Various implementations described herein relate to correcting errors in Dynamic Random Access Memory (DRAM). A memory controller uses an Error Correcting Code (ECC) to store an encoded data word within a DRAM die. The DRAM die is communicatively coupled the memory controller by a memory data bus. The DRAM die includes on-die error correction for data bits stored in the DRAM. Upon reading the encoded data word, the memory controller corrects and detects one or more errors. The one or more errors are introduced by at least one of the on-die error correction of the DRAM die or the memory data bus.

A convolutional code rate matching method and a communication apparatus are provided. A puncturing pattern of a second codeword at a second code rate is obtained based on a puncturing pattern of a first codeword at a first code rate. A second puncturing location set of the second codeword is a subset of a first puncturing location set of the first codeword. When a transmit device decreases a code rate from the first code rate to the second code rate, a redundant bit is sent at a location of a complementary set of the second puncturing location set relative to the first puncturing location set. Compared with the first puncturing location set, the second puncturing location set may obtain more incremental redundant bits, to decrease a channel encoding rate. This can improve decoding performance of a convolutional code.

Consistent with a further aspect of the present disclosure, previously encoded data is stored in a memory, and an encoder accesses both input data and previously encoded data to generate new encoded data or a new codeword. Each codeword is stored in a row of the memory, and with each newly generated codeword, each previously stored code word is shifted to an adjacent row of the memory. In one example, the memory is delineated as a plurality of blocks including rows and columns of bits. When generating a new code word, randomly selected columns of bits in the memory are read from randomly selected blocks of the memory and supplied to the encoder. In this manner the number of times the memory is access is reduced and power consumption is reduced.

Certain aspects of the present disclosure relate to techniques and apparatus for increasing decoding performance and/or reducing decoding complexity. An exemplary method generally includes receiving, via a wireless medium, a codeword encoded using a tailless convolutional code (TLCC) with a known start state, evaluating a set of decoding candidate paths through a trellis decoder that originate at the known start state of the TLCC, performing, for each of a plurality of the decoding candidate paths, a back trace from a respective end state to the known start state, and selecting one of the decoding candidate paths based, at least in part, on path metrics generated while performing the back trace. Other aspects, embodiments, and features are also claimed and described.

A non-systematic convolutional decoder of a convolutionally encoded multi-level data stream includes a shift register and two or more paths of exclusive-OR (XOR) gates, arranged to reconstruct an original input information stream, each path having a quantiser arranged to quantise the signal to two levels, and a set of XOR gates arranged to match an encoding path in an associated convolutional encoder, and a selector arranged to feed an output from each path to a single input of the shift register. If the paths have differing values at their output, the selector may choose the value from the path based upon a function of the multi-level signals associated with each path, such as the path with the largest absolute signal level. The decoder provides a simple means for decoding signals while allowing the signal to also or instead be decoded using e.g. a Viterbi decoder if higher performance is required.

Devices, systems and methods for convolutional precoding and decoding of polar codes are disclosed. An example method for error correction in a data processing system includes receiving a noisy codeword, the codeword having been generated based on an outer stream decodable code and an inner polar code and provided to a communication channel or a storage channel prior to reception by the decoder, the stream decodable code characterized by a trellis, and performing, based on the trellis, a list-decoding operation on the noisy codeword vector to generate a plurality of information symbols, the list-decoding operation being configured to traverse through a plurality of states at one or more stages of a plurality of decoding stages.

A transmission apparatus includes a signal processing circuit configured to obtain information data bits to be transmitted; add known information data bits to the information data bits to generate first data blocks; perform error-correction coding on the first data blocks to generate first coded data blocks including parity data blocks such that the first coded data blocks satisfy a first code rate; remove the known information data bits from the first coded data blocks to generate second coded data blocks, the second coded data blocks satisfying a second code rate different from the first code rate; and modulate the second coded data blocks using a modulation scheme to generate a modulated signal, which is then transmitted. A number of the known information data bits depends on a number of the information data bits such that the first code rate is fixed regardless of the number of the information data bits.

Data convolution for geographically diverse storage is disclosed. Data and corresponding convolutions of data can employ erasure coding to improve robustness of access to information represented in the data. For a peer group of chunks employing a given erasure coding scheme, access to the information represented in the data can be via accessible chunks and/or recovery of a less-accessible chunk, e.g., via a deconvolution operation, via a decoding operation, via a mix of deconvolution and decoding operations. The mix of deconvolution and decoding operations can enable recovery of a less-accessible chunk that cannot be recovered by either a deconvolution or decoding operation alone. This can improve access to information represented in less-available data.

A method of telecommunications includes the steps of receiving an encoded block having a plurality of values, dividing the received encoded block into a plurality of received segments, each received segment comprising at least two of the values, decoding each received segment by providing, for each received segment, a plurality of estimated encoded sequences, each estimated encoded sequence comprising at least two data units, merging estimated encoded sequences for consecutive segments to provide a plurality of candidate sequences, and selecting one of the plurality of candidate sequences by performing a closest fit calculation between the received encoded data block and each of the candidate sequences. The method is suitable for use in software-defined radios.

A transmission apparatus includes a signal processing circuit configured to obtain information data bits to be transmitted; add known information data bits to the information data bits to generate first data blocks; perform error-correction coding on the first data blocks to generate first coded data blocks including parity data blocks such that the first coded data blocks satisfy a first code rate; remove the known information data bits from the first coded data blocks to generate second coded data blocks, the second coded data blocks satisfying a second code rate different from the first code rate; and modulate the second coded data blocks using a modulation scheme to generate a modulated signal, which is then transmitted. A number of the known information data bits depends on a number of the information data bits such that the first code rate is fixed regardless of the number of the information data bits.