Method for the interference-resistant transmission of communications

(74) Assignee Onsagers Patentkontor AS, Oslo

<td>(54) Designation of the invention</td><td>PROCEDURE FOR RISK-RESISTANT TRANSFER OF INFORMATION.</td>

(57) Summary

The present invention relates to a method for noise-resistant transmission of voice signals processed via vocoder, over radio using scattered frequency bands (SSMA) or rapid radio frequency shift. In view of the danger of intrinsic noise this imposes when very many stations use the same frequency band or the same radio frequency channel collective simultaneously, it is proposed with the system internal frame rate (RT) to activate the transmitter (SR) and receiver (ER) during a frame rate period only when also a utility information occurs. In this way it is possible, on average, to save about 40% and more than 60% of the transmission time, respectively, on alternate voice connections and on duplex operation, thereby reducing the system noise in a corresponding way.

<img file="NO113579C_D0001.tif" />

(56) Cited publications

U.S. Patent No. 4057690.

<img file="NO113579C_D0002.tif" />

BACKGROUND OF THE INVENTION 1. Field of the Invention The invention relates to a method for noise-resistant transmission of voice signals processed via vocoder, over radio using scattered frequency bands or by rapid radio frequency shift.

For the use of tactical radios in the military field, a great resistance to intentional interference of any kind is crucial. For this, one must make use of transmission methods where the transmission signals can not be so easily influenced by strong noise transmitters. Such a transfer method is provided e.g. at the disposal of the so-called SSMA technique. This term is derived from the English term "Spect Spectrum Multiple Access Modulation". One application of this method is e.g. reported in DE-PS 2 121 117. Here, the utility signal to be transmitted is strongly interrupted in its frequency band by means of a marking modulation, and this interruption is again canceled on the receiving side by means of a correlator. As a result, strong noise signals in the form of some spectral lines are considerably attenuated relative to the utility signal. Another method is to make use of a very rapid change of radio frequency during transmission, so a powerful noise transmitter will have difficulty in tuning sufficiently quickly on the radio frequency channel just used. Radio systems that use a radio frequency jump method are e.g. known from US-A-4,037,159. The noise resistance of signals to be transmitted is the greater the smaller the width of the signal band of the signal can be measured relative to the transmission bandwidth. For this reason, it is advisable to increase the noise resistance of voice transmission by using vocoders in addition to the described method techniques. The speech signal is then analyzed on the sending side by means of a vocoder analyzer for its speech-specific parameters, and only these are then transmitted in coded form to the receiving side. On the receiving side, the original speech signal is then synthesized in a vocoder synthesizer by means of these parameters in conjunction with a pulse-shaped and a noise-shaped spur function. In addition to reducing the width of the utility band, the encoded form of the signal to be transmitted simultaneously lends itself to a cipher that provides a high degree of secrecy, and which by

1 3579 military applications are usually required in addition to the noise resistance already mentioned.

According to what practice shows, the mentioned methods for noise-resistant transfer of information sometimes give rise to self-disruption. When using the SSMA technique, many stations use the same frequency band at the same time. For a particular receiver, which, with its associated correlator, receives only a signal determined by the marking modulation as a utility signal, all the utility signals transmitted simultaneously to other stations are expressed as noise or noise signals at the receiver's input. In other words, the number of stations in operation at the same time, referring to the predetermined frequency band, must remain limited to such a target that a receiver will still be able to receive its assigned signal impeccably from the noise.

The same applies when using a rapid radio frequency shift. Here one must assume that there is in total only a limited number of radio frequencies in the order of approx. 500 available. Of these 500 radio frequencies, for example. a selection of about 120 radio frequency channels is made available to one of several radio circuits. For many reasons, an overall control of several radio circuits with the aim of realizing a synchronous jumping method is not possible. Due to the available frequencies, it is also not possible to assign a reserved channel bundle to each radio circuit. Therefore, since the radio circuits do not run synchronously in the jump phase or in the frequency program, there must therefore be a computable probability of variable width overlaps between the information blocks of the same frequency emitted by the different radios for different radio circuits.

In practice, the already mentioned redundancy-reducing speech transmission using a vocoder provides an acceptable speech quality when less than 2% of all transmission blocks are disturbed, if the block length is sufficiently small and the speech synthesizer becomes aware of the disturbance. At a collective of 128 radio burst frequencies, this amount of noise would be exceeded already by simultaneous operation of more than 3 radio circuits within the interfering field strength range.

It is the object of the invention to provide a method of introducing a new solution whereby the intrinsic disturbance is substantially reduced, and thus, by providing a transmitted frequency band or collective of radio frequency channels, it is possible, if necessary, to obtain a substantially increase the permissible number of stations currently operating simultaneously.

