TY - JOUR
T1 - ViterbiNet
T2 - A Deep Learning Based Viterbi Algorithm for Symbol Detection
AU - Shlezinger, Nir
AU - Farsad, Nariman
AU - Eldar, Yonina C.
AU - Goldsmith, Andrea J.
N1 - Funding Information:
Manuscript received May 26, 2019; revised November 6, 2019; accepted January 30, 2020. Date of publication February 14, 2020; date of current version May 8, 2020. This work was supported in part by the U.S.—Israel Binational Science Foundation under Grant 2026094, in part by the Israel Science Foundation under Grant 0100101, and in part by the Office of the Naval Research under Grant 18-1-2191. This article was presented at the 2019 IEEE International Workshop on Signal Processing Advances in Wireless Communications (SPAWC), Cannes, France. The associate editor coordinating the review of this article and approving it for publication was X. Chen. (Corresponding author: Nir Shlezinger.) Nir Shlezinger and Yonina C. Eldar are with the Faculty of Math and CS, Weizmann Institute of Science, Rehovot 7610001, Israel (e-mail: nirshlezinger1@gmail.com; yonina@weizmann.ac.il).
Publisher Copyright:
© 2002-2012 IEEE.
PY - 2020/5
Y1 - 2020/5
N2 - Symbol detection plays an important role in the implementation of digital receivers. In this work, we propose ViterbiNet, which is a data-driven symbol detector that does not require channel state information (CSI). ViterbiNet is obtained by integrating deep neural networks (DNNs) into the Viterbi algorithm. We identify the specific parts of the Viterbi algorithm that depend on the channel model, and design a DNN to implement only those computations, leaving the rest of the algorithm structure intact. We then propose a meta-learning based approach to train ViterbiNet online based on recent decisions, allowing the receiver to track dynamic channel conditions without requiring new training samples for every coherence block. Our numerical evaluations demonstrate that the performance of ViterbiNet, which is ignorant of the CSI, approaches that of the CSI-based Viterbi algorithm, and is capable of tracking time-varying channels without needing instantaneous CSI or additional training data. Moreover, unlike conventional Viterbi detection, ViterbiNet is robust to CSI uncertainty, and it can be reliably implemented in complex channel models with constrained computational burden. More broadly, our results demonstrate the conceptual benefit of designing communication systems that integrate DNNs into established algorithms.
AB - Symbol detection plays an important role in the implementation of digital receivers. In this work, we propose ViterbiNet, which is a data-driven symbol detector that does not require channel state information (CSI). ViterbiNet is obtained by integrating deep neural networks (DNNs) into the Viterbi algorithm. We identify the specific parts of the Viterbi algorithm that depend on the channel model, and design a DNN to implement only those computations, leaving the rest of the algorithm structure intact. We then propose a meta-learning based approach to train ViterbiNet online based on recent decisions, allowing the receiver to track dynamic channel conditions without requiring new training samples for every coherence block. Our numerical evaluations demonstrate that the performance of ViterbiNet, which is ignorant of the CSI, approaches that of the CSI-based Viterbi algorithm, and is capable of tracking time-varying channels without needing instantaneous CSI or additional training data. Moreover, unlike conventional Viterbi detection, ViterbiNet is robust to CSI uncertainty, and it can be reliably implemented in complex channel models with constrained computational burden. More broadly, our results demonstrate the conceptual benefit of designing communication systems that integrate DNNs into established algorithms.
KW - Symbol detection
KW - machine learning (ML)
UR - http://www.scopus.com/inward/record.url?scp=85081954619&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85081954619&partnerID=8YFLogxK
U2 - 10.1109/TWC.2020.2972352
DO - 10.1109/TWC.2020.2972352
M3 - Article
AN - SCOPUS:85081954619
SN - 1536-1276
VL - 19
SP - 3319
EP - 3331
JO - IEEE Transactions on Wireless Communications
JF - IEEE Transactions on Wireless Communications
IS - 5
M1 - 8999801
ER -