Data processing: speech signal processing – linguistics – language – Speech signal processing – For storage or transmission
Reexamination Certificate
1999-07-21
2002-12-10
Dorvil, Richemond (Department: 2741)
Data processing: speech signal processing, linguistics, language
Speech signal processing
For storage or transmission
C704S201000, C704S223000, C704S500000
Reexamination Certificate
active
06493666
ABSTRACT:
DESCRIPTION OF THE RELEVANT ART
The ability to transmit data rapidly to virtually any place in the world has become one of the defining characteristics of the current information age. This ability comes at a cost, however, because it requires a global infrastructure for providing telecommunications services. Since the origins of the telecommunications infrastructure in local telephone networks, many successful attempts have been made to control costs by making the existing infrastructure perform more tasks and functions. Generally, these attempts have involved either increasing the ability of the existing telecommunications infrastructure to handle more data, or preprocessing raw data so that less data need be transmitted to convey the same amount of information.
Attempts to increase the data handling capabilities of the existing telecommunications infrastructure have included such techniques as the use of time and frequency division multiplexing. In time division multiplexing, multiple low speed channels are combined for transmission over a single high speed channel. In frequency division multiplexing, a large capacity channel is subdivided into a number of lower capacity channels. More recent development of these multiplexing techniques lies in modern packet switched networks, where data streams are broken into small segments that may be routed individually to take advantage of available channel capacity.
Sometimes, however, the available channel capacity between two or more locations is either physically limited or otherwise expensive. A dramatic example of this situation has involved communications between space probes and earth stations. Large quantities of scientific data, and particularly pictures, need to be transmitted from the spacecraft to earth via relatively low capacity channels. The solution to this problem has been the development of compression techniques that reduce the amount of data required to transmit the same amount or approximately the same amount of information.
Compression techniques generally operate by changing the representation of information from one set of data to another set of data. In the process of compression, extraneous or unnecessary information, including redundancies, may be eliminated. Using such techniques, many common representations, such as photographs and human speech recordings, can often be substantially compressed with relatively little loss of information.
Telephone carriers have widely used multiplexing techniques to take advantage of available channel capacity in transmitting human speech. More recently, and particularly with the advent of cellular telephones that operate in environments with limited channel capacity, carriers have also employed compression techniques to increase the transmission capacities of their telecommunications infrastructures. Commonly used compression techniques used are often lossy, i.e., some information is irretrievably lost, and result in changes to the “sound” of a speaker even though the words being uttered are still understandable.
The application of compression techniques has generally focused on individual communications paths and particularly on the endpoints of those paths. If compression results in the reduction in the amount of data that must be transmitted then it is logical to apply compression as close to the source of the data as possible. Thus, for example, pictures are typically compressed within space probes before they are transmitted, and likewise speech is typically compressed in cellular telephone handsets before it is transmitted. This trend has been particularly evident in traditional telephone system environments where, until recently, little computer capability to perform compression has been available at central offices and high capacity relay stations.
Especially in competitive telecommunications environments, there is continuing pressure to reduce costs and increase efficiency by making better use of the existing telecommunications infrastructure. In many cases, however, this infrastructure has recently changed radically, due to the use of various packet switching techniques to transport data. These changes were experienced initially by corporations using packet switched networks to transmit data in bulk, and more recently there has also been an increasing effect of use of these techniques on traditional telephone communications, as exemplified by U.S. Pat. No. 5,751,706 (describing a system and method for establishing a call telecommunications path that, in some embodiments, makes use of packet switched networks to transport telephone calls). These changes have also resulted in the concentration of computer processing capacity at locations where many individual data channels come together in high capacity packet switched networks.
SUMMARY OF THE INVENTION
Data compression techniques typically change the representation of information to remove extraneous or unnecessary information, including, for example, redundancies. Depending on the representation scheme employed, a particular data channel may contain substantial redundant extraneous or unnecessary information, and therefore be highly compressible. Similarly, a group of data channels, particularly if they are carrying related forms of information such as human telephone conversations, may contain substantial redundant extraneous or unnecessary information in common amongst the group of channels and thus be capable of being compressed as a group to a greater degree than if the channels were compressed individually. The availability of substantial computing power at locations where many individual data channels come together in high capacity packet switched networks provides an attractive environment for taking advantage of the compressability of groups of data channels.
Many compression techniques, defined in this specification as including the operations of quantization, scaling, and coding, are known in the art. The degree of compression provided by a particular technique, however, may be highly dependent on the arrangement or structure of the data to be compressed. Simply dispatching data, originating in multiple individual channels, to a compression mechanism will not, in many cases, take advantage of the redundant, extraneous or unnecessary information in common amongst the multiple channels. In contrast, the organization of data before it is compressed, which is a focus of the system and method of the present invention, enables the compression mechanism to take substantial advantage of the “in common” redundant, extraneous, or unnecessary information.
Of course, compression is not an end in itself, but merely facilitates the transmission of representations of sequences of digital data elements to another location where the sequences of digital data elements are approximately or completely reconstructed. As used in this specification, the term “compression-side” refers to the processing of that takes place from the receipt of information from communications channels to the presentation of compressed data to a transmission network for onward transmission. The term “decompression-side” refers to all of the processing that takes place from the receipt of compressed data from a transmission network to the presentation of approximately or completely reconstructed information to communications channels for onward transmission.
In a preferred embodiment, the present invention enables multiple voice-grade telephone channels, such as might be present at a telephone company central office, to be transmitted via packet-switched computer network in compressed form to a destination location where the multiple voice-grade telephone channels are approximately or exactly reconstructed for final transmission to ultimate destinations. A user of one of these voice-grade telephone channels would be at least substantially unaware that his call was transmitted via a packet switched network rather than through customary telephone company facilities.
As used in this specification, the term “channel” refers to both the data transported
Chawan Vijay B
Covington & Burling
Dorvil Richemond
LandOfFree
System and method for processing data from and for multiple... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for processing data from and for multiple..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for processing data from and for multiple... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2964166