JP3092157B2 - Communication signal compression system and compression method - Google Patents

Description

Description: FIELD OF THE INVENTION The present invention relates to communication systems and, more particularly, to communication signal compression that facilitates increasing communication capacity of communication systems.

2. Prior Art Communication systems are well known in the related art. Samuel Morse US Patent 1,647 (1840) and Alexander Graham Bell US Patent 174,
Since the early technology of 465 (1876), the entire telecommunications industry has evolved across the globe and beyond.

Message encoding and synchronization has played an important role throughout the historic development of communication systems. For example, prior to the invention of the telephone, messages were encoded in Morse code, with corresponding pulses carrying the encoded message via telegraph lines, and receiving and decoding the pulses. The response to the message could be transmitted in the reverse operation. Direct real-time communication between the two was not possible because the messages were manually encoded and decoded.

With the advancement of telephones, real-time communication has become possible by electronic encoding of voice patterns into communication signals and by wire transmission of the communication signals between two telephones. Transmission rates of communication signals well above the speed of sound have allowed real-time voice communication between individuals over significant distances without appreciable time delays.

At present, communication signals are not restricted to wired transmission lines, but can be transmitted by microwaves, radio waves and optical fibers. These advances have enabled global real-time communications.
Users have also sought real-time communication services.

Time-division multiplexing has been used to increase the transmission capacity of various carrier media, rather than the usual wireline telephony system transmitting a single telephony signal over a pair of wires. For example, many communication signals can be multiplexed and transmitted via a single optical fiber. That is, a single optical fiber cable can replace hundreds of cable pairs and provide even greater signal transmission capacity.

The same idea has been applied to wireless telephone systems.
Wireless telephone systems are well known in both fixed and mobile applications. In remote rural areas where the installation and maintenance of regular telephone lines is cost prohibitive, wireless telephone systems facilitate the provision of telephone services through wireless transmission between base stations and various subscriber stations. Mobile radio telephone systems are becoming increasingly popular in the form of cellular car telephones that have become widely available.

Wireless telephone systems use a set of selected radio frequencies to transmit communication signals instead of cables. A typical fixed radiotelephone system has thirteen pairs of selected frequencies or channels for signaling between a subscriber station and a serving base station.

Due to the limited frequency band for wireless telephones, time division multiplexing has been used to allow for increased transmission capacity of wireless communication systems. For example, U.S. Patent No. 4,675,863, issued June 23, 1987, "Subscriber RF Telephone Sys
tem For Providing Multiple Speech and / or Data Sign
als Simultaneously Over Either A Single or A Plura
lity of RF Channels ”(corresponding to JP-A-61-218297)
Discloses a fixed / mobile radio telephone system using 26 channel pairs, each capable of transmitting up to four communication signals simultaneously.

Unlike optical fiber transmission, which allows high-speed transmission of communication signals to and from a destination in the gigahertz range, the transmission capacity of radio frequency carriers (channels) is extremely limited.

Audio signal compression techniques have been used to increase the transmission capacity of wireless channels. A technique that has been successful so far is described in US Patent Application No. 667,446 filed November 2, 1984.
No. (corresponding PCT International Publication No. WO / 02726) is a residual excitation linear predictive coding (RELP). RELP makes it possible to compress a 64 kilobits per second voice communication signal into a 14.6 kilobits per second coded signal for transmission over a wireless channel. If the 14.6 kbit / s signal is decoded on the receiving side to reproduce the 64 kbit / s signal, there is almost no perceptible degradation in signal quality.

The basis of the RELP technique is recursive encoding and decoding using harmonics of a human voice having a statistically predictable waveform. However, transmission of data communication signals such as a modem output signal and a facsimile signal cannot utilize the above-mentioned harmonic characteristics unlike voice signal transmission. Therefore, the RELP signal compression technique used for voice signal transmission is not suitable for transmission of facsimile signals, data signals, and the like. It is desirable to use a more appropriate technique for signal compression of the data signal.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a wireless telephone system that is transparent to a user regardless of whether a voice signal, a facsimile signal or a data signal is transmitted.

SUMMARY OF THE INVENTION The present invention provides a radiotelephone system including signal encoding means for compressing communication signals to facilitate time-division multiplexing of the communication signals to facilitate increased communication capacity. I do. This system is characterized by identifying types of communication signals such as voice signals, facsimile signals or modem output signals, and employing different compression techniques according to the types of communication signals. An improved method for compressing both facsimile and modem output signals is provided.

Other objects and advantages of the present invention will become apparent from the following description of preferred embodiments.

Embodiment Referring to FIG. 1, a base station 11 and a plurality of subscriber stations 10 of a wireless telephone system network are schematically illustrated. The base station 11 is configured to simultaneously and concurrently communicate with a number of subscriber stations 10 through transmission and reception of radio waves via a selected frequency. This base station forms the interface with the central office 12 (TELCO). In cases where the installation of telephone lines is not physically feasible or provides telephony services at remote locations where cost is high, the subscriber station 10
Can be fixed. The subscriber station 10 can also be constituted by a mobile unit such as a car telephone.

