JPH06216834A - Telephone signal transmission/reception method for mobile object communication - Google Patents

Abstract

PURPOSE:To provide a mobile object communication method which can reduce the quantity of transmission/reception information on the telephone signals. CONSTITUTION:Each ratio base station 30 covers plural zones, and each mobile radio equipment 100 uses a radio channel where the telephone signals exist to communicate with the station 30. Then a gateway exchange 20 secures the communication between the station 3 and the equipment 100. Both the station 30 and the equipment 100 store each sample value of many time piece signals obtained by fractionizing the telephone signals at the first stage of calling and then compare each sample value of the time piece signals with each stored one in a busy mode. Then only the identification of the coincident time piece signals is transmitted and the telephone signal is reproduced at the receiver side by the identification signal. Thus the quantity of information to be transmitted is extremely reduced and therefore can be used for the wire transmission. Furthermore the mobile radio transmission power can be reduced.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a telephone signal transmitting / receiving method effective in mobile communication such as a system for communicating a signal having a band such as a telephone, and particularly in communication requiring effective utilization of frequency. More specifically, a certain wireless channel is provided, and one of the many mobile wireless devices in the service area is used to set a wireless channel and communicate with the opposite wireless base station. Among, when the other mobile wireless device starts communication with another wireless base station using the same wireless channel, due to the effective use of frequency or radio wave propagation characteristics, respectively, with the mobile wireless device during communication, To provide a method that eliminates the adverse effect on the communication with the wireless base station, and at the same time dramatically reduces the amount of information to be transmitted and improves the effective utilization of the frequency by decreasing the transmission output. To do.

[0002]

2. Description of the Related Art A system suitable for application of the present invention in mobile communication using voice to which a small zone method is applied, or a method for improving effective utilization of frequency by reducing communication information and its related technology. Are described in the following documents.

Reference 1. Ito "Study on mobile phone systems-Proposal of time division time compression FM modulation system-" IEICE Technical Report R
CS89-11 July 1989

Reference 2. Ito "Study on mobile phone system-Theoretical study on time-division time compression FM modulation system"
RCS89-39 October 1989

Reference 3. Ito "Time division time compression multiplexing (T
CM) About the efficient transmission method of telephone signal "IEICE Technical Report SST92-24 June 1992

Reference 4. "Hearing and voice" published by The Institute of Electronics and Communication Engineers, August 1968, pages 298-299 and 416-4
20 pages

Reference 5. Supervised by Amari “Voice / Hearing and Neural Network Model”, published by Ohmsha Co., Ltd. August 1-16-1
8 pages

Reference 6. Nakata "Speech Synthesis and Recognition" Comprehensive Electronic Publishing Company, page 102, July 1980

That is, in Reference 1, the transmission signal (baseband signal) is divided into predetermined time interval units and stored in the memory circuit, and when reading this, it is n times faster than the speed of storing in the memory circuit. Built in mobile radios and radio base stations to read at a predetermined time slot, angle-modulate or amplitude-modulate the carrier wave with the signal accommodated in this time slot, and to transmit and receive intermittently in time , A radio receiving circuit having a receiving mixer which communicates with each other, a radio transmitting circuit having a transmitting mixer, a synthesizer applied to the receiving mixer of the radio receiving circuit, and a synthesizer applying to the transmitting mixer of the radio transmitting circuit A circuit is provided, and the output of the synthesizer applied to each is interrupted, and this interrupted state is transmitted and received. For the wireless base station that communicates with each other for the opposite purpose, the same method as the above is used to synchronize the intermittent transmission and reception with that of the mobile wireless device, and the receiving side extracts only the signal accommodated in the predetermined time slot. Therefore, the signal received by demodulating by receiving and opening the wireless receiving circuit is stored in the memory circuit, and at the time of reading this, by reading at a low speed of 1 / n of the speed stored in this memory circuit, A system example in which a system capable of reproducing a baseband signal which is the transmitted original signal is constructed has been reported.

Next, in Document 2, the TCM as described above is used.
(Time-division time compression multiplexing) -Adjacent channel interference and co-channel interference, which are problems when the FM system is applied to a small zone, are being studied, and the system can be realized by selecting system parameters appropriately. The possibility is shown.

Further, Document 3 proposes, as an efficient transmission method of a TCM telephone signal, a method of paying attention to the redundancy of the telephone signal and reducing a part of the signal to transmit.

[0012] Further, in Reference 4, the nature of the voice uttered by a person is described in detail.

FIGS. 25A to 25E show waveforms of vowel sounds a, d, a, o, and u of a voice. The first formant of the vowel a in FIG.
The formant is at 2400-3000 Hz. (B)
The first formant of the vowel D is 350 to 550 Hz, the second formant is 1500 to 2000 Hz, and the third formant is 2500 to 3000 Hz.
~ 1400Hz, 2700-3100Hz, 420-500Hz, 760-1000H in the case of the vowel sound (d)
z, 1300-2000 Hz, 300-480 Hz, 1000-1400 Hz, 200 in the case of the vowel U of (e).
0 to 3000 Hz.

The frequency component that characterizes such a vowel is called a formant. However, the frequency is slightly different for each person. On the other hand, the consonant appears in a transient manner only for a short time, its frequency component is relatively high, and its energy is extremely small. Figure 26 shows "CH
In the sound waveform of O "(cho), the part where the amplitude is large and conspicuous is the part of the vowel O following CH, and the essential consonant is the fine waveform part that appears at the beginning.

The smallest pronunciational unit that constitutes a word is called a syllable. Most Japanese syllables take the form of (consonant c) + (vowel v), but in foreign languages the structure is complex and typical syllables are (consonant c) + (vowel v) + (consonant c) Such a configuration is adopted. The duration of the syllable is 100-
At 300 ms, the average is 1/8 second and the inter-syllable pause time is 100-200 ms.

Next, the relationship between the energy of voice and the amount of information will be described with reference to FIG. Even if the peak appearing in the waveform of the voice is compressed and removed, the amount of transmitted information does not change much. On the other hand, if the vicinity of 0 of the voice waveform is removed, the amount of information carried will drastically decrease even if a large amplitude area is left as it is. FIG. 27 shows the results of this experiment, in which the horizontal axis represents the amount of removal of the peak or zero portion of the speech waveform, and the vertical axis represents the reduction in word intelligibility caused by such removal.

FIG. 28 clearly shows the removal conditions when the experimental data of FIG. 27 is removed near zero and when the peak is removed. Here, the peak removal amount = 20 log (p ₁ / p ₀ ) (dB) 0 vicinity removal amount = 20 log (p ₂ / p ₀ ) (dB).

Further, Document 4 describes the state of the temporal change of the formant frequency of the audio signal. The result is that the bandwidth of the frequency spectrum of the formant frequency variation is 7.2, 6.7 and 5. for the first formant, the second formant and the third formant, respectively.
A value of 3 Hz is sought.

Reference 5 describes the types of phonemes forming Japan and their durations, and FIG. 29 is shown.

Further, in Reference 5, the acoustic properties of consonants are shown in FIG.
It is described using 0. That is, nasal consonants and fricatives, which have relatively stationary properties, are characterized by a formant frequency similar to that of a vowel and an anti-resonant frequency (anti-formant frequency) generated by a vocal tract branch pipe. By the way, the antiformant frequency of the nasal consonant is [m],
500-1500Hz, [n] 2000-3000
Hz and [ng] are around 3000 Hz. In particular, the fricative sound has energy concentrated in a high frequency region, and its detection is relatively easy. However, in general, for example, a plosive sound and a nasal consonant sound can be distinguished by the transition direction and speed of the formant frequency near the boundary between the consonant sound and the vowel sound, as shown in FIG.

FIG. 30 shows consonant articulatory positions (lips, gums, palate) and articulation methods (voiced plosives, silent plosives, nasal sounds).
Correspondence between the form and frequency spectrum is shown. In the same figure, ba, da, and ga as voiced plosives, pa, ta, and ka as silent plosives, and ma, na, as nasal sounds.
qa is shown respectively. In the frequency spectrum of each consonant, two thick lines substantially parallel to the horizontal axis indicate the first formant with the lower frequency and the second formant with the higher frequency. The characteristic of the figure is that the frequency of the formant changes differently with time for each consonant. Further, it is considered that this amount of change in frequency takes the value shown in Reference 4.

Document 6 also describes in detail the synthesis and recognition of speech, but in particular, it is described that the number of basic phoneme levels constituting Japanese is 1500 or more.

[0023]

DISCLOSURE OF THE INVENTION Problems to be Solved by the Invention
In the system construction example shown in FIG. 1, a general description of a mobile communication system using a TCM signal is given, and although the system can be constructed by this, the signal itself is transmitted instead of being transmitted. From among a large number of time-piece signals that have been subdivided and stored, select ones that have exactly the same or approximate signal characteristics but almost the same signal characteristics (waveform, signal level, etc.) However, there is no description regarding the method of the present invention in which only the identification information possessed by the method is transmitted. Further, although the method proposed in Document 3 certainly enables an efficient transmission method of the telephone signal, there is a great difference in the degree as compared with the method according to the present invention.

Further, although the characteristics of voices emitted by humans are described in detail in Documents 4 and 5, a method of communicating by making the telephone signal a signal divided in time and effectively utilizing the property. However, there was an unsolved problem that it was not disclosed.

Although the document 6 describes the voice synthesis and recognition in detail, there is no suggestion about transmitting and receiving the telephone signal as a signal divided in time with the identification information. There were issues to be solved.

[0026]

Both communication devices, which communicate using telephone signals, own a large number of time piece signals which are stored in advance in a state where the telephone signals are subdivided in time. And the identification information of those time piece signals is shared by both communicating parties, when the telephone signal is transmitted, the signals constituting the communication are represented by the set of time piece signals already owned. At this time, instead of transmitting the telephone signal, it was decided to transmit only the above-mentioned identification information. As a result, mobile communication has become possible with a smaller amount of information than that of the telephone signal.

Furthermore, both or one of the mobile radio and the gateway switch subdivides a telephone signal in advance in time,
When identification information is added to each of the signals, and the signal on the transmission side can be restored by a large number of time-series signals, one of the mobile radio device, the gateway switch, or the radio base station, or these When some of the time piece signals that are held are transmitted to the other party of the call including the identification information and stored in the storage means that the other party has, instead of transmitting the telephone signal in the call, I decided to send only the identification information. Also, when a telephone signal is subdivided in time and transmitted, in both transmission and reception, this time piece signal is time-compressed or time-expanded to create a new time piece signal. , A new time piece signal group is created by various methods, such as by attaching identification information to each and storing it.

[0028]

Function: Telephone signals are transmitted at short time intervals (for example, 10 ms).
When separated by, a part of the signal waveform is separated. If a large number of parts of the waveform of this signal are prepared as time piece signals and are stored in a memory circuit, and the number of time piece signals (hereinafter referred to as the number of samples) exceeds a certain level, the telephone signal is changed. It may be assumed that the constituent signals can all be assembled from the stored time piece signals.

Therefore, if these signals are attached with identification information and stored, and if these are shared by the communicating parties, telephone communication between the parties does not need to transmit or receive actual telephone signals, but the above-mentioned identification is performed. The original telephone signal can be reproduced by transmitting only the information and assembling the time piece signals by arranging them temporally based on the identification information sent by the receiver side. As a result, the mobile telephone communication having the same communication quality as the conventional one has become possible due to the small amount of information compared to the amount of information contained in the telephone signal.

[0030]

BEST MODE FOR CARRYING OUT THE INVENTION The present invention is applicable to a time-division telephone signal transmission system in general, but an example of application to a time division time compression multiplex (TCM) telephone system will be described below.

1, 2 and 3 show the system configuration for explaining the basic operation example of the present invention.

In FIG. 1, reference numeral 10 is a general telephone network, and 20 is a gateway switch for switching and connecting the telephone network 10 and a wireless system. Reference numeral 30 denotes a wireless base station, which is an interface with the gateway switch 20, a circuit for converting a signal speed, a circuit for allocating and selecting a time slot,
There is a control part etc., besides setting and releasing the wireless line,
It has a wireless transmission / reception circuit for exchanging wireless signals with the mobile wireless device 100 (100-1 to 100-n). here,
Between the gateway exchange 20 and the wireless base station 30, there are transmission lines for transmitting communication signals 22-1 to 22-n including the communication signals of the communication channels CH1 to CHn and control signals.

FIG. 2 shows a circuit configuration of the mobile radio 100 which communicates with the radio base station 30. A received signal such as a control signal or a call signal received by the antenna unit is received by the wireless receiving circuit 13 including the receiving mixer 136 and the receiving unit 137.
5, and the output communication signal is the speed restoration circuit 1
38, the control unit 140, and the clock regenerator 141. In the clock regenerator 141, a clock is regenerated from the received signal and the clock is regenerated by the speed conversion circuit 131, the speed restoration circuit 138, the control unit 140, and the timing generator 14.
2 is applied. The control signal transmitted from the wireless base station 30 is transmitted from the output side of the receiving unit 137 to the control unit 14.
Sent to 0.

The speed decompression circuit 138 receives the compressed and delimited time piece signal or the identification information of the time piece signal (the signal format is often a digital format) in the received signal,
In the case of the former time piece signal, restore the speed of the telephone signal,
It is input to the telephone unit 101 as a continuous signal.

In the latter case of the identification information of the time piece signal, a signal is sent to the identification information transmission / reception circuit included in the speed conversion circuit 131 and received from the time piece signal group storage / comparison circuit according to the instruction of the control unit 140. The time piece signal corresponding to the identified information is read and input to the telephone unit 101 as a temporally continuous signal. The telephone unit 101 receives the above-mentioned telephone signal and outputs it from the handset. The process of receiving the identification information of the time piece signal will be described in detail later.

The telephone signal output from the telephone unit 101 is input to the speed conversion circuit 131. Here, after the telephone signal is cut at a constant time interval, the speed thereof is increased (compressed), and the transmission mixer 133 and the transmission unit 134 are provided.
And is applied to the wireless transmission circuit 132 including. However, if the characteristics of the signal are examined and it is judged that it is sufficient to send only the identification information, the speeded up signal will not be sent, and the process in this case will be described in detail later.

The output of the modulator included in the transmission unit 134 is converted into a predetermined radio frequency in the transmission mixer 133,
The signal is transmitted from the antenna unit and received by the wireless base station 30. In order to transmit a radio signal to the radio base station 30 using the time slot permitted to be used by the mobile radio device 100, the timing information from the timing generator 142 shown in FIG. Must have been obtained by

The timing generator 142 supplies the necessary timing to the transmission / reception interrupt controller 123, the speed conversion circuit 131 and the speed restoration circuit 138 by the clock from the clock regenerator 141 and the control signal from the control unit 140. There is.

The mobile radio 100 further includes synthesizers 121-1 and 121-2 and a changeover switch 122.
-1, 122-2 and changeover switches 122-1 and 122
-2 includes a transmission / reception gating controller 123 and a timing generator 142 which generate signals for switching each of -2, and the synthesizers 121-1 and 121-2, the transmission / reception gating controller 123, and the timing generator 142 are control units. It is controlled by 140. Each synthesizer 121
A reference frequency is supplied to the -1, 121-2 from the reference crystal oscillator 120.

