ترغب بنشر مسار تعليمي؟ اضغط هنا

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

114   0   0.0 ( 0 )
 نشر من قبل Tomer Raviv
 تاريخ النشر 2020
  مجال البحث الهندسة المعلوماتية
والبحث باللغة English




اسأل ChatGPT حول البحث

Ensemble models are widely used to solve complex tasks by their decomposition into multiple simpler tasks, each one solved locally by a single member of the ensemble. Decoding of error-correction codes is a hard problem due to the curse of dimensionality, leading one to consider ensembles-of-decoders as a possible solution. Nonetheless, one must take complexity into account, especially in decoding. We suggest a low-complexity scheme where a single member participates in the decoding of each word. First, the distribution of feasible words is partitioned into non-overlapping regions. Thereafter, specialized experts are formed by independently training each member on a single region. A classical hard-decision decoder (HDD) is employed to map every word to a single expert in an injective manner. FER gains of up to 0.4dB at the waterfall region, and of 1.25dB at the error floor region are achieved for two BCH(63,36) and (63,45) codes with cycle-reduced parity-check matrices, compared to the previous best result of the paper Active Deep Decoding of Linear Codes.



قيم البحث

اقرأ أيضاً

We consider probabilistic amplitude shaping (PAS) as a means of increasing the spectral efficiency of fiber-optic communication systems. In contrast to previous works in the literature, we consider probabilistic shaping with hard decision decoding (H DD). In particular, we apply the PAS recently introduced by Bocherer emph{et al.} to a coded modulation (CM) scheme with bit-wise HDD that uses a staircase code as the forward error correction code. We show that the CM scheme with PAS and staircase codes yields significant gains in spectral efficiency with respect to the baseline scheme using a staircase code and a standard constellation with uniformly distributed signal points. Using a single staircase code, the proposed scheme achieves performance within $0.57$--$1.44$ dB of the corresponding achievable information rate for a wide range of spectral efficiencies.
We analyze the achievable information rates (AIRs) for coded modulation schemes with QAM constellations with both bit-wise and symbol-wise decoders, corresponding to the case where a binary code is used in combination with a higher-order modulation u sing the bit-interleaved coded modulation (BICM) paradigm and to the case where a nonbinary code over a field matched to the constellation size is used, respectively. In particular, we consider hard decision decoding, which is the preferable option for fiber-optic communication systems where decoding complexity is a concern. Recently, Liga emph{et al.} analyzed the AIRs for bit-wise and symbol-wise decoders considering what the authors called emph{hard decision decoder} which, however, exploits emph{soft information} of the transition probabilities of discrete-input discrete-output channel resulting from the hard detection. As such, the complexity of the decoder is essentially the same as the complexity of a soft decision decoder. In this paper, we analyze instead the AIRs for the standard hard decision decoder, commonly used in practice, where the decoding is based on the Hamming distance metric. We show that if standard hard decision decoding is used, bit-wise decoders yield significantly higher AIRs than symbol-wise decoders. As a result, contrary to the conclusion by Liga emph{et al.}, binary decoders together with the BICM paradigm are preferable for spectrally-efficient fiber-optic systems. We also design binary and nonbinary staircase codes and show that, in agreement with the AIRs, binary codes yield better performance.
We propose a deep-learning approach for the joint MIMO detection and channel decoding problem. Conventional MIMO receivers adopt a model-based approach for MIMO detection and channel decoding in linear or iterative manners. However, due to the comple x MIMO signal model, the optimal solution to the joint MIMO detection and channel decoding problem (i.e., the maximum likelihood decoding of the transmitted codewords from the received MIMO signals) is computationally infeasible. As a practical measure, the current model-based MIMO receivers all use suboptimal MIMO decoding methods with affordable computational complexities. This work applies the latest advances in deep learning for the design of MIMO receivers. In particular, we leverage deep neural networks (DNN) with supervised training to solve the joint MIMO detection and channel decoding problem. We show that DNN can be trained to give much better decoding performance than conventional MIMO receivers do. Our simulations show that a DNN implementation consisting of seven hidden layers can outperform conventional model-based linear or iterative receivers. This performance improvement points to a new direction for future MIMO receiver design.
Deep learning (DL) based autoencoder is a promising architecture to implement end-to-end communication systems. One fundamental problem of such systems is how to increase the transmission rate. Two new schemes are proposed to address the limited data rate issue: adaptive transmission scheme and generalized data representation (GDR) scheme. In the first scheme, an adaptive transmission is designed to select the transmission vectors for maximizing the data rate under different channel conditions. The block error rate (BLER) of the first scheme is 80% lower than that of the conventional one-hot vector scheme. This implies that higher data rate can be achieved by the adaptive transmission scheme. In the second scheme, the GDR replaces the conventional one-hot representation. The GDR scheme can achieve higher data rate than the conventional one-hot vector scheme with comparable BLER performance. For example, when the vector size is eight, the proposed GDR scheme can double the date rate of the one-hot vector scheme. Besides, the joint scheme of the two proposed schemes can create further benefits. The effect of signal-to-noise ratio (SNR) is analyzed for these DL-based communication systems. Numerical results show that training the autoencoder using data set with various SNR values can attain robust BLER performance under different channel conditions.
We revisit the idea of using deep neural networks for one-shot decoding of random and structured codes, such as polar codes. Although it is possible to achieve maximum a posteriori (MAP) bit error rate (BER) performance for both code families and for short codeword lengths, we observe that (i) structured codes are easier to learn and (ii) the neural network is able to generalize to codewords that it has never seen during training for structured, but not for random codes. These results provide some evidence that neural networks can learn a form of decoding algorithm, rather than only a simple classifier. We introduce the metric normalized validation error (NVE) in order to further investigate the potential and limitations of deep learning-based decoding with respect to performance and complexity.
التعليقات
جاري جلب التعليقات جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
mircosoft-partner

هل ترغب بارسال اشعارات عن اخر التحديثات في شمرا-اكاديميا