Matches in SemOpenAlex for { <https://semopenalex.org/work/W3013670461> ?p ?o ?g. }
Showing items 1 to 80 of
80
with 100 items per page.
- W3013670461 abstract "A 5G cellular system has evolved to support wider bandwidth with more resource elements (REs) in order to accommodate the emerging request for a very high date rate. A baseband modem needs to be developed to achieve less computational complexity per RE. Accordingly, this paper addresses the issue of choosing the least complexity multi-input multi-output (MIMO) symbol detector among multiple candidate symbol detectors whose performance is proportional to their complexity. The problem is formulated into the sequential decision of MIMO detectors for each RE in a transport block (TB) as a Markov decision process with a constraint. The decision at each RE is converted to a complexity score whose sum across all REs in a TB is to be maximized, while a global constraint of block error rate is imposed with respect to the most complex symbol detector, i.e., maximum likelihood (ML). The reinforcement learning is applied to solve this selection problem via the value-based approach using the start-of-the-art deep Q learning method. A 5G new radio link level simulator is integrated with a Q-value network training process. Simulation results show that the near-optimal performance can still be achieved, while only a small number of REs adopt the detector of ML and most of REs are in favor of less complexity detectors. As a result, the amount of Euclidean distance calculations is reduced by more than 3 times with a negligible performance gap. This implies that a baseband modem could be much simplified and ensure to achieve 5G services with less power consumption." @default.
- W3013670461 created "2020-04-03" @default.
- W3013670461 creator A5044969532 @default.
- W3013670461 creator A5056967242 @default.
- W3013670461 date "2020-02-01" @default.
- W3013670461 modified "2023-09-26" @default.
- W3013670461 title "MIMO-OFDM Detector Selection using Reinforcement Learning" @default.
- W3013670461 cites W2022636335 @default.
- W3013670461 cites W2106261932 @default.
- W3013670461 cites W2113012465 @default.
- W3013670461 cites W2116631972 @default.
- W3013670461 cites W2125385012 @default.
- W3013670461 cites W2128548897 @default.
- W3013670461 cites W2286612019 @default.
- W3013670461 cites W4211221179 @default.
- W3013670461 cites W4250589301 @default.
- W3013670461 doi "https://doi.org/10.1109/icnc47757.2020.9049738" @default.
- W3013670461 hasPublicationYear "2020" @default.
- W3013670461 type Work @default.
- W3013670461 sameAs 3013670461 @default.
- W3013670461 citedByCount "1" @default.
- W3013670461 countsByYear W30136704612023 @default.
- W3013670461 crossrefType "proceedings-article" @default.
- W3013670461 hasAuthorship W3013670461A5044969532 @default.
- W3013670461 hasAuthorship W3013670461A5056967242 @default.
- W3013670461 hasConcept C105795698 @default.
- W3013670461 hasConcept C106189395 @default.
- W3013670461 hasConcept C11413529 @default.
- W3013670461 hasConcept C127162648 @default.
- W3013670461 hasConcept C154945302 @default.
- W3013670461 hasConcept C159886148 @default.
- W3013670461 hasConcept C179799912 @default.
- W3013670461 hasConcept C207987634 @default.
- W3013670461 hasConcept C2524010 @default.
- W3013670461 hasConcept C2776257435 @default.
- W3013670461 hasConcept C2777210771 @default.
- W3013670461 hasConcept C33923547 @default.
- W3013670461 hasConcept C41008148 @default.
- W3013670461 hasConcept C56296756 @default.
- W3013670461 hasConcept C57273362 @default.
- W3013670461 hasConcept C65165936 @default.
- W3013670461 hasConcept C76155785 @default.
- W3013670461 hasConcept C94915269 @default.
- W3013670461 hasConcept C97541855 @default.
- W3013670461 hasConceptScore W3013670461C105795698 @default.
- W3013670461 hasConceptScore W3013670461C106189395 @default.
- W3013670461 hasConceptScore W3013670461C11413529 @default.
- W3013670461 hasConceptScore W3013670461C127162648 @default.
- W3013670461 hasConceptScore W3013670461C154945302 @default.
- W3013670461 hasConceptScore W3013670461C159886148 @default.
- W3013670461 hasConceptScore W3013670461C179799912 @default.
- W3013670461 hasConceptScore W3013670461C207987634 @default.
- W3013670461 hasConceptScore W3013670461C2524010 @default.
- W3013670461 hasConceptScore W3013670461C2776257435 @default.
- W3013670461 hasConceptScore W3013670461C2777210771 @default.
- W3013670461 hasConceptScore W3013670461C33923547 @default.
- W3013670461 hasConceptScore W3013670461C41008148 @default.
- W3013670461 hasConceptScore W3013670461C56296756 @default.
- W3013670461 hasConceptScore W3013670461C57273362 @default.
- W3013670461 hasConceptScore W3013670461C65165936 @default.
- W3013670461 hasConceptScore W3013670461C76155785 @default.
- W3013670461 hasConceptScore W3013670461C94915269 @default.
- W3013670461 hasConceptScore W3013670461C97541855 @default.
- W3013670461 hasLocation W30136704611 @default.
- W3013670461 hasOpenAccess W3013670461 @default.
- W3013670461 hasPrimaryLocation W30136704611 @default.
- W3013670461 hasRelatedWork W2017312071 @default.
- W3013670461 hasRelatedWork W2028779886 @default.
- W3013670461 hasRelatedWork W2047436358 @default.
- W3013670461 hasRelatedWork W2067697423 @default.
- W3013670461 hasRelatedWork W2074400693 @default.
- W3013670461 hasRelatedWork W2100519948 @default.
- W3013670461 hasRelatedWork W2170852358 @default.
- W3013670461 hasRelatedWork W2525668591 @default.
- W3013670461 hasRelatedWork W3213838085 @default.
- W3013670461 hasRelatedWork W4313591620 @default.
- W3013670461 isParatext "false" @default.
- W3013670461 isRetracted "false" @default.
- W3013670461 magId "3013670461" @default.
- W3013670461 workType "article" @default.