Method of and system for handling conference calls in digital telephone exchange

• US 4,076,966 A
• Filed: 08/02/1976
• Issued: 02/28/1978
• Est. Priority Date: 08/02/1976
• Status: Expired due to Term

• Current Assignee: Italtel SPA (Italtel Group SPA)
• Original Assignee: Societa Italiana Telecomunicazioni Siemens SPA (Abaco Innovazione SPA)
• Inventors: Bovo, Amilcare, Valsecchi, Giuseppe
• Primary Examiner(s): Claffy, Kathleen H.
• Assistant Examiner(s): Popek, Joseph A.
• Application Number:
  US05/710,636
• Time in Patent Office: 575 Days
• Field of Search: 179/1 CN, 179/18 BC, 179/15 AP, 179/15 BW, 179/15.55 T, 179/15.55 R, 179/15 AT, 340/347 DD
• US Class Current: 370/268
• CPC Class Codes: H04Q 11/04 for time-division multiplexing H04Q11/0001 takes precedence

Abstract

A telephone exchange linked via PCM channels with subscribers engaged in a conference call includes switching circuits for transmitting to each participating subscriber, during an assigned time slot, the albegraic sum of the digitized voice-current samples from all the other participants. The digitized samples are logarithmically coded, each code consisting of a polarity bit, three range bits and four quantum bits; in order to facilitate correct addition or subtraction of the quantum bits of two codes to be summed (unless these codes signify the two lowest-ranking ranges), according to whether their sign bits are identical or different, the quantum bits of a lower-ranking code are shifted in a register toward less significant bit positions in which they would have their original numerical weight when prefaced by the range bits of the higher-ranking code. The range bits are not summed but only those of the higher-ranking code are preserved, their numerical value being incremented by one in the case of addition but left unchanged or decremented by one (if a negative carry is present) in the case of subtraction. If the difference in rank exceeds the number of quantum bits, i.e. four, the lower-ranking code is ignored.

Claims

1. A method of linearly superposing significant portions of logarithmically coded coincident voice samples in a conference call, each voice-sample code consisting of a polarity bit, a plurality of higher-order range bits and a plurality of lower-order quantum bits, the numerical weights of said quantum bits depending on the rank of the code as determined by said range bits, comprising the steps of:

a. separately storing a higher-ranking code and a lower-ranking code derived from a pair of coincident voice samples to be arithmetically combined;

b. downshifting the quantum bits of the lower-ranking code to less significant bit positions in which they would have their original numerical weights if prefaced by the range bits of the higher-ranking code;

c. comparing the polarity bits of the two codes stored in step (a); and

d. subjecting the quantum bits of the higher-ranking code and the downshifted quantum bits of the lower-ranking code to an algebraic summing operation with a sign depending on the relative values of the corresponding polarity bits as determined by step (c).

2. A method as defined in claim 1 wherein the numerical weights of the quantum bits in codes of rank 0 and rank 1 are the same, the quantum bits of the lower-ranking code being downshifted in step (b) by a number of bit positions equal to the difference in rank upon said lower-ranking code being above rank 0 and by a number of bit positions equal to said difference minus one upon said lower-ranking code being of rank 0.

3. A method as defined in claim 2 wherein an additive summing operation in step (d), in the presence of identical polarity bits, is followed by the steps of (e) incrementing by one the numerical value of the combination of range bits of the higher ranking code, (f) downshifting by one bit position the bits constituting the sum of the additively summed quantum bits including any carry resulting from the addition, and (g) juxtaposing the incremented range bits from step (e) with all but the lowest of the downshifted bits from step (f) to form a composite code.

4. A method as defined in claim 2 wherein a subtractive summing operation in step (d), in the presence of different polarity bits, is followed by the steps of (e) determining whether the subtraction of the quantum bits of the lower-ranking code from those of the higher-ranking code produces a negative carry, (f) decrementing the numerical value of the combination of range bits of the higher-ranking code by one in the presence of a negative carry while upshifting by one bit position the quantum bits constituting the difference of the subtractively summed quantum bits unless the higher-ranking code is of rank 1, and (g) juxtaposing the range bits of the higher-ranking code with the differentially obtained quantum bits, as modified in accordance with step (f) but without the highest upshifted quantum bit in the presence of a negative carry, to form a composite code.

5. A system for linearly superposing significant portions of logarithmically coded voice samples in a conference call, each voice-sample code consisting of a polarity bit, a plurality of higher-order range bits and a plurality of lower-order quantum bits, the numerical weights of said quantum bits depending on the rank of the code as determined by said range bits, comprising:

code-receiving means including two storage circuits each including first register means for storing polarity and range bits and second register means for storing quantum bits of respective codes to be arithmetically combined;

first arithmetic means connected to said first register means for determining the difference in ranks between said respective codes;

linearization means connected to said second register means and controlled by said first arithmetic means for downshifting the quantum bits of the lower-ranking code to less significant bit positions in which they would have their original numerical weights if prefaced by the range bits of the higher-ranking code;

second arithmetic means connected to said second register means and controlled by the contents of said first register means for subjecting the quantum bits of the higher-ranking code and the downshifted quantum bits of the lower-ranking code to an algebraic summing operation with a sign depending on the relative values of the corresponding polarity bits; and

evaluation means connected to said first and second arithmetic means for deriving a composite code from the range bits of the higher-ranking code and the quantum bits resulting from said algebraic summing operation.

6. A system as defined in claim 5 wherein said evaluation means comprises a shift register with n stages connected to be loaded with resultant bits by said second arithmetic means, n being the number of quantum bits in each code, and logic circuitry connected to said second arithmetic means to said storage circuits for shifting the bits in said stages in a sense depending upon the magnitudes of said range bits and the presence of a carry due to said algebraic summing operation.

7. A system as defined in claim 6 wherein said evaluation means further comprises a reversible counter connected to said first arithmetic means for receiving therefrom the range bits of the higher-ranking code, said counter being provided with m outputs where m is the number of range bits in each code, and a final register connected to said m outputs and said n stages, said logic circuitry being connected to said counter for modifying the contents thereof in a sense opposite the sense of shifting of the bits in said stages.

8. A system as defined in claim 7 wherein said shift register is provided with an input connected to said logic circuitry for introducing a positive carry into a stage vacated by a downshifting of said resulting bits.

9. A system as defined in claim 7 wherein said evaluation means further includes memory means controlled by a program counter and switchover means under the control of said memory means for actuating said first arithmetic means to load said counter in a first operating phase with the numerical difference between the combinations of range bits of said respective codes and in a second operating phase with the numerical value of the combination of range bits of the higher-ranking code only, said linearization means being controlled by said first arithmetic means through said counter in said first operating phase.

10. A system as defined in claim 9 wherein said logic circuitry further includes comparison means connected to said first register means for determining the relative magnitudes of the polarity and range bits of said respective codes, said program control being responsive to siad comparison means for skipping said first operating phase upon equality of the range bits of said respective codes.

11. A system as defined in claim 10 wherein said logic circuitry further includes other comparison means connected to said second register means for determining the relative magnitudes of the quantum bits of said respective codes, said shift register being provided with a resetting connection controlled by both said comparison means for delivering an all-zero bit combination to said final register upon equality of the range and quantum bits and diversity of the polarity bits of said respective codes.

12. A system as defined in claim 10 wherein said logic circuitry further includes loading means controlled by said comparison means and effective in said second operating phase to enter an all-one bit combination in said shift register for delivery to said final register, upon equality of said polarity bits, in response to a maximum numerical value of the combination of range bits of the higher-ranking code and in the presence of certain numerical values of the lower-ranking code.