Matches in SemOpenAlex for { <https://semopenalex.org/work/W3166300576> ?p ?o ?g. }
Showing items 1 to 99 of
99
with 100 items per page.
- W3166300576 endingPage "26165" @default.
- W3166300576 startingPage "26165" @default.
- W3166300576 abstract "The optical wireless communication (OWC) system has been widely studied as a promising solution for high-speed indoor applications. The transmitter diversity scheme has been proposed to improve the performance of high-speed OWC systems. However, the transmitter diversity is vulnerable to the delay of multiple channels. Recently neural networks have been studied to realize delay-tolerant indoor OWC systems, where long-short term memory (LSTM) and attention-augmented LSTM (ALSTM) recurrent neural networks (RNNs) have shown their capabilities. However, they have high computation complexity and long computation latency. In this paper, we propose a low complexity delay-tolerant RNN scheme for indoor OWC systems. In particular, an RNN with parallelized structure is proposed to reduce the computation cost. The proposed RNN schemes show comparable capability to the more complicated ALSTM, where a bit-error-rate (BER) performance within the forward-error-correction (FEC) limit is achieved for up to 5.5 symbol periods delays. In addition, previously studied LSTM/ALSTM schemes are implemented using high-end GPUs, which have high cost, high power consumption, and long processing latency. To solve these practical limitations, in this paper we further propose and demonstrate the FPGA-based RNN hardware accelerator for delay-tolerant indoor OWC systems. To optimize the processing latency and power consumption, we also propose two optimization methods: the parallel implementation with triple-phase clocking and the stream-in based computation with additive input data insertion. Results show that the FPGA-based RNN hardware accelerator with the proposed optimization methods achieves 96.75% effective latency reduction and 90.7% lower energy consumption per symbol compared with the FPGA-based RNN hardware accelerator without optimization. Compared to the GPU implementation, the latency is reduced by about 61% and the power consumption is reduced by about 58.1%." @default.
- W3166300576 created "2021-06-22" @default.
- W3166300576 creator A5000618392 @default.
- W3166300576 creator A5018685460 @default.
- W3166300576 creator A5064266389 @default.
- W3166300576 creator A5066393817 @default.
- W3166300576 creator A5085279545 @default.
- W3166300576 date "2021-07-30" @default.
- W3166300576 modified "2023-10-17" @default.
- W3166300576 title "Recurrent neural network FPGA hardware accelerator for delay-tolerant indoor optical wireless communications" @default.
- W3166300576 cites W2485863112 @default.
- W3166300576 cites W2613390430 @default.
- W3166300576 cites W2784210264 @default.
- W3166300576 cites W2919338298 @default.
- W3166300576 cites W2953554277 @default.
- W3166300576 cites W2962687477 @default.
- W3166300576 cites W2963042536 @default.
- W3166300576 cites W2963841205 @default.
- W3166300576 cites W3007570230 @default.
- W3166300576 cites W3024364973 @default.
- W3166300576 cites W3097979045 @default.
- W3166300576 cites W3103926693 @default.
- W3166300576 doi "https://doi.org/10.1364/oe.427250" @default.
- W3166300576 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/34614928" @default.
- W3166300576 hasPublicationYear "2021" @default.
- W3166300576 type Work @default.
- W3166300576 sameAs 3166300576 @default.
- W3166300576 citedByCount "6" @default.
- W3166300576 countsByYear W31663005762022 @default.
- W3166300576 countsByYear W31663005762023 @default.
- W3166300576 crossrefType "journal-article" @default.
- W3166300576 hasAuthorship W3166300576A5000618392 @default.
- W3166300576 hasAuthorship W3166300576A5018685460 @default.
- W3166300576 hasAuthorship W3166300576A5064266389 @default.
- W3166300576 hasAuthorship W3166300576A5066393817 @default.
- W3166300576 hasAuthorship W3166300576A5085279545 @default.
- W3166300576 hasBestOaLocation W31663005761 @default.
- W3166300576 hasConcept C111335779 @default.
- W3166300576 hasConcept C11413529 @default.
- W3166300576 hasConcept C127162648 @default.
- W3166300576 hasConcept C147168706 @default.
- W3166300576 hasConcept C154945302 @default.
- W3166300576 hasConcept C2524010 @default.
- W3166300576 hasConcept C2781305180 @default.
- W3166300576 hasConcept C31258907 @default.
- W3166300576 hasConcept C33923547 @default.
- W3166300576 hasConcept C41008148 @default.
- W3166300576 hasConcept C42935608 @default.
- W3166300576 hasConcept C45374587 @default.
- W3166300576 hasConcept C46637626 @default.
- W3166300576 hasConcept C47798520 @default.
- W3166300576 hasConcept C50644808 @default.
- W3166300576 hasConcept C555944384 @default.
- W3166300576 hasConcept C76155785 @default.
- W3166300576 hasConcept C79403827 @default.
- W3166300576 hasConcept C82876162 @default.
- W3166300576 hasConcept C9390403 @default.
- W3166300576 hasConceptScore W3166300576C111335779 @default.
- W3166300576 hasConceptScore W3166300576C11413529 @default.
- W3166300576 hasConceptScore W3166300576C127162648 @default.
- W3166300576 hasConceptScore W3166300576C147168706 @default.
- W3166300576 hasConceptScore W3166300576C154945302 @default.
- W3166300576 hasConceptScore W3166300576C2524010 @default.
- W3166300576 hasConceptScore W3166300576C2781305180 @default.
- W3166300576 hasConceptScore W3166300576C31258907 @default.
- W3166300576 hasConceptScore W3166300576C33923547 @default.
- W3166300576 hasConceptScore W3166300576C41008148 @default.
- W3166300576 hasConceptScore W3166300576C42935608 @default.
- W3166300576 hasConceptScore W3166300576C45374587 @default.
- W3166300576 hasConceptScore W3166300576C46637626 @default.
- W3166300576 hasConceptScore W3166300576C47798520 @default.
- W3166300576 hasConceptScore W3166300576C50644808 @default.
- W3166300576 hasConceptScore W3166300576C555944384 @default.
- W3166300576 hasConceptScore W3166300576C76155785 @default.
- W3166300576 hasConceptScore W3166300576C79403827 @default.
- W3166300576 hasConceptScore W3166300576C82876162 @default.
- W3166300576 hasConceptScore W3166300576C9390403 @default.
- W3166300576 hasFunder F4320334704 @default.
- W3166300576 hasIssue "16" @default.
- W3166300576 hasLocation W31663005761 @default.
- W3166300576 hasOpenAccess W3166300576 @default.
- W3166300576 hasPrimaryLocation W31663005761 @default.
- W3166300576 hasRelatedWork W2067951144 @default.
- W3166300576 hasRelatedWork W2121697916 @default.
- W3166300576 hasRelatedWork W2138540539 @default.
- W3166300576 hasRelatedWork W2167910806 @default.
- W3166300576 hasRelatedWork W2594743321 @default.
- W3166300576 hasRelatedWork W2943796244 @default.
- W3166300576 hasRelatedWork W3157726097 @default.
- W3166300576 hasRelatedWork W3166300576 @default.
- W3166300576 hasRelatedWork W4283820102 @default.
- W3166300576 hasRelatedWork W4322763382 @default.
- W3166300576 hasVolume "29" @default.
- W3166300576 isParatext "false" @default.
- W3166300576 isRetracted "false" @default.
- W3166300576 magId "3166300576" @default.
- W3166300576 workType "article" @default.