A typical system uses 26 predetermined channels in the 450 MHz band. The number of channels is generally limited due to the allocation of frequency bands for wireless telephony. For example, in the United States, the Federal Communications Commission (FCC) regulates frequency allocation.

Communication between the base station and the mobile station is performed through a pair of frequency channels in a specified frequency band. Preferably, the base station transmits on the lower of the two frequencies of the channel pair and receives on the higher of the two frequencies of each pair. In a system that can use 26 frequency channels, the base station is configured to be able to simultaneously transmit and receive through 13 or more different channel pairs.

In order to increase the transmission capacity of such wireless telephone systems, time division multiplexing of communication signals has been used. For example, U.S. Pat. No. 4,675,863 issued on June 23, 1987 (corresponding to Japanese Patent Application Laid-Open No. 61-218297), the contents of which are incorporated herein by reference, corresponds to four radio frequency channels. Describes a wireless subscriber telephone system that enables two-way transmission of telephone signals up to and including:

That is, each frequency channel pair is divided into four time slots so that the base station can simultaneously communicate with four different subscriber stations via a single channel pair. This effectively quadruples the transmission capacity of this radiotelephone system and allows more than 50
It can be transmitted concurrently via a channel pair wireless system. In practice, one of the 52 time slots defined by the 13 channel pairs is used to perform system overhead functions such as assigning specific channels and time slots for communication with the selected subscriber station. Hold for. This type of time division multiplex radio telephone system can easily provide ordinary telephone services to more than 500 subscriber stations.

One such type of wireless telephone system that is operated using time division multiplexing is the Ultraphone system marketed by International Mobile Machines, Inc., the predecessor of the present applicant.

In order to increase the transmission capacity of a pair of radio channels by time division multiplexing, a standard communication signal between subscribers is compressed to fit within a time slot provided by the time division multiplexing system. For example, the system described in U.S. Pat. No. 4,687,863 compresses a normal 64 kilobit / second digital communication signal into an encoded signal of about 14.6 kilobit / second.

Standard analog communication signals are first converted to digital signals at 64 kilobits per second. This signal is preferably converted to an 8-bit byte signal, thereby producing a digital signal at 8 kilobytes / second.

In the conventional system described above, the communication signal is processed in steps of 22.5 microseconds. The result is a 180 byte sample of the 8 kilobyte / second digital communication signal processed in each successive frame of the radiotelephone system channel. The time division multiplexed frame for each channel pair is configured to accommodate a 14.6 kbit / s coded signal. For a frame with a length of 22.5 milliseconds, it is equivalent to 41 8-bit byte information per frame. Thus, for each frame, the information contained in the 180 byte sample must be encoded within 41 bytes for transmission via one of the time slots of the selected frequency channel. Also,
The 41 encoded bits need to be reconstructed into 180 byte samples per frame without any appreciable distortion or loss of the communication signal content upon reception at the receiving station.

For audio signal transmission, residual excitation linear predictive coding (RELP) is used to convert 180-byte samples into 41-byte coded signals that can be reconstructed into equivalent 180-byte samples without practical problems at the receiving end. The use of a scheme is well known. The RELP coding scheme is also referred to in the above-mentioned U.S. Pat. No. 4,675,863, which is incorporated herein by reference and which is incorporated herein by reference. PCT International Application Publication No. WO dated May 9
86/02726) "RELP Vocoder Implemented In Digital
Signal Processors ".

Although RELP coding has produced satisfactory results for voice signals, it is not suitable for coding facsimile communication signals or modem output signals. This is because these signals do not have the harmonic characteristics and the pitch characteristics unlike the audio signals. That is, the iterative algorithm underlying RELP cannot sufficiently facilitate the encoding of such signals. On the other hand, processor hardware suitable for data compression using RELP is also suitable for the compression method of the present invention. For example, TMS 3202 manufactured by Texas Instruments
The 0-type digital signal processor is suitable for realizing the instantaneous compression processing of the present invention.

In order to improve the transmission of facsimile signals and modem output signals in this type of wireless telephone communication system, improved coding techniques and their implementation have been devised. Referring to FIG. 2, an encoding and decoding scheme according to the invention, in particular for facsimile and data signals, is shown schematically.

Communication signal conversion from a standard analog format to a digital signal of 64 kilobits per second (8 kilobytes per second) for processing at a sample of 180 bytes per frame is performed within the entire wireless communication system for processing any communication signal. Is standardized. When transmitting a facsimile signal or a data signal, a digital signal 20 of 8 kilobytes / second is selectively encoded by an encoding processor 22 and is formatted.
Encode to 21. The coded signal 21 is transmitted through one selected time slot of a plurality of system radio frequency channels 24, each including a number of time slots partitioned by time division multiplexing. The receiving side includes a decoding processor 25. The encoded signal 21 is sent to a decoding processor 25, which processes it to reconstruct a communication signal 26 substantially equivalent to the original signal 20. As described below with reference to FIG. 7, the transmitting side and the receiving side include compression selection processors 27 and 28, respectively, that coordinate the encoding and decoding operations, respectively.
In practice, each of the sender and the receiver includes an encoding processor 22 and a decoding processor 25 to enable two-way simultaneous communication. Further, all of the above-mentioned compression selection processor, encoding processor and decoding processor may be constituted by a single microprocessor such as a Texas Instruments TMS 32020 digital signal processor.

