Data-Driven Ensembles for Deep and Hard-Decision Hybrid Decoding

Raviv, Tomer; Raviv, Nir; Be'ery, Y.

doi:10.1109/isit44484.2020.9173971

Cited by 17 publications

(9 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nonetheless, SICNet can also be combined with alternative techniques to allow a DNN-aided receiver to track time-varying channel conditions at a modest overhead. These include the application of meta-learning for optimizing the hyperparameters of the training algorithm [29]; the pre-training of multiple receivers as a deep ensemble [30]; and the usage of soft symbol-level outputs, rather than FEC decoding, as a measure of confidence for producing labels from data, as proposed in [31], [32]. We leave the study of the combination of SICNet with these methods to facilitate online training for future investigations.…”

Section: B Fec-aided Online Trainingmentioning

confidence: 99%

Deep Learning Based Successive Interference Cancellation for the Non-Orthogonal Downlink

Luong

Shlezinger

et al. 2022

IEEE Trans. Veh. Technol.

View full text Add to dashboard Cite

Non-orthogonal communications are expected to play a key role in future wireless systems. In downlink transmissions, the data symbols are broadcast from a base station to different users, which are superimposed with different power to facilitate high-integrity detection using successive interference cancellation (SIC). However, SIC requires accurate knowledge of both the channel model and channel state information (CSI), which may be difficult to acquire. We propose a deep learningaided SIC detector termed SICNet, which replaces the interference cancellation blocks of SIC by deep neural networks (DNNs). Explicitly, SICNet jointly trains its internal DNNaided blocks for inferring the soft information representing the interfering symbols in a data-driven fashion, rather than using hard-decision decoders as in classical SIC. As a result, SICNet reliably detects the superimposed symbols in the downlink of non-orthogonal systems without requiring any prior knowledge of the channel model, while being less sensitive to CSI uncertainty than its model-based counterpart. SICNet is also robust to changes in the number of users and to their power allocation. Furthermore, SICNet learns to produce accurate soft outputs, which facilitates improved soft-input error correction decoding compared to model-based SIC. Finally, we propose an online training method for SICNet under block fading, which exploits the channel decoding for accurately recovering online data labels for retraining, hence, allowing it to smoothly track the fading envelope without requiring dedicated pilots. Our numerical results show that SICNet approaches the performance of classical SIC under perfect CSI, while outperforming it under realistic CSI uncertainty.

show abstract

Section: B Fec-aided Online Trainingmentioning

confidence: 99%

Deep Learning Based Successive Interference Cancellation for the Non-Orthogonal Downlink

Luong

Shlezinger

et al. 2022

IEEE Trans. Veh. Technol.

View full text Add to dashboard Cite

show abstract

“…Nonetheless, ensembles also encompass computational complexity which is linear in the number of base learners, being unrealistic for practical considerations. To reduce complexity, our previous work [ 20 ] suggests to employ a low complexity gating decoder. This decoder allows one to uniquely map each input word to a single most fitting decoder, keeping the overall computation complexity low.…”

Section: A Data-driven Approach To Tbcc Decodingmentioning

confidence: 99%

“…One recent innovation, referred to as the ensemble of decoders [ 20 ], combined the benefits of model-based approach with the list decoding scheme. This ensemble is composed of learnable decoders, each one called an expert.…”

Section: Introductionmentioning

confidence: 99%

Deep Ensemble of Weighted Viterbi Decoders for Tail-Biting Convolutional Codes

Raviv

Schwartz

Be'ery

2021

Entropy

Self Cite

View full text Add to dashboard Cite

Tail-biting convolutional codes extend the classical zero-termination convolutional codes: Both encoding schemes force the equality of start and end states, but under the tail-biting each state is a valid termination. This paper proposes a machine learning approach to improve the state-of-the-art decoding of tail-biting codes, focusing on the widely employed short length regime as in the LTE standard. This standard also includes a CRC code. First, we parameterize the circular Viterbi algorithm, a baseline decoder that exploits the circular nature of the underlying trellis. An ensemble combines multiple such weighted decoders, and each decoder specializes in decoding words from a specific region of the channel words’ distribution. A region corresponds to a subset of termination states; the ensemble covers the entire states space. A non-learnable gating satisfies two goals: it filters easily decoded words and mitigates the overhead of executing multiple weighted decoders. The CRC criterion is employed to choose only a subset of experts for decoding purpose. Our method achieves FER improvement of up to 0.75 dB over the CVA in the waterfall region for multiple code lengths, adding negligible computational complexity compared to the circular Viterbi algorithm in high signal-to-noise ratios (SNRs).

show abstract

“…In (An et al, 2020), the Reed-Solomon neural decoder is introduced, which estimates the error of the received codewords, and adjust itself to do more accurate decoding. Neural Bose-Chaudhuri-Hocquenghem (BCH) codes decoding is introduced in (Kamassury & Silva, 2020;Nachmani & Wolf, 2019;Raviv et al, 2020).…”

Section: Error Correcting Codes With Deep Learningmentioning

confidence: 99%

Recovering AES Keys with a Deep Cold Boot Attack

Zimerman,

Nachmani,

Wolf

2021

Preprint

View full text Add to dashboard Cite

Cold boot attacks inspect the corrupted random access memory soon after the power has been shut down. While most of the bits have been corrupted, many bits, at random locations, have not. Since the keys in many encryption schemes are being expanded in memory into longer keys with fixed redundancies, the keys can often be restored. In this work, we combine a novel cryptographic variant of a deep error correcting code technique with a modified SAT solver scheme to apply the attack on AES keys. Even though AES consists of Rijndael S-box elements, that are specifically designed to be resistant to linear and differential cryptanalysis, our method provides a novel formalization of the AES key scheduling as a computational graph, which is implemented by a neural message passing network. Our results show that our methods outperform the state of the art attack methods by a very large margin.

show abstract

Data-Driven Ensembles for Deep and Hard-Decision Hybrid Decoding

Cited by 17 publications

References 18 publications

Deep Learning Based Successive Interference Cancellation for the Non-Orthogonal Downlink

Deep Learning Based Successive Interference Cancellation for the Non-Orthogonal Downlink

Deep Ensemble of Weighted Viterbi Decoders for Tail-Biting Convolutional Codes

Recovering AES Keys with a Deep Cold Boot Attack

Contact Info

Product

Resources

About