiner 150 recombines the remaining components output from the formant extracting unit 120 without being encrypted and the encrypted formant components encrypted and output from the substitution encryption encryptor 130 in the form of voice packets. Through the transmission to the voice packet decoding system.

Meanwhile, a voice packet decryption system for decrypting a voice packet including an encrypted formant component transmitted from the voice packet encryption system of FIG. 3A is illustrated in FIG. 3B. This system applies a substitution cipher as a decoding scheme for decoding the formant component.

Referring to FIG. 3B, the VoIP voice packet decoding system according to the preferred embodiment of the present invention includes a vocoder restoration unit 210, a formant synthesis unit 220, a substitution code decoding unit 230, a substitution table generation unit 240, and a packet. It is composed of a separation unit 250, receives a voice packet containing an encrypted formant component from a voice packet encryption system through a network, and decrypts and outputs this voice packet.

The packet separator 250 separates the encrypted formant component and the remaining components among the components included in the voice packet transmitted from the voice packet encryption system shown in FIG. And the residual component is transmitted to the formant synthesis unit 220.

The substitution code decryption unit 230 decrypts the encrypted formant component transmitted from the packet separation unit 250 based on the substitution table transmitted from the substitution table generation unit 240 and transmits the decrypted formant component to the formant synthesis unit 220.

The substitution table generation unit 240 generates a substitution table based on the shared key and transmits the substitution table to the substitution password decoding unit 230. The substitution table generator 240 may generate a weak key according to the position and the key of the formant. In order to prevent this, an appropriate constraint must be applied, and the secret key is seeded for each frame. To generate another substitution table. In addition, for normal operation, proper synchronization between the transmitter and receiver must be premised.

The formant synthesis unit 220 receives the residual components transmitted from the packet separator 250 and the decoded formant components transmitted from the substitution code decoder 230, and synthesizes the two components in the form of a voice packet. To the vocoder recovery unit 210.

The vocoder reconstructor 210 uses a vocoder reconstructor having the characteristics shown in FIG. 2 and reconstructs a voice packet transmitted from the formant synthesizer 220 to an original voice signal and outputs a 'voice output'. For example, the vocoder reconstruction unit 210 performs voice compression / restore using G.723 and G.729.

Hereinafter, an encryption and decryption method of the VoIP voice packet encryption and decryption system of the present invention including the above configuration will be described with reference to the flowcharts shown in FIGS. 4A and 4B.

<Encryption method>

Referring to FIG. 4A, first, the shared key used in the substitution table generation unit 140 properly exchanges the secret key by the H.245 capability exchange method proposed in the recommendation. In this environment, the vocoder compression unit 110 compresses the input 'voice input' (that is, the voice signal) and outputs it to the formant extraction unit 120 in the form of a voice packet (step S11). Thereafter, the formant extractor 120 extracts only the formant component among the components included in the voice packet, and transmits the formant component to the pre-cryptographer 130, and transmits the remaining components to the packet combiner 150 (step S12). ).

Meanwhile, the substitution table generation unit 140 generates an appropriate substitution table using the exchanged shared key and transmits the substitution table encryption unit 130. Subsequently, the substitution code encryption unit 130 encrypts the formant component transmitted from the formant extraction unit 120 using the substitution table, and transmits the encrypted formant component to the packet combiner 150 (step S13). The packet combiner 150 recombines the encrypted formant component and the unencrypted residual component into the format of the voice packet (step S14).

<Decryption method>

Referring to FIG. 4B, the packet separator 250 separates the encrypted formant component and the remaining residual components from the voice packet transmitted through the network, and transmits the encrypted formant component to the substitution code decryption unit 230. The remaining components are transmitted to the formant synthesis unit 220 (step S15). Subsequently, the substitution code decryption unit 230 decrypts the formant component appropriately encrypted using the substitution table transmitted from the substitution table generation unit 240 and transmits the formant component 230 to the formant synthesis unit 230 (step S16). At this time, the substitution table is transmitted to the substitution code decryption unit 230 in a state where synchronization is set through a secret key.

Subsequently, the formant synthesizing unit 220 combines the decoded formant component and the residual component again in the form of a voice packet and transmits it to the vocoder restoration unit 210 (step S17). The vocoder recovery unit 210 restores the voice packet transmitted from the formant synthesizing unit 220 to the original voice signal and outputs a 'voice output' (step S18).

Although the preferred embodiment of the present invention described above has been described separately from each other by a voice packet encryption system and a decryption system, this is to help understanding of the technical idea of the present invention, and the present invention is separated into an encryption system and a decryption system. Note that it does not have an independent configuration. That is, the voice packet encryption system and decryption system of the present invention are configured in the same configuration to implement the encryption and decryption of the voice packet, it is composed of one encryption / decryption system capable of encryption and decryption. In addition, although the technical spirit of the present invention has been described in detail in the above-described preferred embodiment, it should be noted that the above implementation is for the purpose of description and not of limitation. In addition, those skilled in the art will understand that various embodiments are possible within the scope of the spirit of the present invention.

As described above, the present invention proposes a new VoIP voice packet encryption / decryption system using a substitution encryption method that has a low computational amount, thereby requiring an expensive processor that is required in a VoIP voice packet encryption / decryption system using a conventional block encryption method. There is no need. Accordingly, the present invention can implement high-speed encryption of voice packets while minimizing the technical difficulties and economic losses associated with the prior art.

Claims (8)

A vocoder for compressing a speech signal and outputting the speech signal in the form of a speech packet; A formant extracting unit for separating and extracting a formant component and a residual component among the voice packet components output from the vocoder; A substitution table generator for generating a substitution table used to encode the formant component; An encryption unit for encrypting the formant component using the substitution table; And And a packet combiner for combining the encrypted formant component and the residual components of the voice packet transmitted from the formant extractor. The voice packet encryption system of claim 1, wherein the formant extracting unit extracts the formant component from one component of the voice packet in one frame unit. The voice packet encryption system of claim 1, wherein the substitution table generation unit updates the substitution table using a shared key every frame. A packet separator for separating the encrypted formant component and the residual component from a voice packet in which an encrypted formant component and a residual component are combined; A substitution table generator for generating a substitution table used to decrypt the encrypted formant component; A decryption unit for decrypting the encrypted formant component using the substitution table; A formant synthesis unit for synthesizing and outputting the decoded formant component and the residual component transmitted from the packet separator; And And a vocoder for outputting a speech signal by restoring the decoded formant component and the residual component synthesized through the formant synthesis unit. (a) compressing the speech signal into a speech packet; (b) extracting the formant component by separating the formant component and the residual component from the voice packet; (c) encoding the formant component using a substitution table; And (d) combining and outputting the remaining components of the voice packet and the encrypted formant component. 6. The method of claim 5, wherein the substitution table is generated using a shared key every frame. (a) separating the encrypted formant component and the residual component from a voice packet in which an encrypted formant component and a residual component are combined; (b) decrypting the encrypted formant component using a substitution table; (c) synthesizing the residual components and the decoded formant components of the speech packet; And and (d) restoring the synthesized residual component and the decoded formant component and outputting a speech signal. 8. The method of claim 7, wherein the substitution table is generated using a secret key every frame.