The encoding processor 22 first interpolates the communication signal 20 with the selected coefficient M 'to increase the number of samples. Next, the sample whose number of samples has been increased is divided into an in-phase component 30 and a quadrature component 31 by multiplying the cos (Ωt) signal and the sin (Ωt) signal, respectively. This allows simultaneous, parallel signal processing of the two bit streams of these signals 30 and 31. Ω is the communication signal 20
Is set to the center frequency of the frequency domain display. The cos
By multiplying the (Ωt) signal and the sin (Ωt) signal respectively, the center of the frequency range of the communication signal shifts from the frequency Ω to 0 Hz.

After the mixing or multiplication, the signals 30 and 31 are subjected to a low-pass filter to remove frequency components equal to or higher than a selected frequency in the frequency domain. As a result, a distortion frequency component and a reverberation frequency component centered on an even multiple of Ω are removed.

Samples 32 and 33 passed through low-pass filter respectively
Is decimated by the selected coefficient M in order to reduce the number of samples in preparation for quantization. Next, the decimated signals 34 and 35 corresponding to these in-phase and quadrature components are quantized to a predetermined number of quantization levels by an adaptive PCM encoder, and quantized signal samples 36 and 37, and a quantization gain parameter Yields G. Next, these quantized signal samples and the quantization gain parameter G are formatted together with the unique word 40 from the compression / selection processor 27 to obtain an encoded output signal 21 having a predetermined frame configuration. Separate quantization gain components can be generated for the in-phase component and the quadrature component, respectively. However, to compress a signal of 8 kilobytes per second into a coded signal of 14.6 kilobits per second, a common quantum It is sufficient to provide a generalization gain value.

The frame structure conforms to the format requirements of the time slot allocated to the communication signal in the time division multiplex channel of the radio telephone system. In a preferred system, one frame is composed of 41 bytes. The unique word 40 carries a distinction between the type of signal being transmitted, ie, voice, facsimile, modem output, timing information, and the like. On the receiving side, the received encoded frame is processed according to this information.

The decoding processor 25 on the receiving side separates the quantized signal samples 36 and 37 and the quantized gain parameters G corresponding to the in-phase and quadrature components, respectively. Next, decoding of these quantized signals 36 and 37 is performed according to the quantization gain parameter G to generate communication signal samples 42 and 43 equivalent to the decimated samples 34 and 35 before quantization, respectively. The signal samples 42 and 43, both in-phase and quadrature, are then
To interpolate one after another.

Next, the interpolated in-phase component signal 44 is mixed with cos (Ωt), and the interpolated quadrature signal 45 is mixed with sin (Ωt). The mixed output signals 46 and 47 are respectively subjected to a low-pass filter having the same cutoff frequency as the low-pass filter on the transmission side.
Each of the filtered output signals 48 and 49 is converted to a coefficient M '
Respectively, and add the decimated signals to obtain a communication signal equivalent to the first 8 kilobyte signal 20.
Reconstruct 26. The encoding and decoding operations can be synchronized, but are not required.

In a preferred embodiment adapted to the system described in U.S. Pat. No. 4,675,863, the encoding processor 22 encodes a digitized 8 kilobyte / second communication signal 20.
Encodes the 180-byte sample into a 41-byte frame structure to prepare for time-division multiplexing transmission via the selected wireless telephone system channel.

As shown in FIG. 3, the split band modem communication signal is 1200 Hz.
(Indicates data transmission from outgoing modem) or 2400Hz
This is a relatively narrow band signal centered on (displaying data transmission from the answering modem). Fig. 4 shows the facsimile communication signal.
As shown in the figure, it usually occupies a wider area around 1800 Hz.

In the case of facsimile signal transmission, signal interpolation is performed by a factor of 3.
Preferably, the number of frame samples is increased to 540 samples / frame. Next, the in-phase component and the quadrature component are mixed by the cos (1800t) signal and the sin (1800t) signal, respectively. Low-pass filtering is performed to remove frequency components of 1400 Hz or more. next,
The in-phase and quadrature components 32 and 33 are
Decimate each by 10, reducing the number of samples to 54 for each component of the frame.

Next, the in-phase and quadrature-phase decimated outputs 34 and 35 are quantized to six quantization levels in an adaptive pulse code modulator, and the decimated samples
This produces an 8-bit quantized representation of 36 and 37 and the quantized gain factor G. The encoded output of 54 byte samples / frame for each of the samples 36 and 37 is encoded into 8-bit bytes, each representing one group of three quantized signal samples. Thus, the 54 samples of each component remaining after decimating represent the 54 quantized values of the 54 samples and the 8-bit gain coefficient G, respectively.
Expressed in 18 bytes.

