JP3073043B2 - Same frequency communication method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to the same frequency communication system used for wireless communication and other voice communication.

[0002]

2. Description of the Related Art When one line is used in wireless communication, only half-duplex communication that talks in sequence can be performed, and two lines are required to perform simultaneous full-duplex communication. Met.

[0003] However, there is also a technique of transmitting data simultaneously on one line, and an example is shown in FIG.

That is, first, a voice signal having a bandwidth fc is filtered to reduce the bandwidth to 1/2, that is, fc / 2, and then cut a high band while leaving a low band. This is A / D converted at the sampling clock frequency fs and stored in the RAM. Then, the digital signal is D / A-converted at 2 fs at twice the sampling clock frequency.

As a result, as shown in FIG. 5, the bandwidth returns to the same as before the filtering, but the speed of the talk can be doubled because the speed of the D / A conversion is doubled.

[0006] If packets are signaled and transmitted alternately in this manner, it is possible to transmit and receive simultaneously while using one line.

However, in this method, only the lower band is left by filtering, and the upper band is cut, so that the intelligibility is lowered, the sound quality is remarkably deteriorated, and speaker identification becomes difficult.

[0008]

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned conventional problems, and its purpose is to reduce the quality of voice so much.
An object of the present invention is to provide an identical frequency communication system capable of simultaneous transmission and reception from both directions.

[0009]

SUMMARY OF THE INVENTION In order to achieve the above object, the present invention performs A / D conversion of an input analog audio signal having a bandwidth of fc using a sampling clock frequency fs, and converts this digital signal into a fast Fourier transform (FFT) or the like. A frequency analysis is performed using a filter bank summation method (hereinafter referred to as an FBS method) which is an application thereof, and the spectrum obtained by this frequency analysis is band-compressed by performing thinning-out and further shifting, and the band-compressed signal is obtained. The signal is synthesized by the inverse FFT or the FBS method and returned to the time domain signal. This time domain signal is stored in the RAM, and the signal in the RAM is converted to a D / A signal at a frequency 2fs, which is twice the sampling clock frequency fs at the time of A / D conversion. A conversion is performed, and this analog signal is transmitted.

[0010] The above FBS method is based on "Digital Pr".
ocessing of Speech Signals ", PRENTICE-
HALL, pp. 250-310 (1979).

Therefore, according to the present invention, D / A conversion is performed at a sampling clock frequency that is twice that of A / D conversion.
By performing the / A conversion, the speaking speed is doubled, and the transmission time can be reduced to half. Since the audio band is originally compressed to 元, when the speaking speed is doubled, the audio band returns to its original state, and there is no chance that the audio will be difficult to hear.

[0012] In the case of communication, if a packet is transmitted as a digital signal and transmitted, the speaking speed is doubled, but the voice band remains the same, so that it is possible to transmit with a data amount of 1/2. Becomes

As a result, the transmission time is halved and full-duplex communication is made possible by repeating transmission and reception alternately even on a line where only half-duplex communication was conventionally possible.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below with reference to the drawings.

What is characteristic of the present invention is that, as in the prior art, only the lower band of the input analog audio signal is left,
That is, band compression is performed without cutting off the upper band at all and leaving the spectrum of the upper band to some extent. That is, for the A / D converted audio signal,
The point is that the frequency analysis is performed using the FFT or FBS method, and the obtained frequency spectrum is subjected to decimation or shift processing by the most appropriate method to perform band compression.

FIG. 1 is a block diagram of a transmitting side in a communication system using the method of the present invention. FIG. 2 is a block diagram of a receiving side in a communication system using the method of the present invention.
FIG. 5 is an explanatory diagram of an operation on a transmission side of a communication system using the method of the present invention.

1 and 3, an analog audio signal S1 having a bandwidth fc supplied to an input terminal 10 is an LPF.
By passing through 12, the high frequency component is cut. This analog signal S2 has a sampling clock frequency f supplied from the sampling clock generator 14.
The signal is converted into a digital signal S3 by the A / D converter 16 using s.

Next, the digital signal S3 is converted into a DSP (Digital Signal) by the frequency spectrum analyzer 18.
(1) The frequency spectrum is analyzed by an analysis process such as FFT or FBS using a processor or the like. For these spectra S4, the spectrum thinning shift unit 20
Is different from the conventional method in that a spectrum is decimated by using a selection method having a feature that a high-frequency spectrum such as selecting every other spectrum is also retained, and the remaining spectrum is further reduced to a band of fc / 2 or less. Shift the band to 1 /
Make 2 As described above, when the DSP is used, it is possible to compress the band while leaving the spectrum of the high band.

