arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1804.06002cs.ITmath.IT
收藏

和积译码器中基于神经量化器的联合量化器优化

Joint Quantizer Optimization based on Neural Quantizer for Sum-Product Decoder

Tadashi Wadayama, Satoshi Takabe

摘要:

宽带数字通信系统的前端设备需要低精度的模数转换器(ADC)来实现,以降低其功耗。本文的目标是提出一种新的联合量化器优化方法,以最大限度地减少低精度量化器匹配的和积算法。其主要思想是引入一个量化器,其中包括一个前馈神经网络和软阶梯函数。由于软阶梯函数是可微的,并且在任何地方都具有非零梯度值,因此我们可以利用反向传播和随机梯度下降方法来训练量化器中的前馈神经网络。在有监督的训练阶段,关于信道输入和解码器输出的预期损失被最小化。实验结果表明,联合量化器优化方法成功地提供了一个8级量化器的低密度奇偶校验(LDPC)码,实现只有0.1 dB的性能损失相比,未量化的系统。

英文摘要:

A low-precision analog-to-digital converter (ADC) is required to implement a frontend device of wideband digital communication systems in order to reduce its power consumption. The goal of this paper is to present a novel joint quantizer optimization method for minimizing lower-precision quantizers matched to the sum-product algorithms. The principal idea is to introduce a quantizer that includes a feed-forward neural network and the soft staircase function. Since the soft staircase function is differentiable and has non-zero gradient values everywhere, we can exploit backpropagation and a stochastic gradient descent method to train the feed-forward neural network in the quantizer. The expected loss regarding the channel input and the decoder output is minimized in a supervised training phase. The experimental results indicate that the joint quantizer optimization method successfully provides an 8-level quantizer for a low-density parity-check (LDPC) code that achieves only a 0.1-dB performance loss compared to the unquantized system.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