A radio base station 30 is shown in FIG.
N-channel communication signal 22-1 with the gateway switch 20
22 to 22-n are connected to the signal processing unit 31 that forms an interface on the transmission path. Therefore, the communication signals 22-1 to 22-n sent from the gateway switch 20 are transmitted to the wireless base station 30.
Is input to the signal processing unit 31. The signal processing unit 31 includes an amplifier for compensating for transmission loss,
So-called 2-line to 4-line conversion is performed. That is, the input signal and the output signal are mixed and separated, and the input signal from the gateway switch 20 is sent to the signal speed conversion circuit group 51. Further, the output signal from the signal speed restoration circuit group 38 uses the same transmission line as the input signal in the signal processing unit 31, and the gateway exchange 20
Sent to. The input signal from the gateway switch 20 is input to the signal speed conversion circuit group 51 including many signal speed conversion circuits 51-1 to 51-n, and is divided at a predetermined time interval to convert the speed (pitch). The received signal is sent to the signal allocation circuit group 52. Here, if the characteristics of the signal are examined and it is determined that it is sufficient to send only the identification information, the output of the signal speed conversion circuit group 51 will not be sent, but the functions of these signal speed conversion circuit group 51 have already been described. It is similar to the speed conversion circuit 131 of the mobile wireless device 100 described above, except that a large number of telephone signals are simultaneously processed. The process when the output of the signal speed conversion circuit group 51 is not transmitted will be described in detail later.

As for the signal transmitted from the radio base station 30 to the gateway switch 20, the output of the radio reception circuit 35 is input to the signal speed restoration circuit group 38 via the signal selection circuit group 39, and the speed (pitch). The converted telephone signal is restored and input to the signal processing unit 31. In this case, the received signal may be only the identification information instead of the telephone signal, and the process in this case will be described in detail later.

The control of the radio receiving circuit 35 or the output of the call signal is performed by the signal selection circuit group 3 including the signal selection circuits 39-1 to 39-n for selecting the signal for each time slot.
9 is input, and the call signal is separated corresponding to each call channel CH1 to CHn. This output is restored by the signal speed restoration circuit group 38 including the signal speed restoration circuits 38-1 to 38-n provided for each channel to restore the signal speed (pitch), restore the telephone signal, and then perform the signal processing. After being input to the unit 31 and undergoing 4-line to 2-line conversion, this output is sent to the gateway exchange 20 as communication signals 22-1 to 22-n.

Next, the function of the signal speed conversion circuit group 51 (FIG. 3) will be described. An input signal such as a voice signal or a control signal, which is a one-sided telephone signal divided into a fixed time length, is stored in a storage circuit, and the speed is changed at the time of reading this, and the reading is performed at a speed 15 times faster than the case of storing it. This makes it possible to compress the time length of the signal. The principle of time compression of the signal speed conversion circuit group 51 is the same as the case of reproducing a voice recorded by a tape recorder at high speed. In practice, for example, a CCD (Charge Coupled Device) is used.
), BBD (Bucket Brigade Device) can be used, and a memory used in a television receiver or a tape recorder for compressing or expanding the time axis of conversation can be used. These are publicly known, and recently, it is possible to use a large-capacity semiconductor memory element in the form of an LSI, which has made remarkable progress in recent years.

CCD exemplified by the signal speed conversion circuit group 51
The above circuit using BBD or BBD can be used as it is to realize the time extension function of the signal speed restoration circuit group 38. In this case, the clock from the clock generator 41 and the control signal from the control unit 40 are used. This can be realized by receiving the timing signal from the timing generator 42 that generates the timing and setting the reading speed to be slower than the writing speed.

The control or voice signal output from the gateway switch 20 via the signal processing unit 31 is input to the signal speed conversion circuit group 51, and the speed (pitch) conversion processing is performed, and then the time slot. It is applied to the signal allocation circuit group 52 which allocates signals separately.

The signal allocating circuit group 52 is a buffer memory circuit, which stores the high-speed signals for one segment outputted from the signal speed converting circuit group 51, and outputs from the timing generating circuit 42 given by the instruction of the control section 40. The signal in the buffer memory is read out with the timing information of and is sent to the wireless transmission circuit 32. As a result, when the communication signals are viewed as channels, they are arranged in series without overlap in time series, and when all the control signals or call signals described later are implemented, it is as if they were continuous signal waves. become.

The above signals are sent to the wireless transmission circuit 32. The state of this compressed signal will be described with reference to FIG.

The output signals of the signal speed conversion circuit group 51 are input to the signal allocation circuit 52 and given time slots in a predetermined order. SD in Figure 4 (a)
, SD2, ..., SDn indicate that the speed-converted communication signals are assigned to each time slot. Here, in one time slot, a sync signal and a time-compressed speech signal or / and a control signal are accommodated as shown in the figure. When the call signal is not installed, an empty slot signal is added to the call signal portion only with the synchronization signal, or in some systems, no signal including the carrier wave is transmitted. In this way, as shown in FIG. 4A, in the wireless transmission circuit 32, 1 is set in the time slots SD1 to SDn.
The signals that make up the frame will be applied to the modulator circuit. The multiplex signals time-serialized to be transmitted are angle-modulated in the wireless transmission circuit 32, and then transmitted to the space from the antenna section.

Regarding the transmission of the control signal between the radio base station 30 and the mobile radio 100 prior to the call in the incoming and outgoing call of the telephone, either in the band of the telephone signal or out of the band is used. It is possible. FIG. 5 shows these frequency relationships. That is, in the figure (a), an example of the out-of-band signal is shown, and as shown in the figure, the low frequency side (250 Hz) or the high frequency side (3850 Hz) can be used. This signal is used, for example, when it is desired to send a control signal during a call. FIG. 5B shows an example of the in-band signal, which is used at the time of making and receiving calls.

Although the above-mentioned examples are all for tone signals, it is possible to transmit various kinds of signals at high speed by increasing the number of tone signals or by modulating the tones to form subcarrier signals. Becomes

Although the above is the case of the analog signal, when the digital data signal is used as the control signal, it is also possible to digitally encode the voice signal and time-division multiplex both of them for transmission. The circuit configuration in this case is shown in FIG. FIG. 6 shows an example in which the voice signal is digitized by the digital encoding circuit 91, the data signal and the data signal are multiplex-converted by the multiplex conversion circuit 92, and applied to the modulation circuit included in the wireless transmission circuit 32. Then, if the opposite receiver receives the signal and the demodulation circuit performs the reverse operation to that shown in FIG. 6, the audio signal and the control signal can be separately taken out.

On the other hand, the signal sent from the mobile radio 100 is received by the antenna section of the radio base station 30 and input to the radio receiving circuit 35. FIG. 4B schematically shows the upstream input signal. That is, the time slots SU1, SU2, ..., SUn represent transmission signals addressed to the radio base station 30 from the mobile radios 100-1, 100-2, ..., 100-n. The contents of each of the time slots SU1, SU2, ..., SUn are shown in detail in the lower left of FIG. 4 (b) and consist of a call signal and / or a control signal. However, the synchronization signal can be omitted depending on the case where the distance between the mobile wireless device 100 and the wireless base station 30 is small or the signal speed.

The control signal of the input signals arriving at the radio base station 30 is immediately added from the radio reception circuit 35 to the control section 40. However, depending on the size of the speed conversion rate, it is also possible to add the signal from the output of the signal speed restoration circuit group 38 to the control unit 40 after performing the same processing as the call signal. The call signal is applied to the signal selection circuit 39. The signal selection circuit group 39 includes a control unit 4
According to the instruction of the control signal from 0, a timing signal from a timing generation circuit 42 that generates a predetermined timing is applied, and a synchronization signal, a call signal or a control signal is separately output for each time slot SU1 to SUn.

Each of these signals is sent to the signal speed restoration circuit 38.
Is input to. This circuit has a function of performing inverse conversion of the speed conversion circuit 131 (FIG. 2) in the mobile radio device 100 on the transmission side, whereby the original signal is faithfully reproduced and transmitted to the gateway exchange 20. Become.

Hereinafter, an aspect of transmitting the signal space according to the present invention will be described with reference to a required transmission band and a relationship between the required transmission band and adjacent radio channels.

As shown in FIG. 3, the control signal from the control unit 40 is applied to the radio transmission circuit 32 in parallel with the output of the signal allocation circuit 52. However, depending on the size of the speed conversion rate, after the same processing as the call signal is performed, the signal allocation circuit 5
It is also possible to add from the output of 2 to the wireless transmission circuit 32.

Next, also in the mobile radio 100, as shown in FIG.
As shown in FIG. 5, the circuit configuration is required when the communication path among the functions of the wireless base station 30 is one channel.

Original signal, eg voice signal (0.3 kHz-
The frequency distribution on the output side when 3.0 kHz) passes through the signal speed conversion circuit group 51 (FIG. 3) is as shown in FIG. That is, as described above, the speed of the audio signal is 1
If converted to 5 times, the frequency distribution of the signal is expanded to 4.5 kHz to 45 kHz as shown in FIG. Although the frequency distribution of the signal is expanded here, it should be noted that the shape of the waveform simply undergoes a similarity transformation in which the frequency axis is stretched, and the waveform itself does not change. This is necessary when determining the value of the multiple load gain.

Now, FIG. 7 shows a case where the control signal is simultaneously transmitted using the lower frequency band of the voice signal which is a time piece signal. Control signals (0.2 to 4.0 kHz) and call signal CH1 (4.
(Represented as SD1 at 5 to 45 kHz) is contained in a time slot, eg SD1. The same applies to audio signals accommodated in the other time slots SD2 to SDn.

That is, the time slot SDi (i =
2, 3, ..., N) is a control signal (0.2 to 4.0 kH)
z) and the communication signal CHi (4.5 to 45 kHz). However, the signals in each time slot are arranged in time series, and the signals in a plurality of time slots are not added to the wireless transmission circuit 32 at the same time.

The above control signal is not implemented when a time slot for the control signal is provided at the beginning of the frame, and when the lower frequency band is used for another signal, the communication signal is not used. Near the frequency band of (4.1 to 4,4
KHz or 46-46.5 kHz).

Furthermore, when the compressed time piece signal is not transmitted, only the identification information of the time piece signal is transmitted, but this identification information is normally in the digital signal format and is transmitted in the frame. Is a specific time slot for carrying a time piece signal, shared with the control signal provided in the first time slot in the frame, and commonly used for identification information of other telephone signals in the frame. Time slot, eg, SUn (S
Dn) is used exclusively. Etc. are possible. However, in any case, the maximum frequency of the signal does not exceed 45 kHz in the above example.

When these call signals are added to the angle modulation section included in the radio transmission circuit 32 together with the control signal, a required transmission band of at least f _C ± 45 kHz is required. However, f _C is a radio carrier frequency. If there are a plurality of wireless channels given to the system, the speedup of signals by the signal speed conversion circuit group 51 is limited to a certain value due to the limitation of these frequency intervals. The frequency interval of a plurality of wireless channels is f
Let _rep be the maximum signal speed due to the speedup of the above-mentioned audio signal be f _H, and the following inequality must be established between them. f _rep > 2f _H On the other hand, in the case of a digital signal, the voice is usually digitized at a speed of about 64 kb / s. Therefore, it is necessary to read the scale of the horizontal axis in FIG. However, the relation of the above equation holds in this case as well.

In addition, from the mobile radio 100 to the radio base station 3
The control signal input to 0 is input to the wireless reception circuit 35, part of its output is input to the control unit 40, and the other is sent to the signal speed restoration circuit group 38 via the signal selection circuit 39. . The latter control signal undergoes speed conversion (conversion to a low speed signal) completely opposite to that at the time of transmission, and then the general telephone network 10
The signal speed is the same as that used in the above, and is sent to the gateway exchange 20 via the signal processing unit 31.

Next, the operation of making and receiving calls in the basic operation of the system according to the present invention will be described by taking the case of a voice signal as an example. However, it is assumed that all transmitted telephone signals are transmitted in the form of time piece signals, and are not sent in the form of identification information of time piece signals (in the case of sent in the form of identification information, it will be described in detail later). ).

(1) Calling from the mobile radio device 100 will be described with reference to the flow charts shown in FIGS. 8 and 9.

When the power of the mobile radio device 100 is turned on, the radio reception circuit 135 of FIG. 2 includes the downlink (radio base station 30 → mobile radio device 100) radio channel (referred to as channel CH1). After capturing the control signal,
The radio channel (hereinafter referred to as channel CH1) is in the reception state. This is possible by capturing the sync signal in the time slot SDn shown in FIG. 4 (a). The control unit 140 sends a control signal to the synthesizer 121-1 so as to generate a local oscillation frequency that enables reception of the radio channel CH1, and also the switch 122.
-1 is also in a state of being tilted and fixed to the synthesizer 121-1 side.

Therefore, turn off the handset of the telephone unit 101.
When hooked (beginning of call) (S201, FIG. 8), the synthesizer 121-2 of FIG. 2 receives from the control unit 140 a control signal for generating a local oscillation frequency that enables transmission of the radio channel CH1. Also, the switch 122-2 is also tilted to the synthesizer 121-2 side to be in a fixed state.
Next, the call control signal output from the telephone unit 101 is transmitted using the radio channel CH1. This control signal is
This is transmitted by means of the frequency band shown in FIG. 5, for example using the time slot SUn.

Now, the call control signal from the mobile wireless device 100 is properly received by the wireless base station 30, and the mobile wireless device 100 receives the call control signal.
If the ID (identification number) of is detected (S202),
The control unit 40 searches for a currently empty time slot.

As a result, for example, time slot SD
1 is available, the time slot SDn of the radio channel CH1 is used for the mobile radio device 100 by the downlink control signal (time slot uplink (mobile radio device 10)).
0 → radio base station 30) SU1, and downlink (radio base station 30 → mobile radio device 100) SD1 corresponding thereto are instructed to be used (S203).

In response to this, the mobile radio 100 shifts to the receivable state in the designated time slot SD1 and SU1 which is the time slot for the uplink radio channel corresponding to the downlink time slot SD1.
(See FIG. 4B). Mobile radio 1
In the control unit 140 of 00, the transmission / reception interrupt controller 12
3 to operate the switches 122-1 and 122-2 (S204). At the same time, a slot switching completion report is sent to the radio base station 30 using the upstream time slot SU1 (S205), and a dial tone is waited for (S206).

The radio base station 30 has a time slot S
In addition to U1, SU3 and SUn are included in one frame and transmitted as an upstream signal from another mobile radio 100. Upon receiving the slot switching completion report (S207), the wireless base station 30 sends a calling signal together with the ID of the mobile wireless device 100 to the gateway switch 20 (S20).
8). On the other hand, in the gateway switch 20, the mobile radio 10
The ID of 0 is detected, and a necessary switch of the switch group included in the gateway switch 20 is turned on (S209),
The dial tone is transmitted to the wireless base station 30 (S21).
0, FIG. 9).

This dial tone corresponds to the radio base station 30.
Is transferred to the mobile wireless device 100 (S211), and the mobile wireless device 100 confirms that the communication path has been set (S212).

