Multiplex communications – Communication techniques for information carried in plural... – Adaptive
Reexamination Certificate
1998-09-08
2001-10-23
Kizou, Hassan (Department: 2661)
Multiplex communications
Communication techniques for information carried in plural...
Adaptive
C370S466000, C370S467000, C137S317000
Reexamination Certificate
active
06307864
ABSTRACT:
CROSS-REFERENCE TO RELATED APPLICATIONS
This application is based upon and claims the benefit of the priority of the prior French patent application 97 11137 filed on Sep. 8, 1997, the contents of which are incorporated herein.
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to information coding, and more specifically to the baseband transcoding of digital information that is to be transmitted over a telephone line.
2. Description of the Related Art
Conventional baseband transcoding has two steps: a coding step that associates a predetermined symbol with a sequence of digital information, and a shaping step that associates an elementary pulse of predetermined duration with each coded symbol. Among the codes typically used is a first type known as 2B1Q (defined in the European standard ETSI ETR 080 and the American standard ANSI T1.601) and a second type known as 4B3T (defined in the European standard ETSI ETR 080 and the German standard FTZ 1 TR 220. The codes are used for the transmission of digital information within a digital network of universal character, such as an integrated services digital network (“ISDN”) (defined in the European standard ETSI ETR 080 and the American standard ANSI T1.601.
The three main characteristics of ISDN are: (1) the capability of simultaneously employing telephone, telematic, and even video-communication services from one access; (2) subscriber-to-subscriber integral digital transmission with digital connections in the various switching centers; and (3) signaling in message mode only between the various functional entities (e.g., terminals, switches, and servers). The architecture of the local network attaching users to an ISDN network includes: on the subscriber side, a network digital terminal (TNR) to which the various user terminals are attached, and on the switch side, a line terminal (TL) that provides substantially the same functions as the TNR terminal. Between the TNR terminal and the TL terminal there is a subscriber line (U interface) that consists of a pair of copper cables whose characteristics (e.g., maximum length, useful bit rate, and error rate) are standardized.
The useful bit rate with ISDN is 144 Kbits/s, which is used by two “B” channels operating at 64 Kbits/s for transporting useful information and a “D” channel operating at 16 Kbits/s for transporting check bits. Thus, the subscriber line alone provides the so-called “narrow band” ISDN connection that allows access at the basic bit rate (2B+D) of 144 Kbits/s. By using codes such as the 4B3T code or the 2B1 Q code, it becomes possible to reduce the spectral band used for transmission.
Transmitters operating with either the 2B1 Q code or the 4B3T code are known. However, when using a particular baseband transcoding, it is necessary to use a network digital terminal containing a coder specific to the type of coding employed. Thus, changing the type of coding requires replacing the network digital terminal with another network digital terminal that is specifically adapted for use with the new code.
SUMMARY OF THE INVENTION
The present invention provides a switchable device for baseband transcoding that allows a single circuit containing a common filter for shaping symbols to go from one code to another code. In a preferred embodiment, the device can switch between the 2B1Q code and the 4B3T code.
One embodiment of the present invention provides a device for the baseband transcoding of digital information that is to be transmitted over a telephone line within an analog signal delivered to the line transformer and consisting of pulses associated with symbols obtained through the coding of the digital information. The device includes a preprocessing circuit that, in the presence of a first value of a selection signal, extracts first successive pairs of bits each containing a first “sign” bit and a second “magnitude” bit from the digital information at a first frequency so as to form for each first pair one of the symbols of a first type of baseband coding. The preprocessing circuit also, in the presence of a second value of the selection signal, performs a second type of coding and delivers at a second frequency each of the symbols of the second type of coding in the form of a second pair of bits containing a first “sign” bit and a second “magnitude” bit of predetermined values. Stated otherwise, irrespective of the value of the selection signal (and thus the type of coding selected), the preprocessing circuit delivers pairs of sign bits and magnitude bits even though conventional 4B3T coding does not provide for designating each of the possible symbols of the code by a pair of bits.
In a second embodiment, the transcoding device also includes first and second memories respectively containing first and second digital samples representative of at least a part of the predetermined pulses that are associated with the symbols of the first and second types of coding.
In a third embodiment, the transcoding device also includes a shaping filter that reads the contents of the first or second memory at the same reading frequency irrespective of the value of the selection signal, and that uses the contents of the corresponding memory and the successive values of the magnitude and sign bits to derive a digital output signal sampled at the reading frequency. The digital output signal is compatible with the coding selected and is representative of the corresponding analog transmission signal. Therefore, a sign bit and a magnitude bit of predetermined values are associated with each symbol of the 4B3T code so as to effect a parallel with the sign and magnitude bits used in the 2B1Q code. Additionally, the sign bit is regarded as a unique first control bit for the common shaping filter and the magnitude bit is regarded as a unique second control bit for the common shaping filter irrespective of the type of coding selected. Further, the shaping filter reads the stored samples at the same reading frequency and the sampled digital output signal is delivered at this same reading frequency, irrespective of the coding used. However, the successive values of the sign and magnitude bits are delivered at different delivery frequencies for the two types of coding.
In a fourth embodiment, all the digital samples stored in the first and second memories are obtained through two different samplings of the same mathematical function. For example, a raised cosinusoid elevated to an integer power (e.g., the power of 4) could be used. Moreover, the product of each frequency of delivery times the number of samples representative of a pulse of the type corresponding to the coding considered is equal to the reading frequency. This makes it possible to use the same basic mathematical function to digitally generate the various pulses of the two types of coding, even though the pulses do not have the same time duration. For example, all the pulses corresponding to the symbols of the 2B1Q coding have a time duration of 25 ps and all the pulses associated with the symbols of the 4B3T coding have a time duration of 16.66 ps. Thus, in practice, 24 samples can be stored to represent one pulse of the 2B1Q type and 16 samples can be stored to represent one pulse of the 4B3T type.
Furthermore, the two frequencies of delivery of the symbols for the two types of coding correspond to a time period of delivery equal to half the duration of the selected pulses of the type of coding. Stated otherwise, for each type of coding, the duration of a symbol is equal to half of the duration of the corresponding pulse. Also, the shaping filter derives the samples of the corresponding output signal at the selected frequency by adding together at the reading frequency samples drawn from the digital samples relating to the current symbol and samples drawn from the digital samples relating to the preceding symbol.
In a preferred embodiment, the shape of the pulses is chosen such that irrespective of the type of coding used, each nonzero pulse has symmetric front and rear edges extending temporally
Cornu Céline
Fensch Thierry
Kohlenberger Gérald
Kizou Hassan
STMicroelectronics S.A.
T T
LandOfFree
Process and device for the baseband transcoding of digital... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Process and device for the baseband transcoding of digital..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Process and device for the baseband transcoding of digital... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2613233