RU2249304C1 - Method for coding voice signal - Google Patents

Abstract

FIELD: communications engineering.

SUBSTANCE: proposed method includes signal division into time intervals, utterance of subscriber whose voice signal is to be encoded being recorded in advance; transmission of these intervals over different communication channels chosen from preset law; filling of spaces between them in each channel with background produced from recorded utterance by dividing it into time intervals of same length and interchanging obeying random law.

EFFECT: enhanced degree of transferred signal protection against unwanted persons.

1 cl, 8 dwg

Description

The invention relates to communication technology and can be used to encode speech signals.

A known encoding method, including limiting the speech signal, differentiating, dividing the signal spectrum into a number of frequency bands, their subsequent transmission to the subscriber with the permutation of the frequency bands according to a random law (see AS USSR No. 501489, IPC ⁷ N 04 K 1/06, publ. 30.01.76 g.).

The disadvantages of this method are the complexity of the method and the insufficient degree of secrecy of the transmitted information.

Closest to the claimed method is a method of encoding speech signals, which consists in dividing the speech signal into time intervals, dividing the obtained time intervals into a number of subintervals, rearranging them randomly, choosing the duration of the time intervals is performed according to a random law, the choice of the number and duration of the subintervals in the interval produced according to a random law; for the resulting sub-intervals, a choice of the direction of reading signals according to a random law is made y, at the subintervals selected according to a random law, the converted signal is read at a speed exceeding the recording speed, and the temporary pauses resulting from the difference in speeds are randomly located on the time interval and filled with a noise signal, the parameters of which depend on the type of the current pseudorandom sequence ( see RF patent No. 2120700, IPC 7 H 04 K 1/06, publ. 10/20/1998).

The disadvantages of the prototype are the complexity of the method and the ability to obtain an outside observer of the entire signal completely.

An object of the invention is to increase the degree of security of the transmitted signal from prying observers.

This technical problem is achieved by the fact that in the known method of encoding a speech signal, including dividing the signal into time intervals, according to the invention, a subscription of a subscriber’s speech fragment is preliminarily created, whose speech signal is encoded, and the transmission of time intervals is carried out over several communication channels, while time intervals transmit on different channels selected according to a given law, and the gaps between them on each channel are filled with the background created from the recorded fragment of speech, by dividing it into time intervals of the same duration and their permutations according to a random law.

This method will increase the degree of security of the transmitted speech signal from outsiders and simplify the implementation of the method.

The invention is illustrated by drawings, where figure 1 presents a structural diagram of a device that implements the proposed method; figure 2 presents a graph of a fragment of the speech of the subscriber used to create a background, divided into time intervals; figure 3 presents a graph of the encoded speech signal, divided into time intervals; figure 4-7 presents graphs of signals transmitted over used communication channels; on Fig presents a graph of the received and decoded signal.

The structural diagram of a device that implements the method consists of a memory unit 1, a divider 2, background generators 3-6, switches 7 and 16, transmitters 8-11, receivers 12-15 and the adder 17.

An example of a specific implementation of the method.

Before starting the encoding, the subscriber whose speech signal is encoded speaks up any text that is written to memory unit 1 and used by the background generators 3-6 as the base. 3-6 provide background pattern generators, respectively, in the communication channels A, B, C and D by dividing the base signal on the _Q-1 time intervals t, t _p2, t _p3, t _{F4, F5} t, t _{Feeder # 6,} t _{F7, F8} t, t _f9 , t _f10 , t _f11 , t _f12 , etc. and rearrange them randomly. The encoded speech signal using divider 2 is divided into time intervals t ₁ , t ₂ , t ₃ , t ₄ , t ₅ , t ₆ , t ₇ , t ₈ , t ₉ , t ₁₀ , t ₁₁ , t ₁₂ , etc. . the same duration as the background time intervals. Using switch 7, the time intervals of the speech signal are distributed over different communication channels according to a given law, for example:

ACBDCADBDCAB, etc., are transmitted by transmitters 8-11, received by receivers 12-15, by a switch 16 synchronized with switch 7, only time intervals of the encoded speech signal are selected from the total signal flow on all channels and summed by an adder 17 to obtain a decoded speech signal identical to the encoded one.

The time intervals t ₁ and t ₆ is transmitted on channel A transmitter 8 (see Figure 4 -. Continuous line) between background _F5 time intervals t, t _F9, t _F12, t _p2, t _Q-1, t _F4, t _{Feeder # 10,} t _F4 , and t _f5 obtained from the background generator 3 (see figure 4 is a dashed line); the time intervals t _3, t ₈ and t ₁₂ are transmitted on channel B transmitter 9 (see Figure 5 -. continuous line) between background _F4 time intervals t, t _f11, t _F7, t _Q-1, t _F8, t _F12, t _F5 , t _ф1 , and t _ф12 obtained from the background generator 4 (see Fig. 5 is a dashed line); time intervals t ₂ , t ₅ . and t is transmitted on channel ₁₀ to the transmitter 10 (see Figure 6 -. continuous line) between the background time intervals t _{F12, F5} t, t _{Feeder # 10,} t _p3, t _p2, t _{F9, F7} t, t _F4, and _F8 t obtained from the background generator 5 (see Fig.6 is a dashed line); time intervals t _4, t ₇ and t ₉ is transmitted over a channel transmitter 11 D (7 - continuous line) between the background time intervals t _{F12, F5} t, t _{Feeder # 10,} t _p3, t _p2, t _{F9, F7} t, t _f4 , and t _f8 obtained from the background generator 6 (see Fig. 7 is a dashed line), etc.

The switch 16, synchronized with the switch 7, according to a given law is connected to the receivers 12-15, choosing from the entire stream of time intervals, only those that make up the encoded speech signal, sum them up by the adder 17, receiving a decoded speech signal identical to the encoded one (see Fig. .8).

Using the proposed method will allow, in comparison with the prototype, to increase the degree of security of the transmitted information from outsiders and simplify the process of implementing the method.

Claims (1)

A method of encoding speech signals, including dividing the signal into time intervals, characterized in that it preliminarily creates a record of a subscriber’s speech fragment, whose speech signal is encoded, and the transmission of time intervals is carried out through several communication channels, while the time intervals are transmitted over different channels selected for a given law, and the gaps between them on each channel are filled with a background created from a recorded fragment of speech by dividing it into time intervals of the same duration and rearranging them according to luchaynomu law.

Images