The technology relates to bandwidth constrained communication systems with neural network based detection. In some embodiments, a bandwidth constrained equalized transport (BCET) communication system comprises: a transmitter comprising an error control code encoder, a pulse-shaping filter, and a first interleaver; a communication channel; and a receiver comprising a neural network processing block that processes a received signal. The error control code encoder can append redundant information onto the signal. The pulse-shaping filter can intentionally introduce memory into the signal in the form of inter-symbol interference. The first interleaver can change a temporal order of the symbols in the signal. The error control code encoder can be a low-density parity-check (LDPC) error control code encoder. The neural network can be trained with positive mappings between transmitted and decoded training signals, or negative mappings between training signals and a null space of an LDPC generation matrix.

Methods are disclosed for improving communications on feedback transmission channels, in which there is a possibility of bit errors. The basic solutions to counter those errors are: proper design of the CSI vector quantizer indexing (i.e., the bit representation of centroid indices) in order to minimize impact of index errors, use of error detection techniques to expurgate the erroneous indices and use of other methods to recover correct indices.

In a MIMO detector, a slicing list detection scheme employs a complex plane to represent symbols in an alphabet, in which all decision regions are bounded by either vertical lines or horizontal lines. A K-best scheme accesses the complex plane and an offline-generated lookup table to detect elements of a received vector. At each level, the system prunes all but the K-best candidates from each surviving node through the slicing list detector.

This application discloses a decoding method for an OvXDM system, including: generating an augmented matrix B related to a received symbol information sequence; performing singular decomposition on the augmented matrix B; and performing decoding by using a total least square method, to obtain a decoded output information sequence. This application further discloses an OvXDM system. In a specific implementation of this application, decoding is performed by using the total least square method.

In an aspect of the disclosure, a method, a computer-readable medium, and an apparatus are provided. The apparatus may be a receiver. The receiver receives, from a transmitter, a signal carrying a constellation symbol. The receiver processes the received signal by a soft symbol demapper to generate a first soft output of a first set of bits and a second soft output of a second set of bits. The receiver processes the second soft output by a soft syndrome former to generate a third soft output of a third set of bits. The soft syndrome former corresponds to a shaping code applied at the transmitter. The receiver recovers bits represented by the constellation symbol based on the first soft output and the third soft output.

Methods are disclosed for improving communications on feedback transmission channels, in which there is a possibility of bit errors. The basic solutions to counter those errors are: proper design of the CSI vector quantizer indexing (i.e., the bit representation of centroid indices) in order to minimize impact of index errors, use of error detection techniques to expurgate the erroneous indices and use of other methods to recover correct indices.

Sequence detectors and detection methods are provided for detecting symbol values corresponding to a sequence of input samples obtained from an ISI channel. The sequence detector comprises a branch metric unit (BMU) and a path metric unit (PMU). The BMU, which comprises an initial set of pipeline stages, is adapted to calculate, for each input sample, branch metrics for respective possible transitions between states of a trellis. To calculate these branch metrics, the BMU selects hypothesized input values, each dependent on a possible symbol value for the input sample and L>0 previous symbol values corresponding to possible transitions between states of the trellis. The BMU then calculates differences between the input sample and each hypothesized input value. The BMU compares these differences and selects, as the branch metric for each possible transition, an optimum difference in dependence on a predetermined state in a survivor path through the trellis.

The present disclosure relates to an encoded signal demodulation method, apparatus, and device. Some embodiments of the present disclosure are beneficial to improving demodulation performance.

There is provided a decoder for sequentially decoding a data signal received through a transmission channel in a communication system, said data signal carrying transmitted symbols, said decoder comprising a symbol estimation unit (301) configured to determine estimated symbols representative of the transmitted symbols carried by the received signal from information stored in a stack, said symbol estimation unit (301) being configured to iteratively fill the stack by expanding child nodes of a selected node of a decoding tree comprising a plurality of nodes, each node of the decoding tree corresponding to a candidate component of a symbol of said data signal and each node being assigned a metric, the stack being filled at each iteration with a set of expanded child nodes and being ordered by increasing values of the metrics assigned to the nodes, the selected node for each iteration corresponding to the node being assigned the lowest metric in the stack, the decoder comprising a metric determination unit (302) configured to determine an initial metric for each child node of said set of expanded child nodes, wherein the decoder further comprises a modified metric calculation unit (303) configured to calculate a modified metric for at least one of the expanded child nodes from the metric associated with said expanded child node and a weighting coefficient, said weighting coefficient being a function of the level of said node in the decoding tree, the decoder assigning said modified metric to said at least one of the expanded child nodes.

Sequence detectors and detection methods are provided for detecting symbol values corresponding to a sequence of input samples obtained from an ISI channel. The sequence detector comprises a branch metric unit (BMU) and a path metric unit (PMU). The BMU, which comprises an initial set of pipeline stages, is adapted to calculate, for each input sample, branch metrics for respective possible transitions between states of a trellis. To calculate these branch metrics, the BMU selects hypothesized input values, each dependent on a possible symbol value for the input sample and L>0 previous symbol values corresponding to possible transitions between states of the trellis. The BMU then calculates differences between the input sample and each hypothesized input value. The BMU compares these differences and selects, as the branch metric for each possible transition, an optimum difference in dependence on a predetermined state in a survivor path through the trellis.