A communication system in which multiple shaping codes are selectively and iteratively used to encode a data frame such that possible energy inefficiencies associated with the use of constant-probability codes and/or transmission of dummy constellation symbols can be relatively small. In an example embodiment, the used shaping codes have different respective code rates, and a code selector of the shaping encoder operates to select one of the shaping codes by adaptively matching the rate of the code to the effective rate needed to efficiently encode the unprocessed portion of the data frame. The encoding is carried out in a manner that enables the shaping decoder to unequivocally determine the shaping codes that have been used for encoding each particular data frame based on the same rate-matching criteria as those used by the shaping encoder. At least some embodiments advantageously lend themselves to being implemented using circuits of relatively low complexity.
We disclose a transmitter that uses at least first and second fixed constellations in which the same bit-words are assigned to different respective constellation symbols of different respective transmit energies. The transmitter generates an outgoing data frame by first generating two data frames using the first and second constellations, respectively, and then selecting the one of the two data frames that has the lower overall transmit energy and discarding the other. The first and second constellations are constructed in a manner that enables the transmitter to realize a significant shaping gain. Some embodiments of the transmitter are compatible with the use of forward-error-correction coding and provide a shaping gain for the transmission of both information and parity bits. An example embodiment of the transmitter can advantageously be implemented with relatively low complexity by employing constellation mappers and demappers that operate using relatively small look-up tables.
Construction Of Structured Ldpc Convolutional Codes
Protograph construction methods for generating convolutional LDPC code matrices are disclosed in which multi-equation problems of girth maximization are reduced or replaced using other techniques including (with limitation): finding base matrices with a unique set of non-repeating distance parameters, finding the minimum largest such distance parameter among solution-set matrices, and quasi-cyclic lifting of the generated convolutional LDPC code matrix. 4-cycles and select (avoidable) 6-cycles are thereby removed from the resulting convolutional LDPC code matrix, thereby resulting in significant performance gains.
Youtube
Optical Hybrid SPM technology Development and...
Park Systems launched this online event for researchers and scientists...
Duration:
20m 31s
Sang Joon Cho on Park Systems: Favourites in ...
To celebrate the 30th anniversary of the scanning tunnelling microscop...
Duration:
7m 7s
CHEAT ON ME IF YOU CAN Teaser #2 | Go Joon, C...
Catch Cho Yeo Jeong, Go Joon & Kim Young Dae in Cheat On Me If You Can...