Patent US4571736A successfully refreshed from Google Patents.
US4571736A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Digital communication system employing differential coding and sample robbing
Assignee: SOUTHWESTERN LOUISIANA A UNIVERSITY OF, University of
•
Filing: 1983-10-31
•
Granted: 1986-02-18
•
Expires: 2003-10-31
Abstract
Abstract Speech bit rate reduction by robbing, e.g. periodically not transmitting, a percentage of speech data samples in a communication system employing differential pulse code modulation (DPCM) or adaptive differential pulse code modulation (ADPCM). To overcome feedback interference introduced in the code by sample robbing, a robbed sample is replaced by its estimate so that the coding process continues in a normal manner. The estimate is established on the basis of autocorrelation statistics of the speech data samples. At the receiving end of the communication system, the robbed sample is estimated by the same process and re-estimated again using delayed interpolation after one period of delay. The technique is particularly useful where graceful degradation is desired under heavy traffic loading on the data channel, and is found most beneficial when the bit rate is about 24kb/s or lower where a relatively larger quantizing noise masks the interpolation noise introduced by sample robbing. The technique also facilitates handling speech packet loss in packet-networks. The signal-to-noise ratio was improved up to 3 db using a DPCM coder and 1.5 db using an ADPCM coder.
Inventors
- 👤 Jagannath P. Agrawal
- 👤 Subramaniam S. Iyer
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-06 15:40
CPC Classifications
H; H04; H04B; H04B14/00; H04B14/02; H04B14/04; H04B14/046
Claims Analysis
Total Claims: 24
Claim 1
Independent Claim
377 words
A method for improving the signal-to-noise ratio in a digital communication system employing a differential coder using a feedback path at a transmitter thereof for producing, where i is an interger, prediction error samples e i from successive input data samples s i in an input data stream and employing digital interpolation at a receiver thereof to reconstruct data samples s i in an output data stream representative of the input samples s i , said method comprising the steps of: processing at the transmitter at least two successive n-bit input data samples s i supplied thereto for producing, from a corresponding prediction error sample e i being a linear function of at least one order of said successive n-bit input data samples, an encoded prediction error sample e i ' for transmission over a data channel, periodically robbing from the data channel a number of encoded prediction error samples e i ' and replacing in the feedback path of the differential coder each input data sample s i affected by said robbing by an estimated input sample E(s i '), where E(s i ') is a linear function of at least two orders of previously supplied input data samples s i ', reconstructing said input data signals at the receiver by decoding successive prediction error samples e i ' transmitted over said data channel, said decoding being performed by establishing a predicted received sample s i on the basis of the autocorrelation function of at least one order of previously decoded samples s i ', decoding each received data sample as s i '=e i '+s i estimating the received data sample E(s i ') as a function of at least two previously decoded data samples s i ', after receiving at least one additional prediction error sample e i+1 ', . . . and decoding at least one additional data sample s i+1 ', re-estimating the data sample E(s i ') by delayed interpolation thereby to generate a re-estimated data sample E((s i ')), and replacing the decoded data sample s i ' in the output data stream with the re-estimated data sample E((s i ')) thereby to reconstruct the transmitted input data signal s i with an improved signal-to-noise ratio in the communication system.
Claim 2
Independent Claim
377 words
A method for improving the signal-to-noise ratio in a digital communication system employing a differential coder using a feedback path at a transmitter thereof for producing, where i is an interger, prediction error samples e i from successive input data samples s i in an input data stream and employing digital interpolation at a receiver thereof to reconstruct data samples s i in an output data stream representative of the input samples s i , said method comprising the steps of: processing at the transmitter at least two successive n-bit input data samples s i supplied thereto for producing, from a corresponding prediction error sample e i being a linear function of at least one order of said successive n-bit input data samples, an encoded prediction error sample e i ' for transmission over a data channel, periodically robbing from the data channel a number of encoded prediction error samples e i ' and replacing in the feedback path of the differential coder each input data sample s i affected by said robbing by an estimated input sample E(s i '), where E(s i ') is a linear function of at least two orders of previously supplied input data samples s i ', reconstructing said input data signals at the receiver by decoding successive prediction error samples e i ' transmitted over said data channel, said decoding being performed by establishing a predicted received sample s i on the basis of the autocorrelation function of at least one order of previously decoded samples s i ', decoding each received data sample as s i '=e i '+s i estimating the received data sample E(s i ') as a function of at least two previously decoded data samples s i ', after receiving at least one additional prediction error sample e i+1 ', . . . and decoding at least one additional data sample s i+1 ', re-estimating the data sample E(s i ') by delayed interpolation thereby to generate a re-estimated data sample E((s i ')), and replacing the decoded data sample s i ' in the output data stream with the re-estimated data sample E((s i ')) thereby to reconstruct the transmitted input data signal s i with an improved signal-to-noise ratio in the communication system.
Claim 3
Dependent Claim
32 words
A method as recited in claim 1 further comprising the step of synchronizing the transmitter and receiver when transmitting and receiving prediction error sample e i ' for incremental values of i.
