For lossy image compression systems, we develop an algorithm called iterative refinement, to improve the decoder’s reconstruction compared with standard decoding techniques. Specifically, we propose a recurrent neural network approach for nonlinear, iterative decoding. Our neural decoder, which can work with any encoder, employs self-connected memory units that make use of both causal and non-c...

We propose an elastic pipeline that can apply dynamic voltage scaling (DVS) to hardwired logic circuits. In order to demonstrate its feasibility, a hardwired H.264/AVC HDTV decoder is designed as a real-time application. An entropy decoding process is divided into context-based adaptive binary arithmetic coding (CABAC) and syntax element decoding (SED), which has advantages of smoothing workloa...

Context-free inference is a standard part of many NLP pipelines. Most approaches use a variant of the CYK dynamic programming algorithm to populate a chart structure with predicted nonterminals over each span. We can extract a parse tree from this chart in several ways. In this work, we compare two commonly-used decoding approaches (Viterbi and max-rule) with a minimum-bayes-risk (MBR) method w...

Our study aims at bringing a software-based alternative for decoding Reed Solomon over the MPE-IFEC in the DVB-SH system, at the cost of a lower complexity and a good performance. The MPE-IFEC is operating in the Link layer, where the lost packets are considered as erasures. In this context, we propose to recover losses by decoding on a graph the binary image of the Reed Solomon code over a Bin...

In the Software Radio context, the parametrization is becoming an important topic especially when it comes to multistandard designs. This paper capitalizes on the Common Operator technique to present new common structures for the FFT and FEC decoding algorithms. A key benefit of exhibiting common operators is the regular architecture it brings when implemented in a Common Operator Bank (COB). T...

So far in the course, we have seen three types of codes, summarized as follows: 1. Algebraic Codes • Work for worst-case errors. • Have polynomial time encoding and decoding algorithms 2. LDPCs • Work for worst-case errors as well, but a less many errors (smaller distance) compared to the Algebraic Codes. • Have linear time encoding and decoding algorithms and are simpler. 3. Polar Codes • Work...

Document Image Decoding (DID) refers to the process of document recognition within a communication theory framework. In this framework, a logical document structure is a message communicated by encoding the structure as an ideal image, transmitting the ideal image through a noisy channel, and decoding the degraded image into a logical structure as close to the original message as possible, on a...

Frame stacking is broadly applied in end-to-end neural network training like connectionist temporal classification (CTC), and it leads to more accurate models and faster decoding. However, it is not well-suited to conventional neural network based on context-dependent state acoustic model, if the decoder is unchanged. In this paper, we propose a novel frame retaining method which is applied in ...

AND TURBO CODES Javier Garcia-Frias and John D. Villasenor Electrical Engineering Department University of California, Los Angeles Abstract|Hidden Markov models have been widely used to statistically characterize sources and channels in communication systems. In this paper we will consider their application in the context of turbo codes. We will describe simpli ed techniques for modifying a dec...