… Abstract—Although the sphere decoder (SD) is a … fast deep learning (DL)aided SD (FDL-SD)
and fast DL-aided K-best SD (KSD, FDLKSD) algorithms. Therein, the major application of …

… The application of a deep neural network to reduce the computational complexity of the
conventional belief propagation detector is proposed in [8], and the orthogonal approximate …

… learning capability of DL and the near-optimal performance brought by SD, we propose the
sparsely connected deep neural network … least squares with application to sphere decoding,” …

… In this paper, we investigate the use of deep learning (DL) to reduce the detection … Second,
we propose a DL-based list sphere decoding (DL-LSD) algorithm that selects and updates …

… sphere decoding algorithms and their performance metrics are illustrated. Also, various
sphere decoding … suggested the application of deep neural networks (DNN) to massive MIMO …

… reduce the decoding latency of the low complexity SD-SDS. We propose a deep learning-…
To estimate the complex M-QAM symbol, ˆxq using the QR decoder, we use the following …

… that describes a deep learning based SD [23]. The aim of that work is to learn the minimum
… We believe the investigation of this application not only enables a powerfull new application …

… We compare our results with the soft sphere decoder [26]. … of the sphere decoder depends
on the SNR, so we use the … and additions of the sphere decoder is 5.5 ⇥ 106. Considering …

… been published on this topic proving that the application of deep learning to MIMO decoding
can drastically reduce the complexity of the decoder while preserving its performance [23]–[…

… -DNN algorithms when the spheredecoder detection complexity is much greater than that
of the SDS RNN function approximator. This implies that we can use the SDS RNN function …