Mitigation of noise signal contrast in a digital speech interpolation transmission system

• US 4,363,122 A
• Filed: 09/16/1980
• Issued: 12/07/1982
• Est. Priority Date: 09/16/1980
• Status: Expired due to Term

• Current Assignee: Nortel Networks Limited (Nortel Networks Corporation)
• Original Assignee: Northern Telecom Limited (Nortel Networks Corporation)
• Inventors: Joyce, Robert H., Black, David H. A., Le Van, Tiu, Daaboul, Fouad
• Primary Examiner(s): Robinson, Thomas A.
• Application Number:
  US06/187,821
• Time in Patent Office: 812 Days
• Field of Search: 370/79, 370/80, 370/81, 370/113, 370/6, 179/1 B, 179/1 P, 179/1 D
• US Class Current: 370/435
• CPC Class Codes: H04J 3/172 Digital speech interpolation, i.e. DSI

Abstract

A DSI transmission system is described in which noise signal contrast is mitigated by transmitting not only active channels but also channels which carry background noise, the resolution of the background noise channels being varied in dependence upon the number of channels transmitted. In the described system, the DSI transmission is effected in superframes each of which consists of an overhead information frame and a plurality of digital message frames. The overhead information frame contains a word indicating the transmission status of each channel during the superframe, a code indicating the number of bits of each digital message, and maximum segment value information bits for each of the active channels which are transmitted using nearly instantaneous companding. Each of the remaining frames contains a digital message in respect of each active channel. Remaining space in these frames is occupied by digital messages in respect of the background noise channels, which are allocated to the frames by cyclic assignment and are transmitted using compact PCM encoding. For each superframe, active and background noise channels are determined using speech, echo, and off-hook detectors, and the number of bits of each digital message in the superframe is determined in dependence upon the total number of active and background noise channels.

Claims

1. A method of mitigating noise signal contrast in a multi-channel digital speed interpolation transmission system, comprising the steps of:

determining each channel which is active and each channel which carries background noise;

transmitting the active channels; and

transmitting the channels which carry background noise with a resolution which is dependent upon the number of channels transmitted and which is reduced as said number is increased.

2. A method as claimed in claim 1 wherein said determining step comprises, for each channel, the steps of:

detecting speech signals on a transmit path of the channel;

determining whether the channel is in an off-hook state;

detecting echoes on the transmit path of speech signals on a receive path of the channel;

determining that the channel is active if speech signals which are not echoes are detected on the transmit path and the channel is in the off-hook state; and

determining that the channel carries background noise if speech signals are not detected on the transmit path and the channel is in the off-hook state.

3. A method as claimed in claim 2 wherein, for each channel, the step of determining whether the channel is in the off-hook state comprises the steps of:

detecting speech signals on the receive path of the channel; and

determining that the channel is in the off-hook state in dependence upon the detection of speech signals on the transmit and receive paths during a preceding predetermined period.

4. In a multi-channel digital speech interpolation transmission system, the improvement comprising:

means for determining each channel which carries background noise, in addition to each channel which is active; and

means for transmitting, in addition to each active channel, each channel which carries background noise with a resolution which is dependent upon the number of channels transmitted and which is reduced as said number is increased.

5. A digital speech interpolation transmission system for transmitting a plurality of digital signal channels via a transmission link, comprising:

means for forming superframes each comprising a plurality of multi-bit frames;

means for determining, for each superframe, which channels are active and which channels carry background noise (b.n. channels); and

means for transmitting, in a normal loading situation, a digital message in respect of each active channel in each of said frames of the superframe and digital messages in respect of the b.n. channels in said frames of the superframes, the transmitting means comprising;

means for determining for each superframe, in dependence upon the numbers of active channels and b.n. channels for that superframe, numbers of bits for the digital messages of each active channel and each b.n. channel;

means for producing said digital messages, each of the respective number of bits, from the digital signals of each active channel and each b.n. channel; and

means for producing and transmitting, for each superframe, digital information relating to the transmission status of each channel, and the determined number of bits of each of the digital messages, in that superframe.