When this state is entered, the mobile wireless device 100
Since a dial tone is heard from the handset of the telephone section 101, the transmission of the dial signal is started. This dial signal is subjected to speed conversion by the speed conversion circuit 131, and the wireless transmission circuit 13 including the transmission unit 134 and the transmission mixer 133.
From 2, the data is transmitted using the upstream time slot SU1 (S213). Thus, the transmitted dial signal is received by the wireless reception circuit 35 of the wireless base station 30.

The wireless base station 30 has already responded to the call signal from the mobile wireless device 100 to determine the time to be used.
The slot is given, and the signal selection circuit group 39 and the signal allocation circuit group 52 of the radio base station 30 are operated to receive the upstream time slot SU1 and receive the downstream time slot SU1.
The state has shifted to transmitting the signal of the slot SD1.
Therefore, the dial signal transmitted from the mobile wireless device 100 is the signal selection circuit 39-1 of the signal selection circuit group 39.
After passing through, it is input to the signal speed restoration circuit group 38, where the original transmission signal is restored and transferred to the gateway exchange 20 as the call signal 22-1 via the signal processing unit 31 (S21).
4) The call path to the telephone network 10 is set (S21).
5).

On the other hand, an input signal (initially a control signal, a call signal if a call is started) from the gateway switch 20 is subjected to speed conversion by the signal speed conversion circuit group 51 in the radio base station 30, and then the signal allocation circuit. Signal allocation circuit 52-1 of group 52
Has given a time slot SD1. Then, the time of the downlink radio channel from the radio transmission circuit 32
It is transmitted to the mobile wireless device 100 using the slot SD1.

In the mobile radio 100, the radio channel CH
In the time slot SD1 of No. 1, the wireless communication circuit 135 is on standby for reception, and its output is input to the speed restoration circuit 138. In this circuit, the original signal on the transmitting side is restored and input to the handset of the telephone section 101.
Thus, a call is started between the mobile wireless device 100 and the ordinary telephone in the ordinary telephone network 10 (S21).
6).

However, the voice signal (telephone signal) of the caller, which will be described later, is already owned in the form of a time piece signal group in which the mobile radio 100 or the radio base station 30 has added identification information to its memory circuit. In some cases, in order to improve frequency utilization, identification information may be collectively transmitted prior to the start of a call. In this case, although the number of time slots used is short (several seconds), a plurality of time slots are used at the same time instead of one. These controls are also performed by the wireless base station 30.
Will be done.

The telephone call section 101 of the mobile radio 100 is used for ending the call.
By hooking the handset of the device on (S217),
The end signal and the on-hook signal from the control unit 140
The signal is transmitted from the wireless transmission circuit 132 to the wireless base station 30 via the speed conversion circuit 131 (S218), the control unit 140 stops the operation of the transmission / reception interrupt controller 123, and the switches 122-1 and 122- 2 is fixed to the output ends of the synthesizers 121-1 and 121-2, respectively.

On the other hand, in the control unit 40 of the radio base station 30,
When the call end signal from the mobile wireless device 100 is received, the call end signal is transferred to the gateway switch 20 (S219), the switches of the switch group (not shown) are turned off to end the call (S).
220). At the same time, the signal selection circuit group 3 in the radio base station 30
9 and the signal allocation circuit group 52 are opened.

In the above description, the control signals are exchanged between the radio base station 30 and the mobile radio 100 without passing through the signal conversion circuit group 51, the signal speed restoration circuit group 38, etc., but this is explained. For the sake of convenience, the signal speed conversion circuit group 51 and the signal speed restoration circuit group 38 are the same as for the audio signal.
Communication can be performed without any trouble through the signal processing unit 31 and the signal processing unit 31.

(2) Incoming call to the mobile wireless device 100 The mobile wireless device 100 is in a standby state with the power turned on.
In this case, as described in the section of calling from the mobile radio 100, the downlink control signal of the radio channel CH1 according to the procedure defined by the system is in the standby state.

An incoming call signal from the general telephone network 10 to the mobile radio 100 via the gateway switch 20 is transmitted to the radio base station 30.
Suppose you have arrived. Similar to the voice signal, these control signals pass through the signal speed conversion circuit group 51 and are transmitted to the control unit 40 (FIG. 3) through the signal allocation circuit group 52, similarly to the voice signal. Then, the control unit 40 uses the empty slot, for example, SD1 of the downlink time slots of the radio channel CH1 addressed to the mobile wireless device 100, and the mobile wireless device 10
0 ID signal + incoming call signal display signal + time slot use signal (for transmission from the mobile radio 100, for example, S
(Use SU1 corresponding to D1). In the mobile wireless device 100 that has received this signal, it is transmitted from the receiving unit 137 of the wireless receiving circuit 135 to the control unit 140. Control unit 1
At 40, since it is confirmed that this signal is an incoming signal to the mobile radio device 100 of its own, at the same time as making a ringing tone from the telephone unit 101, it waits at the instructed time slot SD1, SU1. Transmission / reception intermittent controller 1
23, and switches 122-1 and 12
Turn on and off 2-2. Thus, the call is ready to be made.

(3) System operation in the case of a telephone signal sent in the form of identification information

(3.1) Estimation of the type and total number of time piece signals

(3.1.1) Types of phonemes forming Japanese and their durations FIG. 29 quoted from Document 5 shows the types of phonemes forming Japanese and their durations. With reference to this, the type of the time piece signal related to the present invention is estimated.

Hereinafter, the type of the time piece signal—which will be provided with different identification information—will be investigated to estimate how much the type will be. As an example, a case where the time length of the time piece signal is 10 ms is obtained.

(3.1.2) The time length of the time piece signal is 1
In order to be able to assemble all kinds of voice signals using the time piece signal group at 0 ms (80 samples), at least all of FIG. 29 must be included as the time piece signal. For example, [a]
Considering the case of creating the time fragment signal of, the duration of the phoneme [a] has an average of 10 at the beginning, in the middle, and at the end.
6, 116, 148 ms, deviations are 26, 26,
It is known to be 35 ms. As can be seen from the literature 4 and FIG. 25, the waveform of the phoneme [a] has only slight changes in the beginning and end of the word, and it can be considered that there is almost no change in the word. Set the following conditions for faithful reproduction.

Considering that the maximum frequency of a voice (telephone) signal is 3 kHz and the Nyquist time (1/6000 is fine, the maximum frequency is 3.4 kHz in the case of a cable, 1/8000 sec, that is, 0 ．
If there is a time difference every 125 ms, 125 μs), it is regarded as another signal. As a result, each sample of the signal is considered as a different signal, and noise in the signal sampling process is not mixed (however, quantization noise is mixed).

The phase difference between the signals is the same signal without being considered as another signal. This is a result of utilizing the fact that the auditory sense of the human ear has a weak ability to separate the phase difference, but it is technically easy to correct the phase difference.

The deviation of the phoneme duration is considered to be the repetition of the same signal and is ignored.

The signal level difference is separately set as level information (maximum 8 dB). Therefore, the standard value is stored in the memory circuit.

Under the conditions (1) to (8), for example, the phoneme [a] has a word beginning 106 ms of 848 types and a word of 116 ms.
There are a total of 2800 types of 928 types and 1024 types of endings of 1024 types. When the above calculation is carried out for all the phonemes shown in FIG. 29 to obtain the sum of the time piece signals, Σ [a] _{10 ms} = 27664 (1) is obtained.

Next, the type of time piece signal when phonemes are connected is estimated. This is not enough to create all kinds of speech signals by the above simple phoneme time piece signalization alone. This is because the words have the phonological comrade connection shown in FIG. For example, phoneme [a] is connected to the next phoneme. [A] and [a], [a] and [i], [a] and [u], ..., [a] and [b]. Less than,
Estimate this type.

Since the duration of the time piece signal is 10 ms, for example, considering the connection of [a] and [a],
[A] is (10-0.125, (79 samples)) ms
If [i] lasts 0.125 ms (1 sample) after that, then [a] then becomes (10-0.125 × 2,
(78 samples)) [i] is 0.12 after sustaining ms
If it lasts 5 × 2 ms (2 samples) ..., Finally, after [a] lasts 0.125 ms (1 sample),
[I] is (10-0.125, (79 samples)) ms
There will be a total of 80 types of cases that persist.

[A] and [i], [a] and [u],
Since there are a total of 19 types such as [a] and [e], ..., It can be seen that when all combinations are taken, 80 × 19 = 1520 types are generated. When vowels and consonants are concatenated by the same calculation as above, [i] and [a], [i] and [i], ..., [i] and [b], [u] and [a], [u] ]
, [I], ..., [u] and [b], [c] and [a],
[C] and [i],…, [c] and [b], …………………
When [o] and [a], [o] and [i], ..., [o] and [b] are obtained, the type of the time piece signal is 80 × 19 × 5 = 7600 (2) as a whole. .

Next, the type of concatenation of consonant and vowel is calculated.
In Japanese, there is no case where consonants are concatenated (consonant [kya] is [k] [i] [y] [a]). For example, [k] is a vowel [a], [i], [ Concatenate with u], [e] and [o]. Therefore, as for the type, 80 × 5 = 400 Similarly, when the type in which consonants [s], [sh], ..., [B] are connected is obtained, 80 × 5 × 14 = 5600 (3) is obtained as a whole. Therefore, when the above three cases are summed up,
The type of time piece signal is Σ [a] _10ms × [a] _10ms = 40864 (4) That is, it becomes about 40,000. Therefore, the type of identification information is about 40,000. Further, the capacity required to store the time piece signal group in the storage circuit is obtained. Assuming that the time piece signal length is 10 ms and is stored with the same quality as the PCM 64 kbps transmission, a capacity of 64 kb × 0.01 = 0.64 kb is required to store the one hour piece signal. Therefore, the capacity required to store the 40,000 time piece signal group in the memory circuit is 0.64 kb × 40864 = 26.15 Mb.

(3.2) Examination of the effect of reducing the total number of identification information and the redundancy of the voice signal The effect of reducing the redundancy in the voice signal when the identification information is transmitted instead of the transmission of the voice signal will be examined. It has
The relationship between the time piece signal length and the total number of identification information, the number of bits required for transmitting the identification information, the required amount of information, and the like may be examined.

(3.2.1) Number of bits required for transmitting identification information and required amount of information (time piece signal length 10
When the time piece signal length is 10 ms, the type of the time piece signal is about 40,000 from the equation (4), and the total number of identification information is 40,000.
That is, 16 bits are added, and 8 bits of level information is added to the total of 24 bits. Therefore, the amount of information to be transmitted per second is 24 × 100 = 2400 bps (5), that is, 2.4 kbps. Since this amount is reduced to about 1/27 as compared with 64 kbps of PCM transmission, the transmission of the identification information has a considerable effect of reducing the redundancy of the voice signal.

(3.2.2) Number of bits required for transmitting identification information and required amount of information (time piece signal length 1 m
s) When the time piece signal length is 1 ms, the kind of time piece signal is about 30,000 from the equation (1), and the total number of identification information is 30,000, that is, 15 bits, and level information 8 bits is added to this. The total is 23 bits. Therefore, the amount of information to be transmitted per second is 23 × 1000 = 23000 bps (6), that is, 23 kbps. Since this amount is reduced to about ⅓ as compared with 64 kbps of PCM transmission, in this case, the transmission of identification information is not as effective in reducing the redundancy of voice signals as in the former case.

The above results are displayed as shown in FIG. In addition, in FIG. 10, the time piece signal length is 1, 5, 20,
The results obtained in the same manner are shown for 40 and 80 ms. Examination of FIG. 10 reveals the following points. The longer the time piece signal length, the smaller the required information amount to be transmitted. The longer the time piece signal length, the larger the amount of information to be transmitted. As a result, it is necessary to increase the capacity of the memory circuit and improve the search capability. Is desirable but outside the requirements,
There is a signal delay problem and it cannot be made too large.

Comprehensively judging (1) to (3), it is judged that the optimum time piece signal length is about 10 to 40 ms. However,
Even in this case, (i) the capacity of the memory circuit is still large. (Ii) It is considered necessary to further reduce the required information amount. Therefore, a method of reducing will be considered below.

(3.2.3) Study on Reduction of Types of Time Piece Signals The types of time piece signals of 30,000 to 40,000 obtained in (3.2.1) and (3.2.2) are appropriate. Is it a correct value? If this value is large, the storage capacity increases, the search capability increases, and the required information amount to be transmitted also increases, which has a great influence on the construction of the system.

The content of FIG. 10 shows advantageous results as compared with other voice transmission systems. However, this advantage is for wireline transmission, and for wireless transmission, especially mobile radio, further reduction of voice signal redundancy will be required. This is because, as is well known, the technology for reducing the signal speed of digital mobile radio has made great progress and is currently 4-5.
This is because kbps is also in a practical state. However,
As a result of the low bit rate of digital mobile radio, it is assumed that the transmission quality is slightly degraded.

In this respect, if the system parameters of FIG. 10 are used, in the system for transmitting the identification information, PCM 64 kbps at a signal speed almost the same as that of digital mobile radio.
Since it is assumed that transmission quality comparable to signals is secured, it seems that there is an advantage in this respect.

In the following, it is assumed that a slight deterioration in transmission quality is allowed, and a study on reduction of the types of time piece signals will be made.

According to Reference 6, the number of phoneme levels required for speech synthesis is about 1500. Why did such a big difference come out? The cause will be considered below.

In Document 6, the types of phonemes [a], [i], ... Are one each, whereas in the time piece signal, for example, [a] is 2,800 types. In the case of the phonemes to be connected, for example, [a] and [i], the type is set to 1 in Document 6, while the time piece signal is set to 1520 types in the case of [a] and [i]. Although the level information is set to 8 bits, actually, this bit amount is not necessary, and 4 bits at most is sufficient.

From the above, (3.2.1) and (3.
It is considered that the time piece signal types 30,000 to 40,000 obtained in 2.2) have considerable redundancy. Then, we will examine how much reduction is possible below.

(A) Types of time piece signals of simple phonemes and concatenated phonemes As can be seen from Reference 3, the waveform of each phoneme is "There is a change in the etymology and ending, and there is almost no change in the word. There are many cases. ”Moreover, since the change of the beginning and the ending of the word is relatively simple, it can be corrected by the level information, and the following assumption is made. i) The beginning and ending are considered as different signals every 1 ms (8 samples). ii) Although there is little change in the word compared with the beginning and end of the word, here, with a margin, it is considered as another signal every 1 ms (8 samples). iii) It is determined that the connected phonemes can be regarded as different signals every 1 ms (8 samples) as in the case of simple phonemes.

(B) Simple phoneme in the case where the time piece signal length is 10 ms and the types of phonemes to be connected Based on the assumption set in (a), the time piece signal length of the phoneme is 10
In the case of ms, the types of simple phonemes and connected phonemes are obtained. According to FIG. 25, the duration of the phoneme [a] is 106, 116, and 148 ms at the beginning, inside, and ending, respectively.
There are a total of 312 types of 58 types for ms and 148 types for endings of 148 ms.

