Pulse or digital communications – Spread spectrum – Direct sequence
Reexamination Certificate
1998-03-31
2001-05-01
Pham, Chi (Department: 2631)
Pulse or digital communications
Spread spectrum
Direct sequence
C370S209000, C375S340000, C375S341000
Reexamination Certificate
active
06226318
ABSTRACT:
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an arrangement for detecting code vectors which are transmitted in a single frequency, multiple transmitter network.
BACKGROUND OF THE INVENTION
Data communication systems, such as cable systems, typically include a head end which transmits data to a plurality of subscribers over a cable system. Typically, the cable system is at least partially buried and has a cable main trunk carrying data directly from the head end, cable branch lines branching out of the main trunk, and cable subscriber lines carrying data between the cable branch lines and the subscribers. Considerable labor is required in running subscriber lines from cable branch lines to subscribers, particularly for those subscribers who are located at distances such as 1,000 feet or more from the cable branch lines.
Instead of running subscriber lines from cable branch lines to subscribers, transmitters could be located periodically along the cable branch lines in order to transmit data over the air between cable branch lines and subscribers. Thus, the substantial labor which is necessary to connect a subscriber to a cable branch line is materially reduced. However, care must be exercised in locating such transmitters. For example, if a subscriber is covered by only one transmitter, there may be areas within the premises of the subscriber where reception is poor.
The possibility of poor reception can be lessened by locating the transmitters sufficiently close to one another that the premises of each subscriber is covered by two or more transmitters. Unfortunately, because each transmitter operates at the same carrier frequency, and because of the variable distances between a subscriber's premises and the transmitters that cover the subscriber's premises, the same data may arrive at a reception site within a subscribers premises at different times and with different phases. As a result, interference, referred to herein as ghosting, is produced.
If signal amplitude versus frequency of the received signal at a reception site in a subscriber's premises covered by two transmitters is graphed, an interference pattern can result. In the case where the reception site is located at an equal distance from both transmitters, the interference pattern has the shape shown in FIG.
1
. This interference pattern may be sometimes referred as 100% ghosting. In this interference pattern, the signal amplitude of the received signal is characterized by periodic, sharply defined nulls at which the signal is substantially undetectable, particularly in the presence of noise. That is, noise in the channel establishes a signal detection threshold above the horizontal axis as viewed in
FIG. 1
, such that any frequency components of the transmitted signal near or at the nulls will be difficult or impossible to detect because the signal to noise ratio at these points is too low. Moreover, when the received signal is processed through an equalizer, the signal to noise ratio can worsen, making signal detection even more difficult.
It is known how to adequately receive signals in the presence of white noise. For example, trellis encoding and Viterbi decoding may be used to encode and decode transmitted data adequately when white noise is present, because this type of coding and decoding performs well under white noise conditions. Unfortunately, trellis encoding and Viterbi decoding do not work particularly well in the presence of non-randomly distributed noise, such as may be present in an environment experiencing 100% ghosting.
The present invention is directed to a coding and decoding arrangement which is particularly useful in a single frequency, multi-transmitter network in which 100% ghosting is present.
SUMMARY OF THE INVENTION
In accordance with the present invention, a receiver receives a signal containing at least one of a plurality of code vectors. A transform is applied to the signal producing a plurality of multi-coefficient spectra. Data elements are derived from the multi-coefficient spectra. More particularly, the data elements are derived from the coefficient in the multi-coefficient spectra having the largest magnitude.
In a more detailed aspect of the present invention, the received signal is multiplied by a plurality of coset leaders to produce a plurality of multiplication results. The coset leaders correspond to cosets into which the code vectors of the plurality of code vectors may be divided. The resulting multi-coefficient spectra are analyzed in order to derive the data elements corresponding to the at least one code vector.
In a further more detailed aspect of the present invention, the received signal may contain at least first and second code vectors. Data elements are determined corresponding to one of the first and second code vectors based upon the magnitudes of the coefficients of the multi-coefficient spectra. This one code vector is subtracted from the received signal to produce a subtraction result. Data elements are then determined corresponding to the other of the first and second code vectors based upon the magnitudes of the coefficients of multi-coefficient spectra derived from the subtraction result.
In a still further more detailed aspect of the present invention, the received signal may contain a plurality of transmitted code vectors. A received signal transform of the received signal is produced such that the received signal transform contains a coefficient for each of at least some of the code vectors of the predetermined set of code vectors. Data elements are determined corresponding to one code vector from the coefficient of the received signal transform having the largest magnitude. This code vector is subtracted from the received signal to produce a subtraction result. A subtraction result transform is produced such that the subtraction result transform contains a coefficient for each of at least some of the code vectors of the predetermined set of code vectors. Data elements are determined corresponding to another code vector from the coefficient of the subtraction result transform having the largest magnitude.
In yet a further more detailed aspect of the present invention, a received signal contains a plurality of code vectors which belong to a predetermined set of code vectors. The code vectors in the predetermined set of code vectors are divided into cosets, and the cosets are arranged into groups of cosets. Each code vector contained in the received signal belongs to a corresponding group of cosets. A window is applied to a first subset of the cosets. The received signal is multiplied by a coset leader corresponding to each coset within the window in order to produce a received signal coefficient spectrum for each coset within the window. Data elements are determined from a received signal coefficient that corresponds to a coset within the window and that has the largest magnitude. The code vector corresponding to these data elements is subtracted from the received signal to produce a subtraction result. The window is slid to cover a second subset of the cosets. The subtraction result is multiplied by the coset leaders corresponding to the cosets within the window in order to produce a subtraction result coefficient spectrum for each coset within the window. Data elements are determined from a subtraction result coefficient that corresponds to a coset within the window and that has the largest magnitude.
REFERENCES:
patent: 5751856 (1998-05-01), Hirabayashi
patent: 5949814 (1999-09-01), Odenwalder et al.
patent: 5966370 (1999-10-01), Gunzelmann
patent: 5987076 (1999-11-01), Zehavi et al.
patent: 5991627 (1999-11-01), Honkasalo et al.
patent: 6009552 (1999-12-01), Ariel et al.
patent: 6035007 (2000-03-01), Khayrallah et al.
patent: 0 565 504 (1993-10-01), None
patent: 0 565 506 (1993-10-01), None
patent: WO 95/22859 (1995-08-01), None
International Search Report, dated Mar. 11, 1999, Application No. PCT/US98/24637.
Citta Richard W.
Dicke Stephen M.
Bayard Emmanuel
Pham Chi
Zenith Electronics Corporation
LandOfFree
Detection of code vectors in single frequency, 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 Detection of code vectors in single frequency, multiple..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Detection of code vectors in single frequency, multiple... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2566950