Abstract:
A proactive fault-tolerant scheme which improves performance and energy efficiency for NoCs. The fault-tolerant scheme allows routers to switch among several different fault-tolerant operations. Each operation mode has different trade-offs among fault-tolerant capability, retransmission traffic, latency, and energy efficiency. Another example provides a proactive, dynamic control policy to balance and optimize the dynamic interactions and trade-offs. The example control policy uses example machine learning algorithm called reinforcement learning (RL). The example RL-based controller independently observes a set of NoC system parameters at runtime, and over time they evolve optimal per-router control policies. By automatically and optimally switching among the four fault-tolerant modes, the trained control policy results in minimizing system level network latency and maximizing energy efficiency while detecting and correcting errors.
Abstract:
Systems and techniques for forward error correction decode processing power reduction are described herein. A first codeword is identified in a network transmission. A conditional decoding window is generated that begins with the first codeword. The conditional decoding window is aligned with a frame of the network transmission. A report is generated that includes an indication of relevance of the conditional decoding window to a network device. The conditional decoding window and the report are transmitted to the network device. A conditional decoding window is received that includes a first codeword. The first codeword is decoded to determine a second codeword. A report is obtained that includes a codeword relevance bit for the second codeword. Upon a determination that the codeword relevance bit indicates that the second codeword is irrelevant, the second codeword is discarded without decoding the second codeword.
Abstract:
A transmitter includes a data stream encoder layer having an output and a pattern generator having a bit pattern output. The transmitter further includes a first multiplexer having first and second inputs and a first multiplexer output. The first input is coupled to the output of the data stream encoder layer, and the second input is coupled to the bit pattern output of the pattern generator. While at least a portion of the data stream encoder layer is powered down, the pattern generator is configured to provide bit patterns on its bit pattern output, a control signal to the first multiplexer is configured to select the second input of the first multiplexer, and the first multiplexer is configured to output the bit patterns on the output of the first multiplexer.
Abstract:
The present invention provides a method of transmitting broadcast signals. The method includes, encoding, by an encoder, data in plural PLPs (Physical Layer Pipes); processing, by a framing and interleaving block, the encoded data in the plural PLPs, wherein the processing the encoded data further includes: time interleaving, by a time interleaver, the encoded data in the plural PLPs, frame mapping, by a framer, the time interleaved data onto at least one signal frame, and frequency interleaving, by a frequency interleaver, data in the at least one signal frame; and waveform modulating, by a waveform generation block, the frequency interleaved data in the at least one signal frame and transmitting, by the waveform generation block, broadcast signals having the waveform modulated data.
Abstract:
A method, apparatus and system for feeding back early stop decoding are provided. The method includes: a terminal side adjusting encoded TFCI bits, and sending the adjusted TFCI bits to a NodeB side via a TFCI domain of an uplink DPCCH (S302); after sending the adjusted TFCI bits to the NodeB side, the terminal side performing a decoding operation on a downlink DPCH, and feeding back, via an idle TFCI domain of the uplink DPCCH, a decoding result to the NodeB side (304). By applying the technical solution, at least one of the problems in the related art that a NodeB cannot obtain a TFCI in time and a terminal side cannot feed back a downlink decoding result in time during early stop decoding can be solved.
Abstract:
The present invention provides a method of transmitting broadcast signals. The method includes, encoding, by an encoder, data in plural PLPs (Physical Layer Pipes); processing, by a framing and interleaving block, the encoded data in the plural PLPs, wherein the processing the encoded data further includes: time interleaving, by a time interleaver, the encoded data in the plural PLPs, frame mapping, by a framer, the time interleaved data onto at least one signal frame, and frequency interleaving, by a frequency interleaver, data in the at least one signal frame; and waveform modulating, by a waveform generation block, the frequency interleaved data in the at least one signal frame and transmitting, by the waveform generation block, broadcast signals having the waveform modulated data.
Abstract:
The present invention relates to a method of transmitting and a method of receiving signals and corresponding apparatus. One aspect of the present invention relates to an efficient layer 1 (L1) processing method for a transmitter and a receiver using data slices.
Abstract:
Techniques herein support enhanced multi-rate encoding and decoding of signals in multiple formats. In one embodiment, input data is received at a first device at one of a plurality of data rates. Encoder units are activated to produce streams of encoded input data. The encoder units are configured to operate at the same data rate. Differential encoding operations are performed to produce an encoded output stream. The encoded output stream is modulated for transmission to a second device. In another embodiment, a first device receives an encoded data stream that is transmitted from a second device. The modulated data stream includes encoded data at one of a plurality of data rates. Differential decoding is performed on the encoded data by activating one or more of a plurality of decoder units, where each of the plurality of decoder units is configured to operate at the same rate.
Abstract:
Certain aspects of the present disclosure relate to a technique for power saving at mobile stations in Very High Throughput (VHT) systems using signal field bits of a preamble with a limited transmission overhead. The present disclosure proposes a method of indicating to a destination through a preamble cyclic redundancy check (CRC) checksum that the destination may be the intended destination of a transmission packet. The present disclosure also proposes a method of indicating to a destination through a preamble CRC that the destination may not be the intended destination of a transmission packet. In this case, decoding of the received packet can be terminated in order to save power at a receiving device.
Abstract:
A cable transmitter supports a number of low density parity check (LDPC) coding rates, e.g., ¼, ⅓, ⅖, ½, ⅗, ⅔, ¾, ⅘, ⅚, 8/9 and 9/10; and supports a number of quadrature amplitude modulation (QAM) schemes, e.g., 4-QAM, 16-QAM, 64-QAM, 256-QAM, 1024-QAM or higher. For a selected modulation scheme, the cable transmitter selects between using a non-uniform symbol constellation or a uniform symbol constellation as a function of a selected coding rate.