When the above calculation is carried out for all phonemes shown in FIG. 29, the types of simple phonemes are given by the following equation. Σ [a '] _10ms = 3458 (7)

Next, the types of phonemes to be connected are obtained. The type is given at 1650, by a method similar to that previously found. Therefore, the sum of the types of simple phonemes and connected phonemes is Σ [a ′] _{10 ms} × Σ [a ′] _{10 ms} = 5108 (8). Further, the capacity required to store the time piece signal group in the storage circuit is calculated as 3.27 Mb.

As a result of the above reduction of the number of signals per time,
The peak waveform of the telephone signal may be suppressed (blunted), which may adversely affect the information transmission of the signal.
As shown in FIG. 27, this is not a big problem because the peak removal does not cause much deterioration. The same calculation as above is carried out for the time piece signal length 5, 10, 20, 40 and 80 ms.
Is performed, the result shown in FIG. 15 is obtained.

(3.3) Description of the structure and operation of the speed conversion circuit 131 and the speed restoration circuit 138

(3.3.1) Outline of Operation FIG. 11 shows an internal circuit configuration diagram of the speed restoration circuit 138, the telephone unit 101 and the speed conversion circuit 131. The speed conversion circuit 131 includes a time piece signal generation circuit 169,
A telephone signal storage circuit 171, a time compression circuit 173, a time piece signal group storage / comparison circuit 181 having a function of accumulating and storing identification information of a time piece signal and collating and comparing.
And an identification information transmitting / receiving circuit 182 for transmitting / receiving the identification information,
It is composed of the switch 177.

Now, the telephone signal from the telephone unit 101 is
The time piece signal generation circuit 16 which is output from the transmission circuit 107
9 is input. Here, the telephone signal is normally disconnected at regular time intervals. The disconnected telephone signal is input to the telephone signal storage circuit 171. Here, after being stored for a certain period of time and after timing adjustment, a part is sent to the time piece signal group storage / comparison circuit 181 and the rest is sent to the time compression circuit 173.

The signal input to the time piece signal group storage / comparison circuit 181 is compared and collated with each of the already stored time piece signal groups, and identification information is added when it is determined that the signal has not been stored yet. Are stored. There are two possible methods for this case. i) Normalize the power of the time piece signal (unify it to a certain level). ii) The phase of the time piece signal is stored in a form that can be easily compared with other signals later.

The signal input to the time compression circuit 173 is
It is time-compressed with a fixed value determined by a normal system and sent to the wireless transmission circuit 132. However, the same identification information that is added when these time piece signals are stored in the time piece signal group storage / comparison circuit 181 from the identification information transmitting / receiving circuit 182 is added. Note that the identification information is added to the lower sideband of the time-compressed time piece signal (thus coexisting with the control signal, but the coexistence is possible because the information amount of the signal is small. See FIG. 5. ).

If the time piece signal group storage / comparison circuit 18
If the signal input to 1 is found to be the same as one of the time-series signal groups already stored as a result of comparison (however,
If the waveforms are the same even if the power levels and phases of the telephone signals are different, it is determined that they are the same time piece signal. In the time piece signal group storage / comparison circuit 181, the signal input to the control unit 140 is Is the same as the signal already stored,
Notify the other party of communication that the time piece signal should not be sent, and that only the identification information of the time piece signal that has already been stored should be sent.

Upon receipt of this notification, the control section 140 transmits from the identification information transmission / reception circuit 182 a signal obtained by adding level information or the like to the identification information of the time piece signal already stored in the time piece signal group storage / comparison circuit 181. To send a control signal to the switch 177 so that the terminal a of the switch 177 is turned on and the terminal b is turned off. As a result, the output signal of the time compression circuit 173 is not transmitted, but the identification information (and level information, phase information, etc.) from the identification information transmission / reception circuit 182 is added. Hereinafter, when there is no misunderstanding, it is simply referred to as identification information. ) Will only be sent.

The radio base station 30 which has received the signal from the mobile radio 100 described above executes the following operation. Demodulation of the received compressed time piece signal Store the received time piece signal in the memory circuit

Needless to say,
The feature of storing the time piece signal in the storage circuit is to add and store the identification information transmitted in addition to each time piece signal. As a result, it becomes possible to share the information stored in the time piece signal group storage / comparison circuit 181 of the mobile wireless device 100 with each other. That is, if the mobile radio device 100 instructs, for example, “information of the identification information 1001”, the radio base station 30 has its own signal speed restoration circuit group 38.
Therefore, it is possible to select the time piece signal which is exactly the same as that of the mobile wireless device 100. Then, the actual telephone signal is a telephone signal waveform having appropriate power and phase by adding "level information etc." included in the identification information to this time piece signal (however, it is a fragmentary signal in terms of time). Is played.

Next, the speed restoration circuit 138 will be described.
The speed restoration circuit 138 is composed of switches 165 and 167, a telephone signal storage circuit 161, a time extension circuit 162 and a telephone signal reproduction circuit 163. Wireless reception circuit 13
The signal sent from the V.5 to the speed recovery circuit 138 is a compressed time piece signal (including identification information). Identification information for identifying the time piece signal. In the case of the time piece signal compressed by the switch 165 operated by the control signal of the control unit 140, the terminal a is turned on, and the telephone signal storage circuit 161.
Sent to. In the case of identification information, the terminal b of the switch 165
Is turned on and is sent to the identification information transmitting / receiving circuit 182.

Now, part of the compressed time piece signal sent to the former telephone signal storage circuit 161 is part of the time extension circuit 1.
At 62, the time is expanded to form a time piece signal having the original time length. Further, the remaining time piece signal is sent to and stored in the time piece signal group storage / comparison circuit 181 together with the identification information sent at the same time. The time piece signal sent to the telephone signal storage circuit 161 and extended by the time extension circuit 162 turns on the a side of the switch 167 at a constant timing based on an instruction from the control unit 140, and the telephone signal reproduction circuit 163. Is input to the receiving circuit 106 of the telephone unit 101. The telephone signal input to the receiving circuit 106 outputs voice.

On the other hand, the identification information sent to the latter identification information transmission / reception circuit 182 is the time piece signal group storage / comparison circuit 18
1 is applied, and based on this, a time piece signal corresponding to this is extracted from the time piece signal group storage / comparison circuit 181, and after adjusting the level and the like according to the level information and the like, via the terminal b of the switch 167. The voice signal is input to the telephone signal reproduction circuit 163, the same process as the former is followed, and the voice is output from the reception circuit 106. The input to the telephone signal reproduction circuit 163 is possible because the switch 167 is turned on at a certain timing based on an instruction from the control unit 140.

FIG. 12 shows a time piece signal group storage / comparison circuit 1
81 and the internal configuration of the identification information transmission / reception circuit 182 are shown. The signal comparison circuit 501 interfaces with the time piece signal group storage / comparison circuit 181 and an external circuit. The inputs to the signal comparison circuit 501 are the signals from the telephone signal storage circuit 171 and the time expansion circuit 162 of FIG. 11, and the output is to the switch 167. The output of the signal comparison circuit 501 is applied to the storage circuits 501 and 502. Although both are writable and readable memory circuits, the memory circuit 502 stores control information in the memory circuit 503.
Is used when you want to send and receive identification information all at once. Of course, when the capacity of the memory circuit is large, the memory circuits 502, 5
One is not necessary to divide into 03.

The identification information transmission / reception circuit 182 includes an identification information transmission circuit 522 and an identification information reception circuit 523 which interface the circuit with an external circuit. The input to the identification information receiving circuit 523 is the switch 16 of FIG.
5 from the wireless reception circuit 135 (FIG. 2), and the output from the identification information transmission circuit 522 is to the switch 177. The identification information creation circuit 521 creates identification information to be added to the time piece signal based on the information from the signal comparison circuit 501, and sends it from the identification information transmission circuit 522. The identification information analysis circuit 524 obtains the identification information received from the communication partner from the identification information reception circuit 523, reads the time piece signal stored in the address of the storage circuit 502.503 indicated by the identification information, and outputs the signal. The comparison circuit 501 outputs it to the switch 167 according to an instruction from the identification information analysis circuit 524. The interface between the identification information transmission / reception circuit 182 and the time piece signal group storage / comparison circuit 181 is connected as shown in the figure, and the above-mentioned functions can be executed.

Further, the time piece signal group storage / comparison circuit 181
And the identification information transmission / reception circuit 182, the control unit 1 shown in FIG.
In addition to transmitting and receiving control signals to and from 40, clock signals and timing signals are obtained from the clock regenerator 141 and the timing generator 142, respectively.

(3.3.2) Method of inputting voice (telephone signal) of the caller to the memory circuit of the mobile radio device

(A) Initial Setting Operation A method in which the user of the mobile wireless device 100 inputs his own voice (telephone signal) into the memory circuit of the mobile wireless device 100 prior to the calling operation will be described.

Prior to the calling operation, the mobile radio 100 is set to the "initial setting mode". This can be set by dialing a specific number in the dial section of the telephone section 101 of the mobile wireless device 100. In this "initial setting mode", the calling circuit 107 of the telephone section 101 asks the caller to utter a "call signal test pattern", and this is stored in the time piece signal group storage / comparison circuit 181. It is an action.
In the "call signal test pattern", for example, 50
It may be a reading of a monophonic table containing sounds, voiced sounds, semi-voiced sounds, choruses, etc., or it may be part of a novel containing all types of audio elements so that any word can be composed. However, it is necessary to carefully read the monophonic table for about 20 minutes and a part of the novel for about 30 minutes. This is because every kind of time piece signal must be stored so that only the identification information needs to be transmitted in every telephone signal thereafter. In addition, since every voice must be stored as a voice (telephone signal) of the owner of the mobile wireless device 100, the "initial setting" is performed even when the emotion is different or the voice is slightly different due to a cold (nasal voice). It is desirable to store in "mode".

When the controller 140 recognizes that the mobile wireless device 100 is in the "initial setting mode", the controller 140
Sets the circuit of the telephone unit 101 so that the output of the transmission circuit 107 is input to the time piece signal generation circuit 169.
In this state, the voice signal of the "call signal test pattern" is input to the time piece signal generation circuit 169. The time piece signal generation circuit 169 obtains a control signal from the control unit 140, cuts the voice signal at a constant time interval (for example, 20 ms), sends the cut voice signal to the telephone signal storage circuit 171, and temporarily After storing, it is sent to the time piece signal group storage / comparison circuit 181 at a fixed timing. The time piece signal group storage / comparison circuit 181 stores these time piece signals after adding identification information. These stored time piece signals are signals different from each other, and the same signal as that already stored is discarded and not stored.

The method of storing different time piece signals as described above has already been described in (3.3.1) i) and ii). As a result of the above "initial setting mode", the voice (telephone signal) of the user of the mobile wireless device 100 is added to the hour piece signal group storage / comparison circuit 181 in the form of a time piece signal, and identification information is attached to each piece. Will be entered.

(B) Time-series signal group storage during initial setting
Operation of the Comparator Circuit 181 Since the voice signal of the calling party is input to the time / slice signal group storage / comparison circuit 181 in the form of a time slice signal, storage is started. Upon storage, identification information is added to these time piece signals. The time piece signals to be stored are different from each other, and the same signal is discarded and not stored. This can be done by the following method.

First, the input time piece signal is sampled at the Nyquist time interval of the highest frequency of the signal or at a time interval shorter than that. Then, the level information of the sampling signal is digitized and stored in time series. In this case, the amount of information to be stored is 160 samples when the Nyquist time interval is 1/8000 seconds in the first time piece signal of 20 ms, for example, since the signal is made into time piece signals, which is not so large.

Next, when the second time piece signal is sent, the level information of the previously stored first time piece signal is sequentially compared in time series. That is, each sample value of the digital signal, for example, each amplitude value a ₁ , a ₂ , a ₃ , a ₄ from the first sample signal to the fourth sample signal of the first time piece signal, and the second time piece The amplitude values a ₁ ′, a ₂ ′, a ₃ ′ and a ₄ ′ of the signals from the first sample signal to the fourth sample signal are compared with each other, and even one of them has identification information. And remember.

Further, when the third time piece signal is sent, its amplitude value and the level information of the previously stored first and second time piece signals and the amplitude value of the third time piece signal are stored. Each comparison is compared with each amplitude value from the first sample signal to the fourth sample signal, and if any one of them is different, identification information is added and stored. This comparison operation is similarly performed for the fourth and subsequent time piece signals input next.

As a result, the time piece signal group storage / comparison circuit 1
In 81, a number of pieces of level information of time piece signals having different characteristics from each other (about 5000 as described in equation (8), but this value slightly fluctuates due to individual differences) are stored with identification information attached respectively. To be done.

(C) Description of Operation of Time Piece Signal Group Storage / Comparison Circuit 181 During Calling Party Call (While Telephone Signal is being Transmitted) FIGS. 13 and 14 show that the calling party is talking (during telephone signal transmission). ), The operation flow of the time piece signal group storage / comparison circuit 181 which plays a main role in transmitting the identification information in place of the telephone signal to be transmitted to is shown.

When the mobile radio 100 is powered on,
Each component of the mobile wireless device 100 enters the operating state, and the time piece signal generation circuit 169, the telephone signal storage circuit 171, the time piece signal group / comparison circuit 181, the identification information transmission / reception circuit 182.
Etc. also start operation.

From the telephone signal storage circuit 171, sample values (amplitude values) x _{1 to} x ₄ of the time piece signal LX of the transmission signal (hereinafter referred to as the transmission time piece signal LX) are stored and compared with the time piece signal group. It is taken into the circuit 181 (S101, FIG. 13).
Here, for the sake of explanation, one transmission time piece signal LX is set to 4
Can be represented by sample values x _{1 to} x ₄ . In practice, it is represented by a much larger number, for example 160 sample values.

In the first sample comparison, the first sample signal x ₁ of the transmission time piece signal LX (its amplitude, that is, the sample value is x _1, and hereinafter, this is referred to as the transmission first sample signal x ₁ ). Is a first sample signal a _1j (amplitude, that is, sample value is a _1j , j = 0, ± 1, ± 2 , ± 3, ..., which is hereinafter referred to as the stored first sampled signal a _1j ), and a _{1j for} which x ₁ = a _1j is obtained. That is, the value of j is specified (S102).

[0144] Already the first sample value is equal to the amplitude (x ₂₎ of the second sample signal x ₂ for transmission (hereinafter this transmission second sample signal x ₂₎ from a subgroup of a _1j In order to obtain the stored second sample signal a _2i (hereinafter referred to as the stored second sample signal), its amplitude (a _2i , i = 0, ± 1, ± 2, ± 3, ...) A comparison is made to obtain a _{2i such} that x ₂ = a _2i . That is, the value of i is specified (S103).

Similarly, in the third sample comparison,
Is the third sampled signal x₃And a_1j, A_2iSub glue
Stored third sample signal a belonging to_3h(H = 0, ± 1,
± 2, ± 3, ...) are compared and the amplitudes of both are the same.
The value is specified, a_3h(S104), the fourth sample
In comparison, the transmitted fourth sampled signal x_Four And a_1j，
a _2i, A_3h4th sample signal belonging to the subgroup of
Issue a_{4 g}(G = 0, ± 1, ± 2, ± 3, ...)
Then, the value of g at which the amplitudes of both are the same is specified, and a_{4 g}To
After obtaining (S105), the specified a of j, i, h, and g_1j，
a_2i, A_3h, A_{4 g}Has the first to fourth sample signals stored in
The stored time piece signal LA is equal to the transmission signal time piece signal LX.
It is confirmed that the match (LA = LX) occurs (S106).

