GB2103454A

GB2103454A - Radio frequency systems

Info

Publication number: GB2103454A
Application number: GB08222092A
Authority: GB
Inventors: Larry Joseph Stagg; John Malcolm Robinson
Original assignee: General Electric Co PLC
Current assignee: General Electric Co PLC
Priority date: 1981-07-31
Filing date: 1982-07-30
Publication date: 1983-02-16

Abstract

A radio frequency transmission system in which speech signals are digitally encoded and multiplexed at 2 for transmission as multilevel amplitude and phase modulation of a carrier signal, in which at least some of said speech signals are encoded by voice-model speech encoders 1. <IMAGE>

Description

SPECIFICATION Radio frequency transmission systems The present invention relates to radio frequency transmission systems.

Speech signals, for example in a telephone system, are commonly encoded for transmission by sampling their instantaneous amplitude at a repetition rate of 8 Kbits/sec and deriving an eight-bit pulse code group to represent each sampled amp !itude, giving an encoded signal bit rate of 64 Kbits/sec. On major links of a transmission path these encoded signals may be interleaved or multi p- lexed, with the addition of signalling bits, to provide in one example a single stream of binary digits at a rate of some 1400M bits/sec representing 1920 speech signals. This stream of digits is then transmitted by modulation of a carrier signal having a frequency of the order of, say, 6 G Hz.

In one modulation scheme presently being developed, known as 1 6QAM, the multiplexed signals are split into four parallel bit streams each at a bit rate of 35 Mbits/sec, and, in any one bit period, the bit values in the four streams are applied in pairs to determine the amplitudes oftwo carrier signals in phase quadrature, these carrier signals then being combined as a single carrier signal whose phase and amplitude in that bit period may have any one of 16 phase/amplitude combinations in dependence upon the four bit values.

In terms of bandwidth utilisation this modulation scheme is roughly comparable in efficiency to that achieved with frequency-division multiplexing of analogue speech signals each having a bandwidth of 4KHz.

Proposals for improving bandwidth utilisation in respect of digitally encoded speech signals include increasing the number of phase/amplitude combinations to 32 or 64, but with these schemes it is correspondingly more difficult to distinguish the combinations from one another on demodulation.

According to the present invention in a radio fre quencytransmission system in which speech signals are digitally encoded and multiplexed fortransmission by modulation of a carrier signal, at least some of said speech signals are encoded by voice-model speech encoders.

According to another aspect of the present invention in a radio frequency transmission system in which digitally encoded signals at a basic digit rate equivalent to one sampled-waveform amplitude coded speech signal are multiplexed fortransmission as multilevel amplitude and phase modulation of a radio frequency carrier signal there are provided a plurality of voice-model speech encoders each arranged in respect of analogue speech signals on a respective input path to provide digitally coded output signals representing during each of a succession of time intervals a model of a vocal tract substantially to reproduce said analogue speech signals, and digital multiplexer means to combine said digitally coded output signals to provide a stream of digits at substantially said basic digit rate which constitute one of said digitally encoded signals fortransmission over the system.

A radio frequency system in accordance with the present invention will now be described by way of example with reference to the accompanying drawings, of which : Figure 1 shows schematically a transmitter arrangement for the system, Figure 2 shows schematically a receiver arrangement for the system, and Figure 3 shows an alternative form for part of the transmitter arrangement of Figure 1.

Referring to Figures 1 and 2 of the drawings the speech signals, instead of being simply amplitudesampled and coded, as in previously proposed systems, are voice-model coded, for example by means if linear predicitive coding analysers 1, to provide in respect of each speech signal a digitally coded signal at a bit rate of, say, approximately 4 Kbits/sec. Such voice-model coding provides, in respect of each of a succession of 20 m secs "frame" intervals a stream of some 80 bits, denoting, in respect in respect of the speech signal during that interval, whether it was voiced or unvoiced, its pitch, and a set of electric filter characteristics as accurately as possible mimicking the acoustic characteristics of the vocal tract which produced the speech signal.Up to sixteen of these voice-model coded signals may be multiplexed together in a multiplexer 2 to form a 64 Kbits/sec stream, equivalent to one sampledamplitude coded signal, for one input to a conventional digital multiplexer 3. The multiplexed digit stream is then applied to a multilevel coding arrangement 4 and thence to a 16 QAM transmitter 5.

The receiver 6 of the system feeds a decoder 7 which is followed by a demultiplexer 8. A 64 Kbits/sec digit stream from the demultiplexer 8 may be further demultiplexed in a circuit 9 and the individual 4 Kbits/sec digit streams applied to individual speech synthesizers 10 orto one or more time shared synthesizers.