Therefore, as shown in FIG. 5, the entirety of the 36 8-bit bytes D1 to D36 and the 8-bit gain coefficient G are formatted into a 41-byte frame in preparation for data transmission by the wireless communication system. The remainder of the 41 byte frame is filled with a 16-bit unique word U, a 4-bit error check code C and 12 unused bits X.

After transmitting the frame via the wireless communication system, the receiving side reconstructs the frame of the format, and
The bit gain coefficient G is separated into an in-phase component of 18 bytes and a quadrature component of 18 bytes. Next, the above-mentioned quantized signal 36
And 37 are decoded according to the quantized gain factor G, resulting in 54 8-bit samples 42 and 43 each containing information approximately equal to the decimated in-phase component signal 34 and the quadrature component signal 35 at the transmitting end.

Next, the decoded in-phase sample 42 and quadrature-phase sample 43 are interpolated by a coefficient 10 respectively, and the number of samples is reduced to 54
Increase to 0. This interpolated in-phase component signal 44 is cos
(1800t). Similarly, the interpolated quadrature component signal 45 is mixed with sin (1800t). Each of these mixed component signals 46 and 47 is low-pass filtered to 14
Removes frequency components above 00Hz. Each of the low pass filter outputs 48 and 49 is decimated by a factor of 3 to reduce the number of 180-bit samples to 180 per frame. Finally,
The decimated component signals 50 and 51 are summed to produce a communication signal 26 substantially equal to the source signal at the transmitting end, i.e., signal 20 at 8 Kbytes / sec.

In the case of modem output signal transmission, parameters for data compression are slightly different. Similarly to the case of the facsimile signal, the digital signal of 180 bytes per frame is interpolated by a factor of 3 to increase the number of samples to 540 per frame. The mixture of cos (Ωt) and sin (Ωt) with the in-phase and quadrature components is made by setting Ω to 1200 Hz or 2400 Hz depending on whether the signal is from the calling modem or the answering modem. Do.

Low-pass filter processing is performed at a cutoff frequency of 700 Hz. This cutoff frequency is lower than in the case of low-pass filtering of the facsimile signal because the modem output signal is limited to a narrow band centered at about 1200 Hz and 2400 Hz.

After filtering, the signal is decimated by a factor of 20 to reduce the number of samples to 27 samples / frame. Adaptive PC for in-phase and quadrature-phase component samples
Quantize to 32 levels with M encoder. As a result, 27 5-bit quantized representations D1 to D54 of 27 decimated samples for each of the in-phase component and the quadrature component are obtained together with the 8-bit quantization gain coefficient G. This information, a 16-bit unique word U and a 4-bit error detection code C
The 12 free bits X are formatted into a 41-byte frame configuration to prepare for transmission via the selected time slot of the wireless communication frequency channel pair. This frame configuration for data communication is shown in FIG. Note that the unique word U
Is preferably formatted at the same location regardless of the type of signal.

On the receiving side, the received 41-byte frame is separated into 27 5-bit quantization samples and 8-bit quantization gains G corresponding to the in-phase and quadrature components, respectively. The encoded 5-bit quantized samples are decoded with a quantization gain G to produce 27 8-bit signal samples that are informationally equivalent to the decimated in-phase and quadrature signal components, respectively. The decoded samples are interpolated by a factor of 20 and converted to 540 samples / frame.
These samples are mixed with the cos (Ωt) signal and the sin (Ωt) signal, respectively, and the same cutoff frequency (700H
Perform low-pass filtering in z). The filtered output signals 48 and 49 are decimated by a factor of 3 to obtain 180 samples /
Convert to a frame. These two decimated signals 50 and 51 are added to produce an 8 kilobyte / second communication signal 26 that is informationally equivalent to the original signal 20.

The unique word 40 is used to indicate the type of signal to be processed so that the system uses appropriate compression methods and related parameters depending on the type of signal. That is,
The unique word indicates whether the communication signal to be processed is a voice signal, a facsimile signal, a calling station modem signal, a responding station modem, or the like.

It is easy to modify the prior art wireless communication system described in the above-mentioned U.S. Pat. No. 4,675,863 and the like so that the compression method of the present invention can be used. FIG.
Is the codec (CODE CODE) in the prior art system.
FIG. 5 schematically shows a variant in which C) is replaced by a compression selection processor (CSP) 60 and an associated codec (CODEC) 61, 62, 63, 64;

Generally speaking, CSP60 uses CODECs 61 to 64 one by one. Therefore, all of the CODECs 61 to 64 can be provided together with the CSP 60 on a single microchip, and the adopted parameters and the encoding method can be controlled according to the type of the communication signal. That is, CSP60 and CODEC61
-64 all TMS 32030 manufactured by Texas Instruments
It can be incorporated into a digital signal processor to implement both the encoding selection and the codec operation.