Claim 4
Dependent Claim
32 words
A method as recited in claim 2 further comprising transmitting the predction error signal at a rate at or below an efficiency crossover frequency of approximately 24 to 32 kilobits per seconds.
Claim 5
Dependent Claim
136 words
A method as recited in claim 1 further comprising, in the estimating steps at both the receiver and transmitter, estimating the data sample E(s i ') as E(s.sub.i ')+a.sub.1 s.sub.i-1 '+a.sub.2 s.sub.i-2 '+. . . +a.sub.k s.sub.i-k ' for k greater than or equal to 2 and where a 1 , a 2 , . . . , a k are prediction coefficients, and at the receiver, re-estimating the data sample E((s i ')) as E((s.sub.i '))=b.sub.i-n s.sub.i-n '+. . . +b.sub.i-1 s.sub.i-1 '+b.sub.i+1 s.sub.i+1 '+. . . +b.sub.i+m s.sub.i+m ' for integers n, m each being greater than or equal to 1, and wherein b i--n , . . . , b i-1 , b i+1 , . . . , b i+m are interpolation coefficients obtained by applying a minimum mean square error criteria.
Claim 6
Dependent Claim
138 words
A method as recited in claim 3 further comprising, in the estimating steps at both the receiver and transmitter, estimating the data sample E(s i ') as ps E(s.sub.i ')=a.sub.1 s.sub.i-1 '+a.sub.2 s.sub.i-2 '+ . . . +a.sub.k s.sub.i-k ' for k greater than or equal to 2 and where a 1 , a 2 , . . . , a k are prediction coefficients, and at the receiver, re-estimating the data sample E((s i ')) as E((s.sub.i '))=b.sub.i-n s.sub.i-n '+. . . +b.sub.i-1 s.sub.i-1 '+b.sub.i+1 s.sub.i+1 '+. . . +b.sub.i+m s.sub.i+m ' for integers n, m each being greater than or equal to 1, and wherein b i-n , . . . , b i-1 , b i+1 , . . . , b i+m are interpolation coefficients obtained by applying a minimum mean square error criteria.
Claim 7
Dependent Claim
90 words
A method as recited in claim 1 further comprising, in the estimating steps at both the receiver and transmitter, estimating the data sample E(s i ') as E(s.sub.i ')=a.sub.1 s.sub.i-1 '+a.sub.2 s.sub.i-2 ' where a 1 and a 2 are prediction coefficients, and at the receiver, re-estimating the data sample E((s i ')) as E((s.sub.i '))=b.sub.i-2 s.sub.i-2 '+b.sub.i-1 s.sub.i-1 '+b.sub.i+1 s.sub.i+l ' wherein b i-2 , . . . , and b i-1 , b i+1 , are interpolation coefficients from obtained by applying a minimum mean square error criteria.
Claim 8
Dependent Claim
91 words
A method as recited in claim 3 further comprising, in the estimating steps at both the receiver and transmitter, estimating the data sample E(s i ') as E(s.sub.i ')=a.sub.1 s.sub.i-1 '+a.sub.2 s.sub.i-2 ' for k greater than or equal to 2 and where a 1 and a 2 are prediction coefficients, and at the receiver, re-estimating the data sample E((s i ')) as E((s.sub.i '))=b.sub.i-2 s.sub.i-2 '+b.sub.i-1 +b.sub.i+1 s.sub.i+1 ' wherein b i-2 , b i-1 and b i+1 are interpolation coefficients obtained by applying a minimum mean square error criteria.
Claim 9
Dependent Claim
29 words
A method as recited in claim 1 further comprising, in the processing step, producing said encoded prediction error sample e i ' by differential pulse code modulation (DPCM) techniques.
Claim 10
Dependent Claim
30 words
A method as recited in claim 1 further comprising, in the processing step, producing said encoded prediction error sample e i ' by adaptive differential pulse code modulation (ADPCM) techniques.
Claim 11
Dependent Claim
29 words
A method as recited in claim 3 further comprising, in the processing step, producing said encoded prediction error sample e i ' by differential pulse code modulation (DPCM) techniques.
Claim 12
Dependent Claim
33 words
A method as recited in claim 10 further comprising, in the robbing step, robbing a selected one of every second through tenth prediction error sample e i ' transmitted over said data channel.
Claim 13
Dependent Claim
29 words
A method as recited in claim 3 further comprising, in the processing step, producing said cncoded prediction error sample e i by adaptive differential pulse code modulation (ADPCM) techniques.
Claim 14
Dependent Claim
33 words
A method as recited in claim 12 further comprising, in the robbing step, robbing a selected one of every second through tenth prediction error sample e i ' transmitted over said data channel.
Claim 15
Dependent Claim
29 words
A method as recited in claim 4 further comprising, in the processing step, producing said encoded prediction error sample e i ' by differential pulse code modulation (DPCM) techniques.
Claim 16
Dependent Claim
30 words
A method as recited in claim 4 further comprising, in the processing step, producing said encoded prediction error sample e i ' by adaptive differential pulse code modulation (ADPCM) techniques.