Therefore, the identification information Ma given to the stored time piece signal LA is read from the time piece signal group storage / comparison circuit 181, and the identification information transmission / reception circuit 182 to the switch 17.
It is sent to the wireless transmission circuit 132 via the terminal a of 7 (S
108), the wireless base station 30 receives this, and ends the transmission of the identification information of one time piece signal. This operation is similarly executed for the next time piece signal.

FIG. 14 shows a subroutine for the first sample comparison of x ₁ and a _1j shown in step S102 of FIG. When the first sample signal x _{1 of} the transmission time piece signal LX is fetched from the telephone signal storage circuit 171 into the time piece signal group storage / comparison circuit 181 (S131, FIG. 14), j = 0 is set (S132). ), The stored first sample signal a _1j (j = j) from among the time piece signals (assumed to be 5000) already stored in the initialization operation.
0) is read after correcting the level difference and the phase (S133),
Amplitude comparison is performed with the first sampled signal x ₁ (S1).
34). When x ₁ = a _1j (j = 0) is obtained in this amplitude comparison, the process _proceeds to step 139 and x ₁ = a _1j (j =
0) is recognized and the process proceeds to step S103.

In the amplitude comparison in step S134, the transmitted first sample signal x ₁ is the stored first sample signal a.
When it is determined that the value is larger than _1j (j = 0) (S134>), the value is reset to j = j + 1 = 1 (S1).
35), the transmitted first sample signal x ₁ is compared with the stored first sample signal a _1j (j = 1) (S136), and the transmitted first sample signal x ₁ is stored first sample signal a _1j (j = 1)
If it is larger than (S136>), the step S is performed again.
Returning to 135, j = j + 1, that is, j = 2 is set, and the transmitted first sample signal x ₁ and the stored first sample signal a
_{The 1j} (j = 2) comparison is repeated. In this comparison, the transmitted first sample signal x ₁ and the stored first sample signal a _1j
If no match is obtained, then in S135, j
= 3, 4, ... And comparison is performed, and when x ₁ = a _1j is obtained (S136 =), the process proceeds to step S139, x ₁ = a _1j is recognized, and the process proceeds to step S103.

In the amplitude comparison in step S134, the transmitted first sample signal x ₁ is the stored first sample signal a.
When it is determined that the difference is smaller than _1j (j = 0) (S134 <), the value is reset to j = j-1 = -1 (S134 <).
137), the transmitted first sample signal x ₁ is compared with the stored first sample signal a _1j (j = −1) (S138), and the transmitted first sample signal x ₁ is stored as the first sample signal a _1j ( j
If it is smaller than (−1) (S138 <), the process returns to step S137 again, j = j−1, that is, j = −.
2 is set, and the comparison of the transmitted first sample signal x ₁ and the stored first sample signal a _1j (j = −2) is repeated. In this comparison, if no match can be obtained between the transmitted first sample signal x ₁ and the stored first sample signal a _1j , S13 is further executed.
7, j = -3, -4, ... Is set and compared, and when x ₁ = a _1j is obtained (S138 =), the process proceeds to step S139, x ₁ = a _1j is recognized, and step S
Move to 103.

In the second sample comparison subroutine of step S103, in FIG. 14, x _{1 is set} to x ₂ and a _1j
If a is replaced with a _2i and j is replaced with i, they can be applied as they are. In the third sample comparison subroutine of step S104, x _{1 is set} to x ₃ and a _{1j is set} to a _{3h in FIG.}
If j is replaced by h, it can be applied as it is.
In the fourth sample comparison subroutine of step S105, x _{1 is set} to x ₄ , a _{1j is set} to a _4g , and j is set to j in FIG.
If g is replaced by g, it can be applied as it is. Less than,
The same applies when the number of sample comparisons is increased to the fifth, sixth, seventh, ... In FIG.

In the comparison of steps S134 to S139 (FIG. 14), the transmitted first sample signal x ₁ is always the stored first sample signal a _1j (j = 0, ± 1, ± 2, ± 3, ...).
Should match either. Because every possible value of the stored first sample signal a _1j is 8 bits,
That is, assuming that the total number of the storage time piece signals LA is 5000, the number of storage time piece signals LA is 5000/256 = 19.5. Some sample signals coincide with the first sample signal x ₁ of the transmitted voice. However, in reality, since the time-slice signals LA and LX of the voice signal of the specific individual, the storage time-slice signal LA in which the stored first sample signal a _1j matches the transmitted first sample signal x _1.
It can be estimated that there are 100 or more. Here, many stored first sample signals a _1j (j = 0, ± 1,
± 2, ± 3, ...) Considering their occurrence frequencies, the one with the highest frequency is j = 0, that is, a ₁₀
If it is stored in the time piece signal group storage / comparison circuit 181 as, the comparison operation of FIG. 14 can be completed in a short time.

The process proceeds to the second sample comparison in step S103 (FIG. 13), and in the subroutine shown in FIG.
Then, when x ₁ is replaced with x ₂ , a _1j is replaced with a _2i , and j is replaced with i, the possibility that the transmitted second sample signal x ₂ matches the stored second sample signal a _2i will be examined.

In the first sample comparison (S102), x
₁ = the number of a _1j become a _1j assuming 20, the second sample signal _{a 2i (i = 0, ±} 1, ± 2, ± 3, ...) potential has already been reduced to 20 On the other hand, since the correlation between x ₂ and x ₁ is a voice signal, the correlation is large, and there are 10 possibilities of x ₂ . Then, in the 20 subgroups in which x ₁ = a _1j , x ₂ = a _2i (i = 0, ±
1, ± 2, ± 3, ...) 20/10 = 2, that is, the average probability of decreasing to two. Again, i = 0
The second sample signal to be stored is the stored first sample signal a _1j
If the most frequent subgroup is selected, the comparison work will be completed in a short time. When the subroutine of the second sample comparison (S103) is read in FIG. 14 and applied, steps S134, 136, 138
In the comparison of, for example, when i = 3, x ₂ > a ₂₃ , and in the cycle next to steps S135 and S136, i
= 4, when x ₂ > a ₂₄ , approximately x ₂ =
It may be judged as a ₂₃ . At i = −1, x ₂ <a _2-1 and steps S137 and S138.
In the next cycle of, when i = -2, when x ₂ > a _2-2 , it may be judged and processed approximately as x ₂ = a _2-1 . The above operation flow is the case where the comparison work is most successfully performed, and in reality, a considerable number of sample signals will be compared.

In the operation shown in FIG. 13, the case where the respective sample comparison operations are sequentially performed in the time series in the order of the first (S102), the second (S103), the third (S104), ... , Each sample comparison operation, for example, first, fourth,
The comparison may be performed in a time-sequential manner, such as in the tenth case, at predetermined time intervals. The sample comparison operation described here is similarly performed in the radio base station 30.

(D) Time signal group storage / comparison circuit 18
Storage capacity required for 1 time piece signal group storage / comparison circuit 1 for storing identification information of the time piece signal for each time piece signal length shown in FIG.
Calculate the required capacity for 81. For example, when the time piece signal is 20 ms, in order to store the telephone signal at the speed of 64 kbps, the storage capacity required for one time piece signal is 64 kbps x 0.02 = 1.28 kb, and therefore there are 6758 types. The required storage capacity required to store the time piece signal of (the number of identification information in FIG. 15) is 1.28 kb × 6758 = 8.65 Mb. This capacity is readily available with current technology. Similar calculation is shown in FIG. 15 for other time piece signal lengths of 5 ms and 10 m.
The results obtained are also shown for s, 40 ms, and 80 ms.

(E) Example of system in which operation equivalent to "initial setting operation" is performed during caller's call In the above description, "initial setting operation" is performed prior to caller's call.
It was an example of a system when performing. Depending on the system, the “initial setting operation” may not be performed, and the operation equivalent to the “initial setting operation” may be performed during the call of the caller. Although this is slightly inferior to the above-mentioned system in terms of frequency utilization efficiency described later, it is superior in terms of user convenience improvement and will be described below.

The telephone signal from the telephone section 101 of FIG. 11 is output from the transmitting circuit 107 and input to the time piece signal generating circuit 169. Here, the telephone signal is normally disconnected at regular time intervals. The disconnected telephone signal is input to the telephone signal storage circuit 171. Here, after being stored for a certain period of time and after timing adjustment, a part is sent to the time piece signal group storage / comparison circuit 181 and the rest is sent to the time compression circuit 173.

The signal input to the time piece signal group storage / comparison circuit 181 is compared and collated with each of the already stored time piece signal groups, and identification information is added when it is determined that the signal has not been stored yet. Are stored. In this case, the signal input to the time compression circuit 173 is time-compressed with a constant value determined by a normal system, and the wireless transmission circuit 132
Sent to. However, for these time piece signals, the time piece signal group storage / comparison circuit 1 from the identification information transmission / reception circuit 182.
The same identification information as that given when stored in 81 is added. It should be noted that the identification information is added to the lower detection band of the time-compressed time piece signal (thus coexisting with the control signal, but the coexistence is possible because the information use of the signal is small. See FIG. 5. ).

If the time piece signal group storage / comparison circuit 18
If the signal input to 1 is found to be the same as one of the time-series signal groups already stored as a result of comparison (however,
If the waveforms are the same even if the levels are different, it is determined that they are the same time piece signal).
At 81, the signal that has just been input is the same as the signal that has already been stored in the control unit 140, and the time piece signal should not be sent to the other party of communication, but only the identification information of the already stored time piece signal should be sent. I will contact you.

Upon receipt of this notification, the control section 140 instructs the switch 177 to transmit from the identification information transmitting / receiving circuit 182 the identification information of the time piece signal already stored in the time piece signal group storage / comparison circuit 181. Send a control signal,
The switch 177 is turned on and a is turned off.
As a result, the output signal of the time compression circuit 173 is not transmitted, but only the identification information (and level information is added to this) from the identification information transmission / reception circuit 182. Hereinafter, when there is no misunderstanding, it is simply referred to as identification information. Will be done.

By adopting the method described above, it becomes possible to perform the same operation as the "initial setting operation" described in (a) while the caller is talking. Also,
The explanation in this section has been about the transmission of the identification information of the telephone signal of the uplink (from the mobile radio 100 to the radio base station 30). Conversely, the downlink (from the radio base station 30 to the mobile radio 10
The transmission of the identification information of the telephone signal (to 0) can also be performed by providing the radio base station 30 with a circuit that performs an operation equivalent to that of the time piece signal group storage / comparison circuit 181.

(3.4) System Operation Example When Transmitting Identification Information Hereinafter, a system operation example in which not only the telephone signal but also the identification information of the time piece signal may be transmitted will be described.

Since an example of an outgoing / incoming call operation using the TCM signal has already been described in (1) and (2), the identification information (and its level information) of the time piece signal in the outgoing call operation of the mobile radio 100. Figure 8 shows the operation when sending
This will be described using 9 and 9. Mobile radio 100 off
From the hook operation (S201) to the telephone network 1 of the gateway switch 20
There is no particular change in the operation between the communication path setting to 0 (S215).

Next, the call state (S216) is entered. First, the telephone signals transmitted by both parties use initially compressed time piece signals. Hereinafter, the time piece signal from the mobile wireless device 100 will be described. "Mosimos" from the caller
i, if the duration of "if" is 2 seconds. If the frame length of the TCM signal is 20 ms, the numerical value (sample number) of the time piece signal is 100. These first "mosi"
100 samples representing vowels and consonants are transmitted in the form of a time piece signal (to which identification information is added respectively), and are simultaneously stored in the time piece signal group storage / comparison circuit 181.

The operation of the radio base station 30 receiving these signals is as described above. Now, the next "mosi" is input to the telephone signal storage circuit 171 of FIG. 11 (of course, it will be input in order not in time but in time series). The 100 samples of this signal are almost the same as the sample signals already stored in the time piece signal group storage / comparison circuit 181 for both vowels and consonants.
However, in some cases, the phases are different, and in that case, only that signal is transmitted in the form of a time piece signal in the same manner as described above. Therefore, the time piece signal group memory
The comparator circuit 181 informs the control unit 140 that the corresponding identification information (and the level information of the signal) should be transmitted to the signal determined to be the same as the already stored time piece signal. As a result, these 1
The identification information given to each of the 00 samples is transmitted. These amounts of information are extremely small as compared with the case of transmitting a time piece signal as will be described later, and make a great contribution to the effective frequency utilization.

In the radio base station 30, a circuit (not shown, hereinafter abbreviated as a base station storage / comparison circuit) that operates similarly to the time piece signal group storage / comparison circuit 181 included in the mobile radio 100. ) Is provided, and the time piece signal instructed by the mobile wireless device 100 has already been stored in the circuit, the base station storage / comparison circuit from the mobile wireless device 100.
It is possible to select the time piece signal instructed by, and the original telephone signal can be restored by processing these signals by the signal speed restoration circuit group 38. Hereinafter, although the call signals are transmitted one after another from the mobile wireless device 100, the same processing as above is also executed in the wireless base station 30. As a result, the probability of transmitting the identification information gradually increases.

Although the transmission from the mobile radio device 100 has been described above, the signal transmitted from the radio base station 30 is processed in the same manner. Therefore, the call signal is prolonged (3 minutes or more, the total number of time piece signals generated in this case is 9000), and the compressed time piece signal is rarely transmitted, and only the identification information (and the level information of the signal) is transmitted. This system can be said to be the transmission of a signal, and this system greatly contributes to the effective utilization of frequency.

As described in the equation (8), the type of the time piece signal is 5000 (when the sampling operation is at 1 ms intervals, the number of identification information is 5108 and the same 20 ms when the time piece signal length is 10 ms, as shown in FIG. However, in the following example, the number of pieces of identification information is both 5000.) Therefore, after the above-mentioned call, the time piece signal group storage / comparison circuit 181 of the mobile wireless device 100 stores , It is determined that a time piece signal capable of assembling almost any Japanese language is stored.
If this is insufficient, the number of calls may be further increased and an unstored time piece signal may be stored each time. As a result, as long as the users of the mobile wireless device 100 are not different, it is possible to collectively transmit the identification information as described later.

(3.5) Creation of new time piece signal from received (transmitted) telephone signal In the case to be described below, the present invention is applied to the other party with whom the person owns the time piece signal group and its identification information. This is an effective method when applied when they are not common at all.

In this case, the first signal of the call, eg all telephone signals such as "Hello", are transmitted to each other in the form of time piece signals. Of course, in this case as well, the identification information is attached to each time piece signal. Therefore, if this state continues for several minutes, as described above, the transmission of the signal of only the identification information gradually increases instead of the time piece signal, and finally the communication of only the identification information should be performed. However,
In terms of signal transmission efficiency, the above method is not always effective. Applying the method described below,
This is because after about 30 seconds to 1 minute from the start of the call, the subsequent call can be performed by communication using only the identification information.