In this communication signal compression processing, it is preferable that the wireless communication system uses a desired audio signal compression method such as RELP as a default state. This is because the cancellation tone of the standard echo canceller is generated at 2225 Hz or 2100 Hz at the start of facsimile transmission and modem transmission.

The CSP 60 monitors the communication signal 20 being processed by the voice CODEC 61 and detects the echo canceling device release tone. This is done by checking the first two reflection coefficients of each frame. If these coefficients are within a certain range over a sufficient number of frames, the system switches from speech processing to communication signal processing in the facsimile CODEC 62 and modem CODECs 63-64 with the data compression technique of the present invention.

After switching from the voice processing state, the system first processes the communication signal with the facsimile signal CODEC 62 according to the facsimile signal transmission parameters described above. The communication signal 20 is monitored to detect the presence or absence of facsimile signaling. That is, half of 300 bits / second for the initial handshake operation between the facsimile transceivers.
This is detected by monitoring for the presence of a heavy FSK signal (using 1650 Hz and 1815 Hz).

The detection of this FSK signal is performed as follows. A second order LPC analysis is performed on the signal that produces two reflection coefficients. An average is taken between each reflection coefficient and the corresponding coefficient from the preceding three frames. If the average of these coefficients is inside a predetermined set of boundary values, facsimile transmission is detected and processing is continued according to the facsimile signal CODEC 62, i.e., using the facsimile parameters by the compression technique described above. However, if the FSK signal is not detected within a predetermined time window of, for example, 4.725 seconds, the system switches to the appropriate modem signal CODEC 63 or 64.

When the release tone and the absence of the FSK signal are detected, the calling modem signal CODEC 63 is used. The unique word transmitted in each frame is signal-processed in the receiving subscriber station, and the frame is decoded as voice data, facsimile data and modem transmission or modem response data. When the receiving subscriber station detects the reception of the unique word of the calling modem signal indication, the selection processor 60 of the receiving subscriber station switches to the mode of using the answer modem signal CODEC64.

Processor 60 monitors the transmission direction communication signal 20 for facsimile signaling as well as the energy in the transmission direction. If the energy loss persists for a predetermined period of time, i.e., 67.5 milliseconds for the modem output signal and 22.5 seconds for the facsimile signal, the processor recognizes this and the system returns to the default state and the communication signal is converted to the voice signal CODEC Process as audio signal. Regardless of whether the FSK signal is detected or not, the energy in the transmission direction is continuously monitored to determine the end of facsimile signaling or modem signaling, and the system is reset to voice signaling.

Although the data compression method of the present invention has been described in the above embodiments in relation to a specific wireless communication system, the present invention can be easily applied to another system in which parameters, frequency, transmission medium, frame synchronization and frame configuration are changed. Can be adapted to The parameters used in this data compression method were determined in consideration of compatibility with the system described in the above-mentioned US Pat. No. 4,675,863. By changing the set of these parameters, it is possible to effectively enable the compression of a communication signal that can be encoded into an informationally equivalent data signal by the method described above.

[Brief description of the drawings]

FIG. 1 is a schematic diagram of a wireless communication system that can use an improved data compression process according to the techniques of the present invention. FIG. 2 is a schematic illustration of data compression and decoding of a communication signal according to the present invention. FIG. 3 is an illustration of the frequency domain of a split band modem communication signal. FIG. 4 is an illustration of the frequency domain of a normal facsimile communication signal. FIG. 5 is a schematic diagram of a frame configuration used for transmitting a compressed facsimile signal according to the present invention. FIG. 6 is a schematic diagram illustrating a frame structure used for transmitting a compressed modem output signal according to the present invention. FIG. 7 is a schematic diagram of an implementation of the improved coding system in a wireless communication system.

──────────────────────────────────────────────────続 き Continuing on the front page (72) Scott Day Kurtz, Inventor, New Jersey, 08054, Mount Laurel, West Bluebell Lane 104 (72) Inventor Brian M. McCarthy, United States, Pennsylvania 19444 Emerson Drive, Lafayette Hill, 235 (72) Inventor James M. Cresse, Sagemore Drive, Marlton, 08053, New Jersey, USA 1504 (56) References JP-A-4-150233 (JP, A) (58) Field surveyed (Int. Cl. ⁷ , DB name) H04B 14/00-14/04 H04J 3/00-3/06

Claims (27)