Claim 17
Dependent Claim
29 words
A method as recited in claim 6 further comprising, in the processing step, producing said encoded prediction error sample e i ' by differential pulse code modulation (DPCM) techniques.
Claim 18
Dependent Claim
30 words
A method as recited in claim 6 further comprising, in the processing step, producing said encoded prediction error sample e i ' by adaptive differential pulse code modulation (ADPCM) techniques.
Claim 19
Dependent Claim
29 words
A method as recited in claim 7 further comprising, in the processing step, producing said encoded prediction error sample e i ' by differential pulse code modulation (DPCM) techniques.
Claim 20
Dependent Claim
33 words
A method as recited in claim 18 further comprising, in the robbing step, robbing a selected one of every second through tenth prediction error sample e i ' transmitted over said data channel.
Claim 21
Dependent Claim
30 words
A method as recited in claim 7 further comprising, in the processing step, producing said encoded prediction error sample e i ' by adaptive differential pulse code modulation (ADPCM) techniques.
Claim 22
Dependent Claim
33 words
A method as recited in claim 20 further comprising, in the robbing step, robbing a selected one of every second through tenth prediction error sample e i ' transmitted over said data channel.
Claim 23
Independent Claim
399 words
An apparatus for improving the signal-to-noise ratio in a digital communication system employing a differential coder which utilizes a feedback path at a transmitter thereof for producing, where i is an integer, prediction error samples e i from successive input data samples s i in an input data stream and employing digital interpolation at a receiver thereof to reconstruct data samples s i ' in an output data stream representative of the input samples s i , said apparatus comprising: means for processing at the transmitter at least two successive n-bit input data samples s i , s i-1 , . . . supplied thereto thereby to produce, from a corresponding prediction error sample e i being a function of at least one order of said successive n-bit input data samples, an encoded prediction error sample e i ' for transmission over a data channel. means for periodically robbing from the data channel a number of encoded prediction error samples e i ' and replacing in the feedback path of the differential coder each input data sample s i affected by said robbing by an estimated input sample E(s i '), where E(s i ') is a linear function of at least two order of previously supplied input data samples s i ', means for reconstructing said input data signals at the receiver by decoding successive prediction error samples e i ' transmitted over said data channel, and further including means for establishing a predicted received sample s i on the basis of the autocorrelation function of at least one order of previously decoded samples s i ', means for decoding each received data sample as s 1 '=e i '+s i means for estimating the received data sample E(s i ) as a function of at least two orders of previously decoded data samples s i ', after receiving at least one additional prediction error sample e i+1 ', . . . and decoding at least one additional data sample s i+1 ', means for re-estimating the data sample E(s i ') by delayed interpolation thereby to generate a re-estimated data sample E((s i ')), and means for replacing the decoded data sample s i ' in the output data stream with the re-estimated data sample E((s i ')) thereby to reconstruct the transmitted input data signal s i with an improved signal-to-noise ratio in the communication system.
Claim 24
Independent Claim
399 words
An apparatus for improving the signal-to-noise ratio in a digital communication system employing a differential coder which utilizes a feedback path at a transmitter thereof for producing, where i is an integer, prediction error samples e i from successive input data samples s i in an input data stream and employing digital interpolation at a receiver thereof to reconstruct data samples s i ' in an output data stream representative of the input samples s i , said apparatus comprising: means for processing at the transmitter at least two successive n-bit input data samples s i , s i-1 , . . . supplied thereto thereby to produce, from a corresponding prediction error sample e i being a function of at least one order of said successive n-bit input data samples, an encoded prediction error sample e i ' for transmission over a data channel. means for periodically robbing from the data channel a number of encoded prediction error samples e i ' and replacing in the feedback path of the differential coder each input data sample s i affected by said robbing by an estimated input sample E(s i '), where E(s i ') is a linear function of at least two order of previously supplied input data samples s i ', means for reconstructing said input data signals at the receiver by decoding successive prediction error samples e i ' transmitted over said data channel, and further including means for establishing a predicted received sample s i on the basis of the autocorrelation function of at least one order of previously decoded samples s i ', means for decoding each received data sample as s 1 '=e i '+s i means for estimating the received data sample E(s i ) as a function of at least two orders of previously decoded data samples s i ', after receiving at least one additional prediction error sample e i+1 ', . . . and decoding at least one additional data sample s i+1 ', means for re-estimating the data sample E(s i ') by delayed interpolation thereby to generate a re-estimated data sample E((s i ')), and means for replacing the decoded data sample s i ' in the output data stream with the re-estimated data sample E((s i ')) thereby to reconstruct the transmitted input data signal s i with an improved signal-to-noise ratio in the communication system.
References Cited (Backward Citations)
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1989-02-17 | 1989-08-19 | 1990-02-18 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1993-02-17 | 1993-08-19 | 1994-02-18 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 1997-02-17 | 1997-08-19 | 1998-02-18 | $7700.00 | Window Closed / Lapsed |