(3.5.1) In the case of creating a new time piece signal without compressing or expanding the telephone signal FIG. 16 (a) shows the first telephone signal of the call. For example, in the case of "Hello", it is estimated that all the first "Hello" are time-series signals. Assuming that the time piece signal length is 10 ms and "if" takes 1 second, the total number of time piece signals is 1 s / 10 ms = 100, that is, 100 time piece signals are connected and transmitted. Identification information [00001] ~
[00100] is attached.

FIG. 16B shows an example in which different time piece signals are created from the time piece signals of FIG. 16A. (B) shows a case where a 9/10 part of the time piece signal [00001] and a 1/10 part of the time piece signal [00002] are added to create a new time piece signal [100100291]. However, [100100291] indicates the identification information of the newly created time piece signal. Here, the first 1 is the type of synthesis, and 001002 is the time piece signal [00001].
And the synthesis from the time piece signal [00002]. In addition, 90% of 91 is [00001] and 10% is [0
0002]. FIG.
The same applies to (c) and (d). Therefore, it can be seen that the method as described above can generate about 1000 new time piece signals per second. If these are stored together with common identification information for both transmission and reception, it is possible to store 10 times as many time-series signals as compared with the conventional method.

Next, a method of creating a new time piece signal from adjacent time piece signals will be advanced one step, and a method of creating a new time piece signal from non-adjacent time piece signals will be described with reference to FIG. .

The time piece signal [000] in FIG.
01] in the upper right-handed coarse diagonal line portion and the time piece signal [00002] in the lower right-handed rough diagonal line portion of 5/10 are added together, and the new time piece signal [200] in FIG.
100255], and the finely shaded part of the right downhill of 2/10 of the time piece signal [00099] and the finely shaded part of the right upper part of 8/10 of the time piece signal [00100] in FIG. A case where a new time piece signal [209910028] of FIG. However, [200100255], [2
09910028] indicates the identification information of each newly created time piece signal.

Here, the first 2 of [200100255] is the type of synthesis, and 001002 is the time piece signal [0000].
1] and the time piece signal [00002], and
As for 55, 50% is [0001] and 50% is [000
02]. [20991
The same applies to [0028].

With the method as described above, it is possible to generate a large number of new time piece signals per second. That is,
Explain how many will be. Since each part of the time piece signal [00001] (10 parts if divided by 1 ms) can be combined with each part of the time piece signal [00002] (10 parts if divided by 1 ms), 10 There are streets (the number is limited because the time length must be 10 ms). Next, each part of the time piece signal [00001] is converted into the time piece signal [0000].
It is possible to combine with each part of [3], so 1
There are 0 ways. Furthermore, the time piece signal [0000
3], [00004], ... [0000n] can be combined with each other, so that n is 1 when creating a new time piece signal.
If 0, there are 100 ways. Each part of the time piece signal [00001] may create a new time piece signal by each part of the time piece signal [00002] and each part of the time piece signal [00003] (because the time length must be 10 ms. The number is limited). As a result, the number of new time piece signals generated by the method described with reference to FIGS. 17A and 17B is further increased. , And the number of time piece signals is set to 300, for example.

The same creation method as the above method exists even if the time piece signal [00002] is mainly considered. Considering up to the time piece signal [00100], the types of new time piece signals that can be created are 300 × 100 = 30000. However, it goes without saying that the time piece signal [00100] and the like are combined with the time piece signal sent in the next one second to create a new time piece signal.

It should be noted that the time piece signal is infinitely large in size, but most of them are the same (practically the same), and the same identification information is given after all. Is. That is, the total number of pieces of identification information is within 5000 to 10,000 as shown in FIG. In that sense, the identification information [200 of FIG.
100255] and the like should be considered as provisional identification information, and the identification information is organized in both storage circuits that communicate with each other.
It is a good idea in terms of storing and transmitting information to keep the number of identification information within 5000 to 10,000.

A method for creating a new time piece signal by temporally reversing the context of time piece signals that are not adjacent to each other will be described with reference to FIGS. 17 (a) to 17 (c). 5/10 of the time piece signal [00001] of FIG.
Rough shaded area in the upper right corner of time and the time piece signal [00002]
When a new time piece signal [300200155] of FIG. 17 (c) is created by adding the rough diagonal line on the right down of 5/10 of FIG. 2/10 of the time piece signal [00099] of FIG.
0100] and a finely shaded part in the upper right corner of 8/10 are added together with the temporal relationship (time series) reversed and a new time piece signal [310009982] in FIG. Show. However, [3002001
55] and [310009982] respectively show the identification information of the newly created time piece signal.

Here, the first 3 of [300200155] is the type of synthesis, and 00201 is the time piece signal [0000].
1] and the time piece signal [00002], which means that the component of the time piece signal [00002] comes first in time. In addition, 55 of 50% is [000
02] means that 50% is a component from [00001]. The time piece signal [31000 in FIG.
The same applies to 9982]. In this case as well, as described above, there is no limit to the generation of new time piece signals, but since there are many identical signals among them, it is necessary to organize them.

By using the above method, it is judged that it is possible to store a time piece signal group capable of assembling an arbitrary telephone signal issued by the caller in about 30 seconds to 1 minute after the start of the call. Therefore, after that, if both parties who communicate only the identification information of the time piece signal as the call signal are transmitted, the original time piece signal is retrieved from the memory circuit using the identification information after reception, and this is further transmitted. Once assembled, the original telephone signal can be reproduced. As a result, the amount of information to be transmitted is drastically reduced, the effective utilization of frequency is improved, the transmission power can be reduced, and the system operation becomes easy.

FIGS. 18 and 19 show a concrete circuit for creating a new time piece signal from continuous time piece signals.
This will be described. 18 and 19 are different from FIG. 11 in that the output of the telephone signal reproduction circuit 163 is input to the reception circuit 106 of the telephone section 101, and the reception telephone signal storage circuit 151-1 and the reception telephone signal storage circuit 1 are provided.
51-2, the output of the transmission circuit 107 of the telephone unit 101 is input to the time piece signal generation circuit 169, and the transmission telephone signal storage circuit 191-1 and the transmission telephone signal storage circuit 191- This is the point input to 2.

First, the telephone signal transmitted from the wireless base station 30 is shown in FIG. 16 (a), and the mobile wireless device 100 receives this signal and changes to the state shown in FIGS. 16 (b), (c) and (d). A method will be described in which time piece signals having different disconnection timings as shown in the drawing are created and different identification information is added to the time piece signals.

In this case, the mobile wireless device 10 shown in FIG.
0 speed restoration circuit 138B and speed conversion circuit 131
In the configuration of B, the received telephone signal storage circuits 151-1 and 1
Of the 51-2, only the former is sufficient and the received telephone signal storage circuit 151-2 is not necessary. Further, the reception time piece signal creation circuit 152-1 and the reception time piece signal creation circuit 1
The former of 52-2 is sufficient, and the delay circuit 153
-1, delay circuit 153-2, received signal synthesis circuit 15
4 is not particularly necessary, and it may be considered that the input / output circuit is simply connected by a line.

When the output of the telephone signal reproducing circuit 163 is input to the reception telephone signal storage circuit 151-1, the telephone signal is temporarily stored here, but sequentially to the reception time piece signal generation circuit 152-1. Sent. The reception time piece signal creation circuit 152-1 creates the time piece signal, and the timing thereof is controlled by the control signal from the control unit 140.
The control unit 140 has previously received an instruction from the wireless base station 30 regarding the new time piece signal generation timing, and transmits the timing according to the instruction to the reception time piece signal generation circuit 152-1. As a result, the reception time piece signal creation circuit 152-1 creates a time piece signal as requested by the radio base station 30, and the output thereof is sent to the reception time piece signal storage circuit 155 and stored therein. The output of the reception time piece signal storage circuit 155 is the time piece signal group storage / comparison circuit 18
1 is transmitted and is stored if there is no identical signal group stored.

The operation of the radio base station 30 on the transmitting side will be described. The transmission unit of the radio base station 30 has a similar configuration to that of the speed conversion circuit 131B of FIG. 19, so the case of transmission by the mobile radio device 100 will be described below. In this case, the telephone signal output from the transmission circuit 107 of the telephone unit 101 is input to the speed conversion circuit 131B, part of which is sent to the time piece signal generation circuit 169, and the other part of which is sent telephone signal storage circuit 191-.
Input to 1. In this case, the transmission telephone signal storage circuit 19
1-1 and the transmission telephone signal storage circuit 191-2,
The former is sufficient and the transmission telephone signal storage circuit 191-2
Is not necessary. Also, the transmission time piece signal creation circuit 192-
The same applies to the 1 and transmission time piece signal generation circuit 192-2. Furthermore, the delay circuit 193-1, the delay circuit 193-2, and the transmission signal synthesis circuit 194 are not particularly necessary, and it may be considered that the input / output circuits are simply connected by the line.

When the output from the transmission circuit 107 is input to the transmission telephone signal storage circuit 191-1, the telephone signal is temporarily stored here, but sequentially to the transmission time piece signal generation circuit 192-1. Sent. Transmission time piece signal creation circuit 19
In 2-1 the time piece signal is created, but its timing is controlled by the control signal from the control unit 140. Thus, the obtained time piece signal is the transmission time piece signal storage circuit 1
After being sent to 95 and stored, its output is transmitted to the time piece signal group storage / comparison circuit 181 and stored if there is no identical signal group to be stored. Next, in the wireless base station 30 and the mobile wireless device 100 as described above,
It will be described that the newly created time piece signal and its identification information are exactly the same.

In FIG. 19, the transmission time piece signal generation circuit 192-1 and the reception time piece signal generation circuit of the radio base station 30 which has the same function as this are also controlled by the control section 140 on the signal transmission side. . In this case, the control unit 140 prepares a certain law in advance regarding the timing of creating a new time piece signal and the method of adding the identification information, and applies this law to both mobile radio devices 100 that communicate with each other. . Therefore, the mobile wireless device 1 that communicates oppositely
In 00, signals to which the same identification information is added at the same time timing from the same telephone signal are stored in the time piece signal group storage / comparison circuit.

Next, a method of creating the time piece signal of FIG. 17 will be described. In this case, all the circuits shown in FIG. 18 and FIG. 19 need to exhibit the functions as given respectively. First, the case of the mobile wireless device 100 that receives a signal from the wireless base station 30 will be described. In this case, the signals stored in the reception telephone signal storage circuit 151-1 and the reception telephone signal storage circuit 151-2 are sequentially input to the reception telephone signal generation circuit 152-1 and the reception telephone signal generation circuit 152-2, respectively. It Of these, the reception telephone signal creation circuit 152-1 creates a time piece signal at the timing instructed by the control signal from the control unit 140, and the time length of the time piece signal at this time is half the normal length. 17 (a) is a portion with a rough diagonal line to the upper right (half the length of the signal [00001] of FIG. 17 (a)).

On the other hand, the reception telephone signal creation circuit 152-2 also creates the time piece signal at the timing instructed by the control signal from the control section 140, and the time length of this time piece signal is 1/2 the normal length. 17 (a) is a part indicated by a rough diagonal line on the right-hand side (signal [0000 of FIG. 17 (a)].
2] 1/2 length).

The above time piece signals are respectively sent to the delay circuit 15
3-1 and the delay circuit 153-2. These delay circuits 153-1 and 153-2 give signals a delay of an amount of time designated by the control signal from the control unit 140. This delay amount is calculated based on the signal [20010] in FIG.
[0255] is created. Therefore, the outputs of the delay circuit 153-1 and the delay circuit 153-2 are input to the reception signal synthesizing circuit 154, and this output is 1/2 of the above.
2 pieces of the time piece signal of the length are concatenated in series in time to become the time piece signal of the normal length, and FIG.
[200100255], a new time piece signal is created from a telephone signal existing in a temporally separated portion.

The new time piece signal created by such a process is temporarily stored in the reception time piece signal storage circuit 155 and then transmitted to the time piece signal group storage / comparison circuit 181.
If the stored signal groups are not the same, they are stored. In this case as well, the law regarding the creation of a new time piece signal is that the control unit on the signal transmitting side has the instruction, and the same time piece signal is stored in the same time piece signal storage / comparison circuit for transmission and reception. The identification information is given and stored.

Further, a method of creating the time piece signal of FIG. 17C will be described. In this case, there is not much difference from that described with reference to FIG. However, the delay amount given to either the delay circuit 153-1 or the delay circuit 153-2 is shown in FIG.
The difference is that it is larger than in the case of (b). For example, in order to create the signal [300200155] of FIG. 17C, it is necessary to increase the delay amount given to the time piece signal [00001] three times as much as when the signal [200100255] of FIG. 17B is created. .

In the above description, a new time piece signal is created from two time piece signals. Actually, in the circuit configuration of the speed restoration circuit 138B shown in FIG. 18, three or more series of reception time piece signal generation circuits 152-1 and the like to be inputted to the reception signal synthesis circuit 154 (in the case of two series in FIG. The number of new time piece signals as described below can be created in a short period of time, and the time required for communicating only the identification information can be further shortened. Create new time piece signal from three or more different time piece signals (time series is forward direction) Create new time piece signal with time series reversed from three or more different time piece signals New time combining and However, in the speed conversion circuit 131B on the transmission side shown in FIG. 19 as well, the transmission time one signal generation circuit 192-1 and the like are shown in FIG.
Naturally, it is necessary to increase the number to 3 or more (in the case of 2 in FIG. 19 as an example) in accordance with the increase in the number in 8 in FIG.

When a new time piece signal is created from two or more time piece signals, it is also possible to change the level of the original signal and then create the new time piece signal. In this case, the occurrence of noise can be reduced or neglected by tapering the change in the level of the time piece signal and adjusting the amplitude at the combining point.

(3.5.2) When Compressing or Expanding Telephone Signal to Create New Time Piece Signal Next, the received (transmitted) telephone signal is compressed or expanded to create a new time piece signal. The method will be described with reference to FIG. In the same figure, the time piece signal length and the signal number (code number) of the original time piece signal are the same as those in FIG. 17, and the explanations already given in FIG. 17 are omitted.

A method of creating the new time piece signal [400100255] of FIG. 20B from the two adjacent time piece signals [00001] and [00002] of FIG. 20A will be described. A new time piece signal such as "ba" or "pa" is created. For example, from the time piece signals “ba”, “pa”, etc. obtained on the time axis t = t ₀ , another time axis t = t
This is a method to obtain a time piece signal starting from ₁ .

In this method, the time piece signal [000
01] in the first half and select this and the time piece signal [0000
2] and almost the first half. Time piece signal [0000
1] does not perform time compression, and the time piece signal [00002] performs time compression. That is, the time piece signal [00002]
The 5.03 ms portion of is compressed to 5 ms by an average of about 1%. However, in this case, the time compression rate is 1 at the head portion (first portion) of the signal, that is, the time compression is not performed, and then the time compression is sequentially added to the tail portion of the signal (the last portion). (Part), the time compression rate is set to about 1.02. As a result, the time compression rate becomes about 1.01.

As described above, the reason why the time compression rate is inclined is that the formant frequency is sequentially increased (shifted to a higher frequency band) as shown in FIG. After performing the above operation, by connecting these two signals,
A new time piece signal [400100255] shown in FIG.
Is created.