(57) [Claims] A communication signal (20) which is compressed and reconstructed on the receiving side to facilitate transmission over a selected carrier medium (24) is transmitted through said selected carrier medium (24) of a communication system to a variety of sources. A method of communicating between points (10, 11), comprising the step of converting said communication signal (20) into two separate signal components, said converting step comprising: Determining the center frequency Ω of the communication signal (2) so as to generate an in-phase signal component (30).
0) with a signal representing cos (Ωt); mixing the communication signal (20) with a signal representing sin (Ωt) to produce a quadrature signal component (31); The signal component (30) and the quadrature signal component (3
1) are quantized, and these signal components (30) and (31) are quantized to the in-phase and quadrature signals (36,3).
7) and an associated quantization gain parameter (G), thereby producing a compressed encoded signal (21) for transmission over the selected carrier medium (24) of the communication system. Transmitting the compressed coded signal (21) via the selected carrier medium (24) of the communication system and receiving the compressed coded signal (21) at a receiving point; The quantized in-phase and quadrature signals (36, 37) of the compressed encoded signal (21) received via the selected carrier medium (24) of the communication system and the quantized gain parameter (G) And the step of determining an approximate center frequency Ω ′ of the compressed coded signal (21) to be decoded. The step of determining the quantum by the quantization decoding according to the quantization gain parameter (G). In-phase and quadrature signals Zumi (36, 3
7) reconstructing to produce reconstructed in-phase and quadrature signals (42, 43); and mixing said reconstructed in-phase signal (42) with a signal representing cos (Ω't). Mixed reconstructed in-phase signal (4
6) mixing the reconstructed quadrature signal (43) with a signal representing sin (Ω't) to produce a mixed reconstructed in-phase signal (47); Adding the reconstructed in-phase and quadrature signals (46, 47) to produce a decoded decompressed communication signal (26). Each of the components (30, 31) is filtered before quantization, and the signal components (30, 3
Removing the frequency components above a predetermined frequency from 1) to produce filtered in-phase and quadrature components (32, 33); and Said quantized in-phase and quadrature signals (36,37) to produce a compressed encoded signal (21).
And formatting the quantization gain parameter (G). 2. The decimated in-phase and quadrature components (34, 35) of each of the filtered in-phase and quadrature components (32, 33) are decimated by a predetermined coefficient M before quantization. The communication method of claim 1, further comprising the steps of: 3. The in-phase and quadrature-phase signal components (30, 30) are interpolated by a predetermined coefficient M 'and mixed with signals representing cos (Ωt) and sin (Ωt), respectively. The communication method according to claim 2, further comprising the step of: 4. A filter for filtering the mixed reconstructed in-phase and quadrature signals (46, 47) before adding them to remove frequency components of a predetermined frequency or more from the signals (46, 47). 4. The method of claim 3 further comprising the step of producing a processed reconstructed in-phase and quadrature signal.
The communication method described. 5. The communication method according to claim 4, further comprising a step of complementing said reconstructed in-phase and quadrature signals with a predetermined coefficient m before mixing. 6. The communication method according to claim 5, further comprising the step of decimating each of said filtered and reconstructed in-phase and quadrature signals (48, 49) by a predetermined coefficient m 'before adding. 7. A process for determining an approximate center frequency Ω of the communication signal (20), a process for determining a type of the communication signal (20), and a process for processing the communication signal (20) based on the type. Assigning a predetermined value for the communication method. 8. The method according to claim 8, wherein said predetermined value assigned to said center frequency Ω is 1800 Hz (FIG. 4), and said coefficients M and M ′.
Are selected as 10 and 3, respectively, the cutoff frequency of the filter processing is selected as 1400 Hz, and the decimated in-phase and quadrature components (34, 35) are determined that the communication signal (20) is a facsimile signal. In this case, each of the six quantization levels is quantized, or the predetermined value assigned to the center frequency Ω is 1200.
Hz (FIG. 3), and the coefficients M and M 'are 20 and 3
Respectively, and the cutoff frequency of the filtering is selected to be 700 Hz, and the communication signal (20) is obtained by dividing the decimated in-phase and quadrature components (34, 35) with the transmission band signal of the divided band modem signal. If it is determined, each of the quantization values is quantized to 32 quantization levels, or the predetermined value assigned to the center frequency Ω is 2400.
Hz (FIG. 3), and the coefficients M and M 'are 20 and 3
And the cut-off frequency of the filtering is selected to be 700 Hz, and the decimated in-phase and quadrature components (34, 35) are divided into the communication band (20) and the response band signal of the divided band modem signal. If determined, each of the quantization values is quantized to 32 quantization levels, or the predetermined value assigned to the center frequency Ω ′ is 18
00 Hz (FIG. 4), the coefficients m and m 'are selected to be 10 and 3, respectively, and the cut-off frequency of the filter processing is determined by the compressed coded signal (21) corresponding to the communication signal (20) by facsimile. If the signal is determined to be a signal and received at the receiving point, the frequency is selected to be 1400 Hz, or the predetermined value assigned to the center frequency Ω ′ is 12
00 Hz (FIG. 3), the coefficients m and m 'are selected to be 20 and 3, respectively, and the cut-off frequency of the filtering is divided by the compressed coded signal (21) corresponding to the communication signal (20). The predetermined value assigned to the center frequency Ω ′ is determined to be 700 Hz when the signal is determined to be the transmission band signal of the band modem signal and received at the receiving point, or the predetermined value assigned to the center frequency Ω ′ is 24.