With 4 Kbits/sec output signals from the voicemodel coders 1 the bandwidth utilisation can cleariy be increased by sixteen times over the present 16 QAM schemes.

As shown in Figure 3 the Analysers 1 shown in Figure 1 alternatively may be arranged in two groups 11 and 12 each having its own set of digitally coded speech input channels but the channels associated with the group 11 being selectively switchable by way of switching circuits 13 to the inputs of the group 12 in place of the channels normally associated with that group 12.

The frame periods for the analysers of the group 11 may be arranged to commence midway through the frame periods for the analysers of the group 12, and the respective outputs may simply be interleaved, or multiplexed together, in output switching circuits 14. When the number of speech channels actually in use is equal to less than the number of channels associated with the group 11 these speech channels may be connected by way of the circuits 13 to both sets of analysers 1 so that in respect of any one active speech channel overlapping sets of speech samples are analysed and the corresponding voice-model coded output signals are made available at intervals of 10 msecs instead of 20 msecs.In the corresponding synthesizer 10 at the receiving end of the system therefore the voice-model parameters may be up-dated twice as often as normal, giving improved quality in the resulting synthesized speech for a bit rate of approximately 8 Kbits/sec. A similar result would of course be obtainable with individual analysers with 10 msec coding cycles.

Even with voice-model coding of this improved quality a bandwidth utilisation eight times that of the present 16 QAM schemes can be achieved.

Further improvement of bandwidth utilization will of course be achieved by increasing the number of levels. Using say a 32 level system with voice-model coding of 8 Kbits/sec gives a further 25% improvement, that is an overall increase often times over the present 16 QAM schemes.

It will be appreciated that the use of multiplexed voice-model coding with lower level digital transmission systems, such as Quadrature Phase-shift Keyed (QPSK) or Reduced Bandwidth Quadrature Phase-shift (RBQPSK) systems, or with other presently used digital code modulation systems, including those using satellite links, a correspdonding improvement in bandwidth utilisation may be achieved.

Claims

1. A radio frequency transmission system in which speech signals are digitally encoded and multiplexed for transmission by modulation of a carrier signal, wherein at least some of said speech signals are encoded by voice-model speech encoders.

2. A radio frequency transmission system in which speech signals are digitally encoded and multiplexed for transmission as multilevel amplitude and phase shift modulation of a carrier signal, wherein at least some of said speech signals are encoded by voice-model speech encoders.

3. A radio frequency transmission system in which digitally encoded signals at a basic digit rate equivalent to one sampled-waveform amplitude coded speech signal are multiplexed fortransmission as multilevel amplitude and phase modulation of a radio frequency carrier signal, wherein there are provided a plurality of voice-model speech encoders each arranged in respect of analogue speech signals on a respective input path to provide digitally coded output signals representing during each of a succession of time intervals a model of a vocal tract substantially to reproduce said analogue speech signals, and digital multiplexer means to combine said digitally coded output signals to provide a stream of digits at substantially said basic digit rate which constitutes use of said digitally encoded signals for transmission over the system.

4. A radio frequency transmission system in accordance with Claim 3 wherein the voice-model speech encoders are of a linear predictive coding type.

5. A radio frequency transmission system in accordance with Claim 3 wherein speech signals are selectively encoded at different bit rates in dependence upon available bandwidth.

6. A radio frequency transmission system in accordance with Claim 3 wherein up to sixteen voice-model coded speech signals each having a bit rate of substantially 4 Kilobit/sec are multiplexed to provide a stream of bits at up to 64 Kilobit/sec, equivalent to one sampled waveform amplitude coded speech signal.

7. A radio frequency transmission system in which digitally encoded signals are transmitted as multi-level amplitude and phase modulation of a radio frequency carrier signal, comprising a plurality of speech signal input paths, a plurality of voicemodel speech encoders each arranged to provide in respect of speech signals on a respective input path digitally encoded output signals representing a model of a vocal tract substantially to reproduce those speech signals, said digitally encoded signals having a substantially lower digit rate than a predetermined digit rate equivalent to that of an amplitude-sampled and encoded speech signal, digital multiplexer means to combine said digitally encoded output signals from said plurality of voice-model encoders to provide a stream of digits at said predetemined digit rate, and further digital multiplexer means to combine said stream of digits with other streams of digits at said predetermined digit rate for transmission over the system.