Next, from the beginning of the above "ba", "b"
Regarding the case of creating the ending of the word “a”, a specific hardware configuration will be described with reference to FIGS. 21 and 22.

The configuration shown in FIG. 21 has the reception telephone signal storage circuits 151-1 and 151-2, reception time piece signal generation circuits 152-1 and 152-2, and delay circuit 153- of FIG.
1, 153-2 and the received signal synthesis circuit 154 are replaced with circuits of the received signal waveform measuring device 250, switches 251-1 and 251-2, a high-pass filter 252-1, and a low-pass filter. A filter 252-2 and received signal compression / expansion circuits 253-1 and 253-2 are added.

The configuration shown in FIG. 22 has the transmission telephone signal storage circuits 191-1 and 191-2, the transmission time piece signal generation circuits 192-1 and 192-2, and the delay circuit 193- of FIG.
1, 193-2 and transmission signal synthesizing circuit 194. The transmission signal waveform measuring instrument 290, switches 291-1 and 291-2, high-pass filter 292-1, low-pass filter. A filter 292-2 and transmission signal compression / expansion circuits 293-1 and 293-2 are added.

The configurations of FIGS. 21 and 22 are different from the configurations of FIGS. 18 and 19 in that they respectively operate as follows.

Received signal waveform measuring device 250: Measures the waveform of the signal stored in the received telephone signal storage circuit 151-1 or 151-2. When the amplitude of the waveform is large, the change is small If the waveform amplitude is small and the change is large, it is judged to be the beginning or end of the waveform.If the waveform amplitude is small and the change is similar to the waveform already stored as a pattern, It is determined that the sound is a consonant, and the result is reported to the control unit 140. Further, when it is determined from the result of measuring the waveform of the signal stored in the reception telephone signal storage circuit 151-1 or 151-2 that the consonant is the beginning of the word or the time piece signal in the word, the consonant is “b”,
In the case of "p" etc., the switches 251-1 and 251-2
To the terminal b side, and the output of the reception telephone signal storage circuit 151-1 or 151-2 is directly input to the reception time piece signal generation circuit 152-1 or 152-2. Also,
When the consonant is "d", "g", "t", "k", "n", etc., the switches 251-1 and 251-2 are connected to the terminal a.
The output of the reception telephone signal storage circuit 151-1 is turned to the side and the output of the high-pass filter 252-1 and the reception telephone signal storage circuit 151 is output.
-2 output is passed through the low pass filter 252-2 and then the reception time piece signal generation circuit 152-1 or 152-
It is reported to the control unit 140 that the input is made to 2. Further, the reception time piece signal generation circuit 152-1 or 152
In step -2, the waveform of the new time piece signal being created and the waveform of the received signal synthesizing circuit 154 are measured, and it is reported to the control unit 140 whether the newly created time piece signal can be used as the time piece signal.

Transmission signal waveform measuring instrument 290 ... The same operation as the reception signal waveform measuring instrument 250 will be carried out for the telephone signal transmitted from the mobile radio device 100.

Switches 251-1 and 251-2 ... Switch 251-in response to a control signal from control section 140.
The operation of tilting the terminals 151-2 to the a side or the b side is executed. Switches 291-1 and 291-2 ... Switch 251
-1 and 251-2 perform the same operation as the mobile wireless device 100.
Carry out the telephone signal transmitted from.

High-pass filter 252-1 ... Passes only the high-frequency component of the signal components of the telephone signal and filters the low-frequency component. Low-pass filter 252-2 ... Passes only the high-frequency component of the signal components of the telephone signal and filters the high-frequency component. High-pass filter 292-1 ... High-pass filter 252-1
Is the same as. Low-pass filter 292-2 ... Low-pass filter 252-2
Is the same as.

Reception signal compression / expansion circuit 253-1 ... Has a characteristic of temporally compressing or expanding the signal from the reception time piece signal generation circuit 152-1. Reception signal compression / expansion circuit 253-2 ... Has a characteristic of temporally compressing or expanding the signal from the reception time piece signal generation circuit 152-2. Transmission signal compression / expansion circuit 293-1 ... Has a characteristic of temporally compressing or expanding the signal from the transmission time piece signal generation circuit 192-1. Transmission signal compression / expansion circuit 293-2 ... Has a characteristic of temporally compressing or expanding the signal from the transmission time piece signal generation circuit 192-2.

Now, the procedure for creating the new time piece signal [400100255] described with reference to FIG. 20 will be described using the circuit configurations of FIGS. 18 and 19 replaced with the partial configurations of FIGS. 21 and 22. In this case, the terminals of the switches 251-1 and 251-2 are tilted to the b side, and the output of the reception telephone signal storage circuit 151-1 or 151-2 directly outputs the reception time piece signal generation circuit 152-1 or 152
-2 is input. Next, the reception time piece signal generation circuit 15
In 2-1 or 152-2, the telephone signal is the control unit 1
The signal is disconnected at the time timing instructed by 40 and converted into a time piece signal. This output is the received signal compression / expansion circuit 253.
-1 or 253-2.

If the signal generated by the reception time piece signal generation circuit 152-1 is [00001], the reception signal compression /
The decompression circuit 253-1 outputs the signal as it is without receiving signal compression / decompression.
-2, the signal created in -2 is [00002], received signal compression / expansion circuit 253-2 receives the signal, and then output. Delay signals 153-1 and 153, respectively.
These two signals are concatenated in series in time by the reception signal synthesizing circuit 154 after shifting the timing after -2 to become one new time piece signal [400100255]. By this operation, "b" near the beginning of the word "b"
a ”is obtained.

By performing the same signal processing as described above using two or more time piece signals further apart in time, from the beginning of "ba" to the beginning of all kinds of "ba", It is possible to create in-words or endings. However, in the case of creating a new time piece signal using three or more time piece signals, two series of constituent elements such as the reception telephone signal storage circuits 151-1 and 151-2 shown in FIGS. It is necessary to have three or more lines.

(3.5.3) When a frequency component of a telephone signal is divided to create a new time piece signal: In addition to compressing or expanding the next received (transmitted) telephone signal, the frequency of the telephone signal Split the components,
A method of creating a new time piece signal will be described with reference to FIG. In FIG. 23, the time piece signal length and the number (code number) of the original time piece signal are the same as those in FIG. 17, and the description that has already been given in FIG. 17 is omitted.

Now, from the two adjacent time piece signals [00001] and [00002] of FIG.
A method of creating the new time piece signal [500100255] in (b) will be described. New time piece signal [50010
0255] are already obtained “da”, “g
"da", "ga", "ta", "k" that have not yet been obtained by using time piece signals such as "a", "ta", and "ka"
This is a method of obtaining a new time piece signal such as a ”. For example, the time piece signal“ d ”obtained on the time axis t = t ₀ in FIG.
a ”,“ ga ”,“ ta ”,“ ka ”, etc., a new time piece signal“ da ”starting from another time axis t = t ₁ ,
This is a method of obtaining “ga”, “ta”, “ka”, and the like. As is clear from FIG. 30, "da", "ga", "t"
In "a", "ka", etc., the first formant shifts to the higher frequency side and the second formant shifts to the lower frequency side with time, so the frequency of the time piece signal is inevitably separated into the high and low frequencies. Above, it is necessary to perform separate processing such as time compression / expansion.

In this case, a desired new time piece signal can be obtained by performing the following operation. The output of the reception telephone signal storage circuit 151-1 passes through the high-pass filter 252-1 and inputs only the second formant signal to the reception time piece signal generation circuit 152-1. The output of the reception telephone signal storage circuit 151-2 passes through the low pass filter 252-2 and inputs only the first formant signal to the reception time piece signal generation circuit 152-2. The output of the reception time piece signal generation circuit 152-1 is input to the reception signal compression / expansion circuit 253-1, where time expansion is performed by about 1% and the result is sent to the reception signal synthesis circuit 154 (delay circuit 153-1). Is the same as thru in this case). The output of the reception time piece signal creation circuit 152-2 is input to the reception signal compression / expansion circuit 253-2, where time expansion of about 1% is performed and the result is sent to the reception signal synthesis circuit 154 (delay circuit 153-2. Is the same as thru in this case).

By the above signal processing, for example, "d
It is possible to create a new head of "da" from the head of "a".

Although the individual explanations have been given in the above (3.5.1) to (3.5.3), in the case of actual signal processing, these are combined and performed. Therefore, the following new signal is created. i) Creation of a received time piece signal, for example the ending of the word "ba" from the beginning of the word "ba". ii) The received time piece signal, eg "ba" to "b.
Creation of other consonants that start with the same consonant, such as "i", "bu", "be" and "bo". iii) In combination of i) and ii)
Creation of "iu", "bio", etc. iv) Creation of other consonants “pa”, “fa”, etc. from the received time piece signal, eg, “ba”. v) Combination of multiple consonants, or combination of multiple consonants in i) to iv) above to create other consonants.

As described above, by receiving a short telephone signal, it is possible to create a time piece signal which has not been obtained from the short telephone signal. Therefore, the telephone signals are exchanged for about 30 to 60 seconds at the longest. Then, only the identification information can be communicated thereafter.

21 and 22 show an example of a circuit configuration for creating a new time piece signal from a certain time piece signal and are simple examples. Therefore, although the circuit configuration becomes complicated, if the operations described in (3.5.1) to (3.5.3) above are performed more precisely and simultaneously,
A new time piece signal closer to the original telephone signal is obtained. Further, the time required for this is also shortened, and even if they are mutually long, it is possible to communicate only telephone signals for about 10 to 20 seconds, and thereafter it is possible to communicate only the identification information.

In the above-described system example of the present invention, the transmitted time piece signal is well received by the receiving side. However, in the mobile radio system, there are cases where radio signals are not properly transmitted to the receiver of the other party due to severe fading due to the influence of topographical features in the space which is the transmission medium. In such a case, time signal group memory
If the contents are stored in the comparison circuit 181 (FIG. 11 and the like), the contents of the time piece signals having the same identification information may differ between the transmitting side and the receiving side. In order to avoid this, for example, the reception signal waveform measuring device 250 shown in FIG. 21 is provided with a function of measuring the S / N of the reception signal, and when the value is equal to or less than a certain value determined by the system, the control signal is transmitted to the other side. Since the voice (telephone) signal sent to and transmitted to the device is severely deteriorated, it is necessary to request not to be stored in the mutual storage circuit.

(3.6) Effective Utilization of Time Slot without Telephone Signal As described above, when only the identification information (and the level information of the signal) is transmitted, the signal in the time slot has 15 bits. The efficiency is extremely poor if it is transmitted as it is. Therefore, effective use of time slots will be described using the previous model. As a method of effective use, there is a method in which a time slot in one frame is not fixedly provided corresponding to a telephone signal, and a dynamic time slot assignment is executed for each frame. As described above, the signal having only the identification information may be shared with the time slot for the control signal and the time slot having only the identification information (shared with other telephone channels) may be installed.

As a result of the above, empty time slots in one frame increase, and this can be utilized for batch transmission of identification information described below.

(3.7) Batch Transmission of Identification Information In the above description, in communication performed by the mobile wireless device 100 and the wireless base station 30, while transmitting a telephone signal (time signal), the identification information is transmitted to each other at the same time. It is a method of transmitting and transmitting the identification information instead of transmitting the time piece signal to the time piece signal stored by the other party. However, if the following conditions are satisfied, it is possible to more effectively transmit the identification information.

A) Communication terminals, especially mobile terminals, must be specified by the user. b) The memory capacity of the memory circuit included in the communication terminal is large. c) The communication traffic of the system is large and the communication terminal can utilize it.

Among these, in the case of a), the number of mobile terminals used per person is increasing with the development of personal communication. Or even if it is not one per person, for example, A
It suffices if the user can be specified, such as Mr. B, Mr. B, and Mr. C. However, in this case, the storage content of the storage circuit of the mobile terminal is required to be tripled. In addition, it is necessary to input its own ID from the telephone unit 101 prior to use, and to make the mobile wireless device 100 operate as if it were its own dedicated terminal. Furthermore, the number of mobile phones such as car phones, cordless phones, and mobile phones housed in fixed networks is becoming one. Further, it is not necessary for each person to be one, as long as the user can be identified. Since b) is extremely easy to obtain at the current state of the art, the conditions for a) are not so stringent. Also, c) is possible in the system of the present invention. Hereinafter, a system example will be described.

(3.7.1) System example for batch transmission of identification information Frame length: 20 ms Multiplicity: 48 (50 time slots in one frame, 2 time slots are synchronization and control signals, And used for transmitting identification information) Information amount of identification information: 10 bits (total number of time piece signals is 1000 types, but level information etc. are not included) Information quantity of time piece signal: transmitted as analog signal To do. Therefore, since there are 1000 20 ms time-series signals, the time required to transmit all of them is 0.02 × 1.
000 = 20 seconds / 1 channel

When the identification information is collectively transmitted, the voice of the caller must be stored in advance in the time piece signal group storage / comparison circuit 181 in the form of the time piece signal. To do so, it remembers the signal of the call that the caller made earlier. The caller of the telephone unit 101 is provided in advance to the caller (see FIG. 1).
0) to generate a reference speech signal (a short sentence group, but includes all necessary time piece signals), and stores this in advance in the form of time piece signals. Etc. are adopted.

Based on these and these conditions, the time required to collectively transmit the identification information in the system example of the present invention is obtained. It should be noted that the timing of batch transmission is the call state S216 (see FIG. 9).
Immediately after entering) is desirable. In the system for transmitting the identification information all at once, the time slot for a call is always an empty time slot. This is because the time slot for communication is used for a short time for batch transmission of identification information, but when it is finished, only the transmission of identification information will be done afterwards, and this will be transferred to the time slot for exclusive use of identification information. Because. When 5000 kinds of time piece signals are transmitted as analog signals using all 48 time slots, the required time is 100 seconds / 48 speech channels = 2.08 seconds. That is, it is completed in about 2 seconds. Further, if transmission is performed using 30 time slots out of 48 time slots, the required time is 100 seconds / 30 speech path = 3.33 seconds. That is, the transmission is completed in just over 3 seconds.

As described above, the time required for batch transmission of the identification information is extremely short, and is completed without the notice of the caller. Alternatively, during the 2-3 second period in which the identification information is collectively transmitted, a time slot different from the call may be assigned and transmitted in parallel with the batch transmission of the identification information. On the other hand, when such an operation is performed by the digital cordless telephone, it becomes as follows. In a digital cordless telephone, the multiplicity (the number of telephone conversions on one wireless carrier) is 4, and 5000 kinds of time piece signals are transmitted in digital form. The amount of information to be transmitted is: 32 kbps × 0.01 s × 5000 = 1600 kb When transmission is performed using all 4 multiplexing, the required time is 1600 kb / (32 kbps × 4) = 12.5 seconds. That is, it takes 12.5 seconds. In this case, the caller may feel uncomfortable that the call cannot be made (or the call does not reach the other party). Therefore, it is not possible to transmit using all 4 multiplexes,
Set the call path so that you can talk for 12.5 seconds,
Another time slot should be assigned for the transmission of the identification information. As a result, the batch transmission of the identification information has no choice but to use the parallel transmission using the 3 multiplex. In this case, the required time is further increased to 1600 kb / (32 kbps × 3) = 16.66 seconds.