00 Hz (FIG. 3), the coefficients m and m 'are selected to be 20 and 3, respectively, and the cut-off frequency of the filtering is divided by the compressed coded signal (21) corresponding to the communication signal (20). The communication method according to claim 7, wherein 700 Hz is selected when it is determined that the signal is a response band signal of a band modem signal and the signal is received at the receiving point. 9. The communication signal according to claim 1, further comprising a step of including a unique word for indicating the type of the signal being transmitted in the compressed coded signal. The communication method according to claim 1, wherein 10. A communication system according to claim 1, wherein said communication signals are different types of signals, and a unique word for indicating the type of the signal being transmitted is formatted at the same time position regardless of the type of the signal. 2. The method according to claim 1, further comprising the step of including in said compressed coded signal.
The communication method described. 11. A communication signal (20) which is compressed and reconstructed on the receiving side to facilitate transmission over a selected carrier medium (24) at various points (10) through said selected carrier medium (24). , 11), comprising a signal compression encoding device (22), the encoding device (22)
Means for converting the communication signal (20) into two individual signal components, and means (27) for determining an approximate center frequency Ω of the communication signal (20); ) To produce the communication signal (2
Means for mixing 0) with a signal representing cos (Ωt), and means for mixing the communication signal (20) with a signal representing sin (Ωt) to produce a quadrature signal component (31). And the in-phase signal component (30) and the quadrature-phase signal component (3
1) are quantized, and these signal components (30) and (31) are quantized to the in-phase and quadrature signals (36,3).
7) and means for encoding the associated quantization gain parameter (G), thereby producing a compressed encoded signal (21) for transmission over the selected carrier medium (24) of said communication system. In the system, the encoding device (22)
However, each of the in-phase and quadrature-phase signal components (30, 31) is subjected to a filtering process before quantization so that the signal components (30, 3)
1) means for removing frequency components above a predetermined frequency from 1) to generate filtered in-phase and quadrature-phase components (32, 33), and said device for transmission via said selected carrier medium (24) of said communication system. Said quantized in-phase and quadrature signals (36,37) to produce a compressed encoded signal (21).
And formatting means for multiplexing the quantization gain parameter (G). 12. The apparatus according to claim 11, wherein said encoding device includes means for decimating each of said filtered in-phase and quadrature components by a predetermined coefficient M prior to quantization. 12. The communication system according to claim 11, wherein 13. The coding apparatus (22) interpolates the communication signal (20) by a predetermined coefficient M ',
13. The communication system according to claim 12, further comprising means for producing said in-phase and quadrature-phase signal components (30,31) by mixing with signals representing t) and sin (Ωt), respectively. 14. A means (27) for determining an approximate center frequency Ω of the communication signal (20) determines the type of the communication signal (20), and based on the determination, determines the center frequency Ω, coefficient M and 14. The communication system according to claim 13, wherein a predetermined value is assigned to M 'and a cutoff frequency of said filter. 15. The quantized in-phase and quadrature signals (36, 37) and the quantization gain of the compressed encoded signal (21) received via the selected carrier medium (24) of the communication system. Format decoding means for separating the parameter (G), means for determining the approximate center frequency Ω 'of the compressed signal (21) to be decoded, and the format decoding means according to the quantization gain parameter (G). Quantizing decoding means for reconstructing the in-phase and quadrature signals (36, 37) to generate reconstructed in-phase and quadrature signals (42, 43); cos (Ω′t) and a mixed reconstructed in-phase signal (4
Mixing means for producing 6); mixing means for mixing said reconstructed quadrature signal (43) with a signal representing sin (Ω't) to produce a mixed reconstructed quadrature signal (47); Summing means for adding the mixed reconstructed in-phase and quadrature signals (46.47) to produce a decoded decompressed communication signal (26). Communication system according to one or more of the claims 11 to 14, characterized in that: 16. A first apparatus having the compression encoding device (22).
Communication station (10) and said signal decompression decoding device (25)
16. The communication system according to claim 15, further comprising a second communication station (11) having a first communication station (10) and receiving the compressed encoded signal (21) from the first communication station (10). 17. A communication system further comprising a plurality of communication stations each having at least one of said signal compression encoding device (22) and at least one of said signal decompression decoding device (25). 16. The communication system according to claim 15, wherein: 18. A method according to claim 18, wherein at least one of said communication signals comprises a plurality of said signal compression encoding devices for encoding and decoding a selected type of signal and a plurality of said signal decompression decoding. Encoding means (25); means for selecting one of the plurality of encoding apparatuses (22) for signal processing according to the type of the communication signal (20) to be transmitted; 18. The communication according to claim 17, further comprising: means for selecting one of the plurality of decoding devices (25) for signal processing according to the type of the coded signal (21). system. 19. A determination means (28) for determining the approximate center frequency Ω 'of the compressed coded signal (21) determines the type of the communication signal and sets a predetermined value based on the type of the communication signal. 19. The communication system according to claim 15, further comprising assigning a center frequency Ω ′. 20. The in-phase and quadrature-phase signal components (3