6. A system as claimed in claim 5 wherein said means for forming superframes is arranged to form each superframe with an additional, overhead information frame in addition to said plurality of frames, said digital information being transmitted in said overhead information frame of each superframe.

7. A system as claimed in claim 6 wherein the means for determining numbers of bits for the digital messages serves to determine a number of bits which is common to all of the digital messages in the relevant superframe, whereby a predetermined number of the digital messages each with the determined number of bits can be transmitted in each of said plurality of frames of the relevant superframe, said predetermined number being greater, in a normal loading situation, than the number of active channels, and wherein the means for transmitting the digital messages comprises means for transmitting the digital message of each active channel in each of said plurality of frames of the relevant superframe and for transmitting the digital messages of b.n. channels in the remaining available space in each of said plurality of frames, the transmission being effected cyclically for the b.n. channels throughout said plurality of frames of the relevant superframe.

8. A system as claimed in claim 7 wherein the means for producing the digital messages of each active channel from the digital signals of the channel comprises a nearly instantaneous companding (NIC) encoder for encoding the digital signals of each active channel in each superframe to produce a maximum segment digital value and a plurality of NIC encoded values, each NIC encoded value constituting a respective one of the digital messages of the active channel, and wherein the means for producing and transmitting said digital information comprises means for transmitting the maximum segment digital value of each active channel in the overhead information frame of the relevant superframe.

9. A system as claimed in claim 7 wherein the means for producing the digital messages of each b.n. channel from the digital signals of the channel comprises means for omitting one or more most significant magnitude bits from each digital signal to produce the respective digital message.

10. A system as claimed in claim 7 wherein said means for transmitting is responsive to an overload situation, in which the number of active channels is equal to or greater than the predetermined number of digital messages which can be transmitted in each of said plurality of frames with a predetermined minimum number of bits for each digital message, to transmit in the relevant superframe only said predetermined number of active channels with each digital message thereof having said predetermined minimum number of bits, b.n. channels not being transmitted in the superframe.

11. A system as claimed in claim 5 wherein said plurality of digital signal channels to be transmitted are derived from at least one source which is asynchronous to the timing of the transmission system, said system further comprising, for each such source, synchronizing means comprising:

a store;

means for writing the digital signals of the channels from the source sequentially into the store;

means for reading the digital signals of the channels sequentially from the store at a higher rate than that at which they are written into the store;

means for producing a stuffing command when the phase of reading from the store lags the phase of writing into the store by a predetermined amount;

means responsive to the stuffing command for inhibiting the sequential reading from the store of digital signals in respect of a predetermined frame of the superframe whereby said predetermined frame is stuffed;

means for transmitting the stuffing command as part of said digital information to identify the superframe including the stuffed frame; and

means at a receiver of the system for reconstructing the timing of the asynchronous source and responsive to the stuffing command to ignore the stuffed information in said predetermined frame.

12. A system as claimed in claim 5 wherein the means for determining which channels are active and which channels carry background noise comprises:

a speech detector for detecting speech signals on a transmit path, for each channel for each superframe;

an off-hook detector for determining for each channel, for each superframe, whether the channel is in an off-hook state;

an echo detector for detecting, for each channel for each superframe, signals on the transmit path which are echoes of signals on a receive path; and

gating means for producing, for each channel for each superframe, a signal representing that the channel is active if speech signals which are not echoes are detected on the transmit path and the channel is in the off-hook state, and a signal representing that the channel carries background noise if speech signals are not detected on the transmit path and the channel is in the off-hook state.

13. A system as claimed in claim 12 wherein the off-hook detector comprises means for detecting speech signals on a receive path, for each channel for each superframe, and means responsive, for each channel, to a plurality of consecutive detections of speech signals for the channel, by said speech detector or by said means for detecting speech signals, occurring within a preceding predetermined number of superframes to produce a signal representing that the channel is in the off-hook state.

14. A system as claimed in claim 12 and including means for producing, for each superframe, from the digital signals of the channels an average signal for each channel, said detectors each being responsive to the average signal for each channel.