In the above description, the batch transmission of the identification information is performed from the mobile wireless device 100 to the wireless base station 30, but conversely, the fixed caller is identified from the wireless base station 30 to the mobile wireless device 100. It is also possible to send the information all at once.
For that purpose, it is necessary to satisfy one of the following conditions. i) The identification information of a specific fixed-side caller is stored in the wireless base station 30. ii) Identification information of a specific fixed-side caller is stored in a storage circuit (not shown) of the gateway exchange 20. iii) The identification information of the fixed party is stored in a specific fixed telephone.

Of these conditions, i) can be used immediately, but in the case of ii), the memory circuit (not shown) of the gateway exchange 20 is accessed each time a call is made, and a specific fixed side is accessed. It is necessary to retrieve / read the identification information of the caller. Further, there is a need for a device that converts a signal into a form suitable for a signal to be placed in a wireless section. In the case of iii), in addition to ii), inter-exchange communication within the telephone network 10 is required.

As is clear from the above description, the present invention is
When applied to a CM system, the transfer of identification information is extremely easy. Although the embodiments of the present invention have been mainly focused on the system using the TCM signal, the application of the present invention is merely TCM.
Not only signals, but any system that can be generally time-divided and transmitted, such as digital cellular phones and digital cordless telephones that have recently been put into practical use, can be applied. Further, since the present invention is a basic communication method, any transmission medium may be used. That is, it is applicable to both the wireless communication system and the wired communication system.

The advantages of the system to which the present invention is applied can be summarized as follows. Although the voice uttered by the caller is not transmitted in the wireless section, the telephone signals that are completely the same as the voice uttered by the other party arrive at the terminals, so the callers can identify each other when applying the present invention. It is possible to make a phone call without being aware that only those are being sent. No new operation is required for the user to perform the. The effective utilization of frequency is greatly improved. The wireless transmission power can be reduced by the amount of information to be transmitted. Mobile radio 10 used exclusively by a specific individual
It is possible to prevent 0 from being used by a third party without permission. This is because the time piece signal of the voice possessed by a specific individual is already stored in the time piece signal group storage / comparison circuit 181, and if a time piece signal that is not stored is input, the transmission operation of the mobile wireless device 100 is performed. It becomes possible by stopping. The system according to the present invention is basically confidential communication, and is an advantageous system in terms of protection of communication privacy.

(4) Calculation of Effective Frequency Utilization by Transmission of Identification Information It is clear from Reference 4 that telephone signals have a lot of redundancy. The operation of the present invention effectively utilizes this redundancy, and embodies a communication system that has never been seen before. The frequency effective utilization rate by transmitting the identification information, which is a substitute for the time piece signal of the present invention, is calculated next as compared with the existing system.

Frame length of TCM (system model)
Is 20 ms, it is possible to create an arbitrary audio signal from a voice element (in the case of the present invention, a time piece signal) having a time length of 20 ms (as described in (c) of (3.3.2),
The required number of voice elements, which is assumed to be a slight difference in sample values, that is, it is assumed that noise is included, is assumed to be about 5000, as already described in equation (8). Therefore, the type of identification information is 5000, that is, about 13 bits. Since 8 bits are required for this as level information, the amount of information required per second is I _TCM = 21b / 0.02 (s) = 1050b. On the other hand, the current digital cordless telephone requires 1 second. Since the amount of information I _DCT is I _DCT = 32 kbps, I _TCM / I _DCT = 1050/32000 = 0.033 Therefore, when only the identification information is transmitted, it is 0 compared with the digital cordless telephone in use. 0.033, that is, 1/30 is sufficient. From the viewpoint of effective frequency utilization, this means that the effective frequency utilization is 30 times.

The above is the comparison with the working digital cordless telephone, but in comparison with the working digital cellular telephone, in this case, the information amount I _DCS required for one second is I _DCS = 10 kbps. _TCM / I _DCS = 1050/10000 = 0.105 Therefore, when only the identification information is transmitted, it is 0.105, that is, compared with the current digital cordless telephone.
About 1/10 is enough. In any case, it became clear that the effective frequency utilization is improved by one digit or more.
At the same time, the transmission power can be reduced to 1/10.

(5) Installation of time-series signal group storage / comparison circuit to gateway switch The transmission / reception of only the identification information, which has already been described in (3) and (4), takes time between the mobile radio 100 and the radio base station 30. This was done by installing a single signal group memory / comparison circuit. However, the operation of the present invention becomes more effective by installing the time-series signal group storage / comparison circuit also in the gateway switch 20, for the following reason.

The mobile wireless device 100 may move to any place within the service area managed by the gateway switch 20. There are many wireless base stations 30 in the service area managed by the gateway switch 20. When the mobile wireless device 100 moves within the service area as described above, the wireless base station 30 that communicates opposite to the mobile wireless device 100 performs the "during-call zone switching" to the adjacent wireless base station to move. It becomes necessary to change the wireless base station 30 that communicates with the wireless device 100. In this case, the adjacent radio base station does not hold the time piece signal and its identification information which are commonly managed by the mobile radio device 100. Therefore, in order to continue the communication of only the identification information, the information stored in the time piece signal group storage / comparison circuit of the radio base station 30 must be transferred to the adjacent radio base station by some method. I won't. The gateway exchange 20 is also provided with a circuit having the same function as the time piece signal group storage / comparison circuit 181 and the identification information transmission / reception circuit 182 of the mobile wireless device 100. Then, in this case, the mobile wireless device 100 and the wireless base station 30 start the operation of sharing the time piece signal and its identification information at the start of the call, but at the same time, the same time piece signal to be shared and its identification information are started. Is also transferred to the gateway switch 20 and stored therein.

In such a state, when the identification information or the like has to be transferred to the adjacent radio base station as described above, the gateway exchange 20 moves to the destination of the mobile radio 100,
That is, it is in a position to easily know the wireless base station to make a new call, and if the identification information or the like is transferred to the wireless base station, it becomes possible to easily continue the communication of only the identification information.

In FIG. 24, the gateway switch 20 and the mobile radio 1
00 time piece signal group storage / comparison circuit 181 and identification information transmission / reception circuit 182, and a time piece signal group storage / comparison circuit 23 and identification information transmission / reception circuit 2
4 shows the configuration in the case of including 4.

[0240]

As is apparent from the above description, in the time division communication system using the telephone signal, the identification information is transmitted to the partner terminal prior to the call as described in detail in the text.
By transmitting only the identification information instead of the telephone signal,
The effective utilization of frequency is greatly improved. Also, even if both parties to communicate do not have the time piece signal or its identification information in advance, as already explained, it is possible to create a large number of new time piece signals from the transmitted call signal, After the call is started, a time piece signal group capable of assembling an arbitrary telephone signal can be provided immediately. Moreover, although the voice uttered by the caller is not transmitted in the wireless section,
Since a telephone signal that is completely the same as the voice uttered by the other party arrives at each other's terminal, the callers can make a call without being aware that only the identification information when the present invention is applied is sent. It is possible. It is also possible to use the reduction of the amount of information to be transmitted to reduce the transmission power. further,
Even if the present invention is applied to the wired transmission system, the same effect as the wireless transmission system can be exhibited. Therefore, the effect of the present invention is extremely large.

[Brief description of drawings]

FIG. 1 is a conceptual configuration diagram showing a concept of a system of the present invention.

FIG. 2 is a circuit configuration diagram for explaining a basic operation of a mobile wireless device used in the system of the present invention.

FIG. 3 is a circuit configuration diagram for explaining a basic operation of a radio base station used in the system of the present invention.

FIG. 4 is a time slot structure diagram for explaining a basic operation used in the system of the present invention.

FIG. 5 is a spectrum diagram showing spectra of a call signal and a control signal.

FIG. 6 is a circuit configuration diagram for multiplexing a voice signal and a data signal.

FIG. 7 is a spectrum diagram showing spectra of a call signal and a control signal.

FIG. 8 is a flow chart showing a flow of a calling operation of the system according to the present invention.

FIG. 9 is a flow chart showing the flow of a calling operation of the system according to the present invention together with FIG.

FIG. 10 is a type diagram showing the types of time piece signals when time piece signals are created every 0.25 ms.

11 is a circuit configuration diagram of a speed restoration circuit, a telephone unit, and a speed conversion circuit, which are the constituent elements of FIG.

FIG. 12 is a time piece signal group storage, which is a component of FIG.
It is a circuit block diagram which shows the internal structure of a comparison circuit and an identification information transmission / reception circuit.

FIG. 13 is a flow chart showing a flow of an operation of obtaining identification information by a time piece signal comparison operation.

FIG. 14 is a flow chart showing a first sample comparison sub-routine in FIG.

FIG. 15 is a type diagram showing types of time piece signals when a time piece signal is created every 1 ms.

FIG. 16 is a time piece diagram showing a method for creating a new time piece signal from adjacent time piece signals.

FIG. 17 is a time piece diagram showing a method of creating a new time piece signal from time piece signals that are not adjacent to each other.

FIG. 18 is a circuit configuration diagram showing a specific portion of a speed restoration circuit that creates a new time piece signal from continuous time piece signals.

FIG. 19 is a circuit configuration diagram showing a specific portion of a speed conversion circuit that creates a new time piece signal from continuous time piece signals.

FIG. 20 is a time piece diagram showing a method of compressing or expanding a received or transmitted telephone signal to create a new time piece signal.

FIG. 21 is a circuit configuration diagram showing a replacement portion of the speed restoration circuit shown in FIG. 18 capable of creating a word middle and word ending from a word head.

22 is a circuit configuration diagram showing a replacement portion of the speed conversion circuit shown in FIG. 19 capable of creating a middle word and a last word from the beginning of a word.

FIG. 23 is a time piece diagram showing a method of dividing a frequency component of a telephone signal to create a new time piece signal.

FIG. 24 is a conceptual configuration diagram of a system in the case where the gateway switch is equipped with a time piece signal group storage / comparison circuit and an identification information transmission / reception circuit.

FIG. 25 is a waveform diagram of vowels.

FIG. 26 is a waveform diagram of a consonant and a vowel (cho).

FIG. 27 is a characteristic diagram showing information transmission amount characteristics when peaks are removed from a telephone signal and when the vicinity of 0 is removed.

28 is a waveform diagram showing waveforms in the case of removing peaks and in the case of removing near zero in FIG.

FIG. 29 is a phoneme diagram showing phoneme types and their durations.

FIG. 30 is a formant frequency diagram showing a correspondence relationship between a consonant articulation form and a frequency spectrum.

[Explanation of symbols]

 10 telephone network 20 barrier exchange 22-1 to 22-n communication signal 23 hour piece signal group storage / comparison circuit 24 identification information transmission / reception circuit 30 radio base station 31 signal processing unit 32 radio transmission circuit 35 radio reception circuit 38 signal speed restoration circuit Group 38-1 to 38-n Signal speed restoration circuit 39 Signal selection circuit group 39-1 to 39-n Signal selection circuit 40 Control unit 41 Clock generator 42 Timing generation circuit 51 Signal speed conversion circuit group 51-1 to 51- n signal speed conversion circuit 52 signal allocation circuit group 52-1 to 52-n signal allocation circuit 91 digital encoding circuit 92 multiplex conversion circuit 100, 100-1 to 100-n mobile radio device 101 telephone unit 120 reference crystal oscillator 121- 1, 121-2 Synthesizer 122-1, 122-2 Switch 123 Transmission / reception gating controller 131, 31B speed conversion circuit 132 radio transmission circuit 133 transmission mixer 134 transmission unit 135 radio reception circuit 136 reception mixer 137 reception unit 138, 138B speed restoration circuit 140 control unit 141 clock regenerator 151-1, 151-2 reception telephone signal storage circuit 152 -1,152-2 Reception time piece signal creation circuit 153-1, 153-2 Delay circuit 154 Reception signal composition circuit 155 Reception time piece signal storage circuit 161 Telephone signal storage circuit 162 Time extension circuit 163 Telephone signal reproduction circuit 165,167 Switch 169 Hours signal generation circuit 171 Telephone signal storage circuit 173 Time compression circuit 177 Switch 181 Hours signal group storage / comparison circuit 182 Identification information transmission / reception circuit 191-1, 191-2 Transmission telephone signal storage circuit 192-1, 192- 2 Transmission time One-sided Creation circuit 193-1, 193-2 Delay circuit 194 Transmission signal composition circuit 195 Transmission time piece signal storage circuit 250 Reception signal waveform measuring instrument 251-1, 251-2 Switch 252-1 High pass filter 252-2 Low range Pass filter 253-1, 253-2 Received signal compression / expansion circuit 290 Transmit signal waveform measuring instrument 291-1, 291-2 Switch 292-1 High-pass filter 292-2 Low-pass filter 293-1 293-2 Transmission signal compression / expansion circuit 501 Signal comparison circuit 502, 503 Storage circuit 521 Identification information creation circuit 522 Identification information transmission circuit 523 Identification information reception circuit 524 Identification information analysis circuit

Claims (2)

[Claims] 1. Radio base means (30) for covering a plurality of zones respectively to form a service area, and mobile radio means for moving across the plurality of zones by the radio base means and radio channels. (100) using a gateway exchange means (20) for exchanging communication with the mobile radio means and a general telephone network connected to the mobile radio means and the gateway exchange means ( 10)
In order to enable at least one of the communications with the telephone means included in the above, both parties of the said communication identify the voice of the person making the communication into a time fragmented signal. Information is stored as a storage time piece signal, and in the communication, the telephone signal to be transmitted is temporally subdivided into a transmission time piece signal), and the storage of the identification information is shared by both parties. When the judgment is made, only the identification information is transmitted, and when the reception side receives the identification information, the stored time piece signal stored with the same identification information as this identification information is read out to be a telephone signal transmitting method. In the case where the telephone signal is subdivided in time and transmitted as a transmission time piece signal, both parties performing the communication perform a time compression / expansion of the transmission time piece signal to share a new time piece signal group. do it Create a meta process, the call signal transmission and reception method in a mobile communication storing denoted by the identification information, respectively. 2. A radio base means (30) which covers a plurality of zones to form a service area, and a mobile radio means which moves across the plurality of zones and uses the radio base means and radio channels. (100) using a gateway exchange means (20) for exchanging communication with the mobile radio means and a general telephone network connected to the mobile radio means and the gateway exchange means ( 10)
In order to enable at least one of the communication with the telephone means included in the above, both parties of the communication identify the voice of the person making the communication into a time-divided time piece signal. Information is stored as a storage time piece signal, and in the communication, the telephone signal to be transmitted is temporally subdivided into a transmission time piece signal), and the storage of the identification information is shared by both parties. When the judgment is made, only the identification information is transmitted, and when the reception side receives the identification information, the stored time piece signal stored with the same identification information as this identification information is read out to be a telephone signal transmitting method. In the case where the telephone signal is subdivided in time and transmitted as a transmission time piece signal, the two components performing the communication divide the frequency component of this transmission time piece signal and divide it from the frequency components of other time piece signals. Got Create a way that both the new time piece signal group by combining the wavenumber component determined in common, the call signal transmission and reception method in a mobile communication storing denoted by the identification information, respectively.