0, 31), each of the signal compression encoding devices (22) further includes means for filtering each of these signal components before quantization so as to remove frequency components of a predetermined frequency or more from these signal components; Each of the mixed reconstructed in-phase and quadrature signals (46, 47) is filtered prior to addition so as to remove frequency components above a predetermined frequency from these signals and filtered. 19. The signal decompression decoding device (25) further comprising filter means for respectively producing a phase and a quadrature signal (48, 49). Communication system. 21. Each of said filtered in-phase and quadrature components (32, 33) is a predetermined coefficient M
Wherein each of the signal compression encoding devices (22) further comprises means for decimating to produce decimated in-phase and quadrature components (34, 35), respectively; and that the reconstructed in-phase and quadrature signals ( 21. The communication system according to claim 20, wherein each of the signal decompression decoding devices (25) further includes means for interpolating (42, 43) with a predetermined coefficient m before addition. 22. The communication signal (20) is converted to the cos (Ωt)
Means for interpolating with the predetermined coefficient M 'to produce said in-phase and quadrature-phase signal components (30, 31) before mixing with the signals respectively representing sin and .OMEGA. Further comprising: means for decimating each of said filtered reconstructed in-phase and quadrature signals (48,49) by a predetermined coefficient m 'prior to addition, said signal decompression decoding Equipment (2
22. The communication system according to claim 21, wherein each of 5) further includes: 23. The encoder (22) includes means (27) for determining the type of the communication signal (20), and the center frequency Ω and the coefficient M are determined based on the type of the signal (20).
And M ′, assigning a selected numerical value to each of the encoding devices (22) to the cutoff frequency and the number of quantization levels of the filtering process, and the decoding device (25) Means (28) for judging the type, and based on the type of the signal (20), a selected numerical value is set as the center frequency Ω ', the coefficients m and m', the cutoff frequency of the filtering and the number of quantization levels. 23. The communication system according to claim 22, wherein allocation is performed for each of the decoding devices (25). 24. The predetermined value assigned to the center frequency Ω is 1800 Hz (FIG. 4), the coefficients M and M ′ are selected to be 10 and 3, respectively, and the cutoff frequency of the filtering is 1400 Hz. And quantizing the decimated in-phase and quadrature components (34, 35) to six quantization levels respectively when the communication signal (20) is determined to be a facsimile signal; or The predetermined value assigned to the frequency Ω is 1200
Hz (FIG. 3), and the coefficients M and M 'are 20 and 3
Respectively, and the cutoff frequency of the filtering is selected to be 700 Hz, and the communication signal (20) is obtained by dividing the decimated in-phase and quadrature components (34, 35) with the transmission band signal of the divided band modem signal. If it is determined, each of the quantization values is quantized to 32 quantization levels, or the predetermined value assigned to the center frequency Ω is 2400.
Hz (FIG. 3), and the coefficients M and M 'are 20 and 3
And the cut-off frequency of the filtering is selected to be 700 Hz, and the decimated in-phase and quadrature components (34, 35) are divided into the communication band (20) and the response band signal of the divided band modem signal. If determined, each of the quantization values is quantized to 32 quantization levels, or the predetermined value assigned to the center frequency Ω ′ is 18
00 Hz (FIG. 4), the coefficients m and m 'are selected to be 10 and 3, respectively, and the cut-off frequency of the filter processing is determined by the compressed coded signal (21) corresponding to the communication signal (20) by facsimile. If the signal is determined to be a signal and received at the receiving point, the frequency is selected to be 1400 Hz, or the predetermined value assigned to the center frequency Ω ′ is 12
00 Hz (FIG. 3), the coefficients m and m 'are selected to be 20 and 3, respectively, and the cut-off frequency of the filtering is divided by the compressed coded signal (21) corresponding to the communication signal (20). The predetermined value assigned to the center frequency Ω ′ is determined to be 700 Hz when the signal is determined to be the transmission band signal of the band modem signal and received at the receiving point, or the predetermined value assigned to the center frequency Ω ′ is 24.
00 Hz (FIG. 3), the coefficients m and m 'are selected to be 20 and 3, respectively, and the cut-off frequency of the filtering is divided by the compressed coded signal (21) corresponding to the communication signal (20). 24. The communication system according to claim 23, wherein it is determined that the signal is a response band signal of a band modem signal, and when the signal is received at the receiving point, 700 Hz is selected. 25. The communication system according to claim 24, wherein the means (27) for determining the approximate center frequency Ω of the communication signal (20) determines a type of the communication signal (20). 26. The communication system according to claim 26, wherein said communication signals are of different types, and further comprising means for inserting a unique word for indicating the type of the signal being transmitted into said compressed coded signal. 12. The communication system according to claim 11, wherein: 27. A communication system according to claim 27, wherein said communication signals are of different types, and a unique word for indicating the type of signal being transmitted is formatted at the same time position regardless of the type of said signal. The apparatus further comprises means for inserting the compressed encoded signal into the compressed encoded signal.
11. The communication system according to 11.