Realizing Neural Decoder at the Edge With Ensembled BNN
Devannagari Vikas, Nancy Nayak, Sheetal Kalyani · IEEE Communications Letters · 2021
We propose extreme compression techniques like binarization, ternarization for Turbo code based Neural Decoders such as TurboAE. These methods reduce memory and computation by a factor of 64 and perform better than the quantized (with 1-bit or 2-bits) Neural Decoders. However, because of the limited representation capability of the Binary and Ternary networks, the performance is not as good as the real-valued decoder. To fill this gap, we further propose to ensemble 4 such weak performers to deploy in the edge to achieve a performance similar to the real-valued network. These ensemble decoders give a saving of 16 and 64 times in memory and computation respectively and help achieve performance the same as real-valued TurboAE.