According to the invention, this task is solved by using this method for noise-resistant transmission of voice-processed voice signals over radio using scattered frequency bands (SSMA) or a fast radio frequency shift (FH) in the pre-processing of voice signals to be transmitted. a determination of the voice signal criterion by means of which, referring to the system internal frame rate, the transmitter (SR) is switched on or off, the transmitter stage by the occurring and non-occurring signal, respectively, in the associated time element, that the referenced to the system internal frame rate, for its transmission to the receiving side, digitized signal sections, optionally sectionally provided with a data security information for error determination, and that the signal criterion (sc¹) on the receiving side at the decoding of the arriving signal sections is recovered, optionally from the data securing information and, referring to the frame period, is interpreted to enable or disable the vocoder synthesizer (VOE) on the receiver's output side depending on a present and not present signal.

It is admittedly already known from US-PS 4 057 690 by means of a noise suppressing link to activate a transmission line only when a utility signal is detected. However, no special noise-resistant voice transmission is used here. Nor can this publication provide any incentive as to how the receiver in a radio connection section should be appropriately switched on only when the transmitter is also working, and this taking into account the difference in time between transmitter and receiver.

The basis of the invention is the recognition that, on average, approx. 40% of the transmission time is unnecessary, even in exchange-rate connections, even if small pieces (from about 20 rn length) are cut out. For duplex connections, the transmission time that can be saved can, under these conditions, even exceed 60%. This saving benefits the noise from the noise. Thus, in simplex operation, almost twice as many drives and in duplex operation nearly three times as many stations in the same radio field, if the remaining error rate still allowed by known systems is to be observed for this. It is also very advantageous in this connection that this saving of transmission time also manifests itself in the form of an increase of the ECM resistance, since the receivers here are also only switched on when there is a signal in the time section given in advance in the internal frame rate.

Using the rapid change of radio frequency, for optimum full utilization of the noise reduction possible in the solution according to the connection, it is necessary to make the length of the frequency jump interval exactly equal to 1, 2, .... n frame periods of the vocoder and synchronize it to frame rate resp, on a multiple of this.

In order to prevent the vocoder synthesizer on the receiving side from being disconnected when the message no signal is caused by a disturbance of the receiver, it is appropriate to interpret the reception side recovered signal criterion in a signal evaluator, if another input receives the received signal. The signal evaluator issues a control signal for disabling the vocoder synthesizer only when no signal is detected at the time of a signal criterion due to noise.

BRIEF DESCRIPTION OF THE DRAWINGS The invention will now be described in more detail by way of example examples shown in the drawing.

Fig. 1 is a block diagram of the transmitting and receiving side of a radio station working with the enabling and disabling of the transmitter and / or receiver according to the invention and is part of a radio system using the frequency jump technique.

Fig. 2 is a block diagram of the transmitting and receiving side of a radio station working with the enabling and disabling of transmitter and / or receiver according to the invention and included in a radio system using the SSMA technique.

FIG. 3 is a more detailed block diagram of a transmit vocoder analyzer provided in FIG. 1 and 2, and FIG. 4 is a more detailed block diagram of the reception-side vocoder synthesizer provided in FIG. 1 oa 2.

The sending side of the block diagram of a radio station shown in FIG. 1, the vocoder analyzer exhibits VOS, which at its input side is controlled by the output of a microphone Mi. The Vocoder analyzer VOS has two outputs. One output delivers the parameters that represent the speech and which in the vocoder CO for the purpose of determining transmission errors are coded redundantly and then the transmitter SR is supplied. The transmitter SR is in the embodiment of FIG. 1 equipped with a device FS for rapid frequency change. In the transmitter SR, the encoded signals are transmitted to the radio frequency plane and emitted by the antenna As.

On the receiving side, the radio frequency signals are received with the antenna Ae and, in the receiver ER, which, in a similar way as the transmitter SR, is equipped with a device FS for fast synchronous frequency change, is transmitted to the baseband plane and then examined in the decoder DE for transmission errors. The decoder DE has two outputs connected to the two inputs of the receiving side vocoder Synthesizer VOE, and at one input the vocoder synthesizer from the decoder DE receives the speech-specific parameters of the speech analyzed in the broadcast side vocoder analyzer, and by means of which the original signal is again built up by synthesis in the synthesizer and radiated through the hearing aid (telephone) H.

In accordance with the invention, the transmit vocoder analyzer VOS at its other output emits a signal criterion sc, which at all times indicates whether a reference signal is present or not available during a frame period. A signal sc is obtained in connection with the analysis and is supplied to the other input to the transmitter SR via a delay link τ, where it is used for switching on and off the transmitter depending on existing and non-existing signals. In other words, the transmitter SR in step with the internal frame rate of the vocoder analyzer VOS during operation is always switched off when there is no signal in the time interval of a frame period. In this case, the device FS for rapid frequency shift is also referred to the frame rate, namely that after a radio frequency shift is constantly made at the transition, from one signal section to another.

In order to enable on the receiving side in the same way to switch the receiver on and off depending on whether or not there is any signal, the receiver and / or the vocoder synthesizer VOE must be able to distinguish between signal and noise. To ensure this, the speech-specific sampling values ​​obtained within a frame period and representing one and one signal sections, on the transmit side at the input of the vocoder CO, in the case of a present signal, are provided with a data security information which enables the decoder DE to detect or also correct transmission errors. and output a corresponding signal via its second output as signal criterion sc 'to the second input of the vocoder synthesizer VOE. Thus, in the vocoder synthesizer VOE, signal criterion sc 'is provided for switching on and off the vocoder synthesizer.