The band-compressed signal S5 from the spectrum thinning shift unit 20 is returned to a time-domain signal S6 by an inverse FFT or an FBS method, which is an application thereof, in a time-domain signal synthesizing unit 22. RAM2
4 and store it. The signal S7 stored in the RAM 24 is D / A-converted by the D / A converter 26 at a clock frequency 2fs which is twice the A / D conversion supplied from the clock generator 25, and the analog signal S8 is further converted. The output signal S9 is obtained by passing through the LPF 28. This output signal S9 is transmitted via the communication device 30.

As described above, by performing D / A conversion at a clock frequency twice that at the time of A / D conversion at the time of D / A conversion,
The speaking speed is doubled, and the transmission time can be halved. Since the band is originally compressed to ２, when the speaking speed is doubled, the band of the sound is returned to the original one, and there is no chance that the sound is difficult to hear because of the conventional win.

In the case of communication, when packetized as a digital signal and transmitted, the speaking speed is doubled, but since the bandwidth is the same as before, it is possible to transmit with a half data amount. It becomes possible.

As a result, the transmission time is reduced by half, and full-duplex communication is made possible by repeating transmission and reception alternately even on a line where only half-duplex communication was conventionally possible.

In FIG. 2, the high frequency component of the signal S 10 received by the communication device 30 is cut by the LPF 32. The low-frequency signal S11 from which the high-frequency component has been cut is A / D-converted into a digital signal S12 by an A / D converter 34 at a sampling clock frequency 2fs from a sampling clock generator 33, and further at the same sampling clock frequency 2fs. The data is written to the RAM 36. RA
The signal in M36 is read out as S13 at a clock frequency fs which is 1/2 of that written from the clock generator 38, and supplied to the frequency spectrum analyzer 40.

In this way, the speaking speed is reduced by half, and returns to the normal conversation speed. Also, the bandwidth is halved.

In order to return the half band to the original band, the spectrum signal S14 obtained by the frequency spectrum analysis unit 40 is supplied to the spectrum shift interpolation unit 42 corresponding to the processing on the transmitting side. Here, the spectrum shifted at the time of transmission is restored by the reverse operation, and the missing spectrum is interpolated.

The signal S from the spectrum shift interpolation unit 42
15 is synthesized into a time-domain signal S16 by a time-domain signal synthesizing section 44, and further D / A-converted by a D / A converter 46 according to the clock frequency fs from the clock generating section 38. The D / A converted analog signal S17 is LP
By passing through F48, the original analog audio signal S1 is obtained.

[0027]

As described above, according to the present invention,
Frequency analysis of spectrum using FFT and FBS methods,
By compressing the band, it becomes possible to perform simultaneous transmission and reception two-way communication without causing significant deterioration of sound quality which is a problem of the related art.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a transmission side of a communication system using the method of the present invention.

FIG. 2 is a configuration diagram of a receiving side of a communication system using the method of the present invention.

FIG. 3 is an explanatory diagram of the method of the present invention.

FIG. 4 is an explanatory diagram of a conventional communication system.

FIG. 5 is an explanatory diagram of a conventional communication system.

[Explanation of symbols]

12 LPF, 16 A / D converter, 18 frequency spectrum analysis unit, 20 spectrum thinning shift unit, 22
Time domain signal synthesizer, 24 RAM, 26 D / A converter, 28 LPF, 30 communication equipment

Claims (1)

(57) [Claims] 1. An input analog audio signal having a bandwidth of fc is digitally converted (A / D converted) by a sampling clock frequency fs, and this digital signal is subjected to a fast Fourier transform (FFT) or a filter bank summation method as an application thereof. , The frequency spectrum obtained by the frequency analysis is thinned out, and the remaining spectrum is band-compressed by performing a shift process to a band of fc / 2 or less, and the band is compressed. Signal
A synthesis process is performed by an inverse FFT or a filter bank summation method to return to a time domain signal, and this time domain signal is stored in a RAM,
A same frequency communication system, wherein a signal in a RAM is D / A converted at a frequency 2fs which is twice the sampling clock frequency fs at the time of A / D conversion, and this analog signal is transmitted.