A method and system for training a neural network are herein provided. According to one embodiment, a method includes generating a first labelled dataset corresponding to a first modulation scheme and a second labelled dataset corresponding to a second modulation scheme, determining a first gradient of a cost function between a first neural network layer and a second neural network layer based on back-propagation using the first labelled dataset and the second labelled dataset, and determining a second gradient of the cost function between the second neural network layer and a first set of nodes of a third neural network layer based on back-propagation using the first labelled dataset. The first set of nodes of the third neural network layer correspond to a first plurality of detector classes associated with the first modulation scheme.

A method of soft-decision decoding including training a machine learning agent with communication signal training data; providing to the trained machine learning agent a signal that has been received via a communications channel; operating the machine learning agent to determine respective probabilities that the received signal corresponds to each of a plurality of symbols; and, based on the determined probabilities, performing soft decision decoding on the received signal.

Aspects of the present disclosure include methods for communication using a MIMO channel, such as an acoustic channel for underwater communication. An acoustic receiver may receive a signal comprising information encoded in at least one transmitted symbol. Using a two-layer iterative process, the at least one transmitted symbol is estimated. The first layer of the two-layer process uses iterative exchanges of soft-decisions between an adaptive turbo equalizer and a MAP decoder. The second layer of the two-layer process uses a data-reuse procedure that adapts an equalizer vector of both a feedforward filter and a serial interference cancellation filter of the adaptive turbo equalizer using a posteriori soft decisions of the at least one transmitted symbol. After a plurality of iterations, a hard decision of the bits encoded on the at least one transmitted symbol is output from the MAP decoder.

A transmission method simultaneously transmitting a first modulated signal and a second modulated signal at a common frequency performs precoding on both signals using a fixed precoding matrix and regularly changes the phase of at least one of the signals, thereby improving received data signal quality for a reception device.

System and method embodiments are provided for a receiver for circularly convolved signals. In an embodiment, a universal decoder for a circularly convolved signal includes a first decoder configured to decode the circularly convolved signal; a second decoder configured to decode a plurality of symbol lengths signal from a first portion of the circularly convolved signal, wherein the plurality of symbol lengths signal is time aligned with the circularly convolved signal before passing through the second decoder; and an adder component configured to sum a first decoder output coming from the first decoder and a second decoder output coming from the second decoder to produce a symbol value from which a log likelihood ratio (LLR) output is obtained.

Methods and apparatus for detecting ACK/NACK bits with dual list-RM decoder and symbol regeneration for PUCCH format 3. In an exemplary embodiment, a method is provided for detected ACK/NACK bits received in a long-term evolution (LTE) physical uplink control channel (PUCCH) Format 3 uplink transmission. The method includes generating Top-M ACK candidates from a descrambled bit stream, regenerating Top-M candidate symbols from the Top-M ACK candidates, calculating channel estimates for the Top-M candidate symbols, combining to the channel estimates generate a combined metric; and searching the combined metric to determine detected ACK bits.

Provided are a reception device, a reception method and a transmission reception system capable of reducing the influence of distortion in a received signal and achieving high demodulation performance without performing a computation process having a great amount of calculations. The reception device receives a signal containing a known signal part and a data part, and includes a conversion unit that converts the signal received by a reception unit into a digital signal, a region determination unit that determines a nonuse region which is a periodic region containing distortion in the digital signal, on a basis of a first digital signal in the known signal part contained in the digital signal and a known signal held in advance, and a demodulation unit that performs demodulation on the digital signal by using a second digital signal in a region other than the nonuse region in the digital signal.

A data transmission device includes a de-interleaver configured to receive, from a host device at a first data rate, a data stream including encoded data, de-interleave the data stream into a plurality of forward error correction (FEC) data streams, and output the plurality of FEC data streams at a second data rate less than the first data rate. Each of a plurality of interleavers is configured to interleave a respective one of the plurality of FEC data streams into an intermediate data stream including first data blocks and second data blocks. An encoder module configured to generate, for each of the intermediate data streams, FEC blocks including a first parity section and a first data section, the first parity section including a first parity bit corresponding to the first data blocks and a second parity bit corresponding to the second data blocks, and the first data section including the first data blocks and the second data blocks, and output the FEC blocks at the second data rate.

A reception device includes: a receiver that receives a multiplexed signal; a first demapper that demaps the multiplexed signal, with a second modulated symbol stream of a second data series being included in the multiplexed signal as an undefined signal component, to generate a first bit likelihood stream of a first data series; a second demapper that demaps the multiplexed signal, with a first modulated symbol stream of the first data series being included in the multiplexed signal as an undefined signal component, to generate a second bit likelihood stream of the second data series; a first decoder that performs error control decoding on the first bit likelihood stream to derive the first data series; and a second decoder that performs error control decoding on the second bit likelihood stream to derive the second data series.

Methods and apparatus for providing a demapping system with phase compensation to demap uplink transmissions. In an embodiment, a method is provided that includes detecting a processing type associated with a received uplink transmission, and when the detected processing type is a first processing type then performing the following operations: removing resource elements containing reference signals from the uplink transmission; layer demapping remaining resource elements of the uplink transmission into two or more layers; phase compensating all layers to generate phase compensated layers; and soft-demapping all phase compensated layers to produce phase compensated soft-demapped bits.