A variant of the radio station of FIG. 1 is shown in the block diagram of FIG. 2 for a radio station in a radio network that works with SSMA technology. Here, the signal sections on the transmit side at the output of the vocoder analyzer VOS are supplied with a modulator M, where at each bit of these signal sections a marking modulation is provided by a quasi-random generator QZ. The utilization of the utility frequency band thus obtained is similarly converted on the receiving side into a modulator M post-coupled output from the receiver ER, with a marking modulation produced in an identical quasi-random generator QZ, and the signal thus obtained is supplied. and timing link AS. The interpretation link AS delivers, via its first output, the signal bits representing the transmitted utility information, to the first input of the vocoder synthesizer VOE. At the second output of the interpretive coupling, a signal occurs only when there is no signal bit and only where noise occurs. This signal is up-integrated into the integrator I for the duration of a signal section. The signal s1 that occurs at the output of integrator I is fed to a rating circuit ES, which generates the signal criterion sc 'at the input of a signal section and delivers it to the other input of the vocoder synthesizer VOE for switching on and off depending on an existing one. resp, non existing signal section.

In order to enable a better understanding of the operation of a transmitter and receiver of the transmitter and receiver in the radio stations of FIG. 1 and 2 in accordance with the invention, the diagrams of FIG. 3 and 4 in more detail the transmit-side vocoder analyzer VOS and the receive-side vocoder synthesizer VOE. The transmitter side analysis portion of the vocoder in the embodiment of FIG. 3 has at the output of the microphone Mi parallel spectral channels SK1, SK2 ... SKn. These spectral channels, by means of band filters, divide the speech band into different sub-bands, which on the output side are rectified and released via a low pass filter for higher frequency parts which are not desirable. Then, these output signals from spectral channels - also referred to as envelope curve sub-signals - are sampled and converted into coded signals in an analog-to-digital converter A / D.

In addition, for the reception-side synthesis of the speech appearing at the output of the microphone, a signal, called a pitch signal, which characterizes the speech base frequency and a signal indicating the character of the sound must be formed. In the analysis section of the vocoder according to FIG. 3, two additional channels are provided, namely a channel SGF with a speech frequency filter and a channel LB with a connection for determining the sound character. The output side signals on the channels SGF and LB are also encoded via an analog-to-digital converter A / D and together the digital outputs from the voice channels SK1 .... SKn united in the multiplexer MUX to a pulse frame with a frame period of about length. 20 ms. The signal criterion sc needed for switching on and off of transmitter SR is provided by means of a sum amplifier SU and a switched threshold value switch SW. The sum amplifier SU sums the signals that occur at the inputs of the spectral channels analog-digital converters A / D. Depending on whether the sum value exceeds the threshold for the threshold switch SW or not, the output signal of the threshold switch announces the criterion signal present and, respectively, signal not present.

The reception side synthesis part of the vocoder has on its input side an intermediate bearing SP and a switch U1 coupled to this. The switch U1 has three switching positions, and in the shown switching position it transmits the input signal directly to the switched demultiplexer DE-MUX during the bypass of the intermediate storage SP. In the middle coupling position, the input of the demultiplexer DE-MUX is connected to the output of the intermediate storage SP, and in the upper coupling position it is connected to the output of the device R, which in the most extreme case delivers a pulse pattern which characterizes the resting state.

The output of the demultiplexer is followed by its own digital analog converter D / A. To these are joined spectral channels SK1, SK2, .... SKn. The signal containing the speech base frequency is applied at the output of the associated digital converter D / A to the control input for the pitch of a tone generator TG. The signal indicating the sound character, in turn, controls a switch U2 which, depending on the speech signal, has or does not have a voice character, supplies all other inputs to the voice channels the output signal of a pitch generator TG or the output of a noise generator RG. The synthesized subband channels of the original voice channel appearing at the output of spectral channels SK1, SK2, ___ SKn are summed in the sum amplifier SU 'and fed to the telephone H on the output side.

As shown in FIG. 4, the vocoder synthesizer VOE on its input side additionally has a signal evaluator SB, the first input of which also receives the arriving signal, while at its second input it receives the signal criterion sc ', which is obtained either in the decoder DE of FIG. 1 or in integrator I of FIG. 2. The signal assessor SB comprises a calculator which examines the signal supplied to it at the first input, as to whether an occurring signal pause is heard or caused by a disturbance of the signal on the transmission path. Only if, at the time of occurrence of the signal criterion signal, the signal assessor does not detect any signal interference, does the signal assessor bring switch U1 to its upper position for connection to the device R. Otherwise, the signal assessor SB controls the switch U1 to the middle switching position, where the demultiplexer DE-MUX receives a preceding,! the intermediate storage SP intermediate stored signal section supplied instead of the disturbed signal section.