Optical Axons for Electro-Optical Neural Networks
Recently, neuromorphic sensors, which convert analogue signals to spiking frequencies, have been reported for neurorobotics. In bio-inspired systems these sensors are connected to the main neural unit to perform post-processing of the sensor data. The performance of spiking neural networks has been...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-10-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/21/6119 |
| _version_ | 1797549644752879616 |
|---|---|
| author | Mircea Hulea Zabih Ghassemlooy Sujan Rajbhandari Othman Isam Younus Alexandru Barleanu |
| author_facet | Mircea Hulea Zabih Ghassemlooy Sujan Rajbhandari Othman Isam Younus Alexandru Barleanu |
| author_sort | Mircea Hulea |
| collection | DOAJ |
| description | Recently, neuromorphic sensors, which convert analogue signals to spiking frequencies, have been reported for neurorobotics. In bio-inspired systems these sensors are connected to the main neural unit to perform post-processing of the sensor data. The performance of spiking neural networks has been improved using optical synapses, which offer parallel communications between the distanced neural areas but are sensitive to the intensity variations of the optical signal. For systems with several neuromorphic sensors, which are connected optically to the main unit, the use of optical synapses is not an advantage. To address this, in this paper we propose and experimentally verify optical axons with synapses activated optically using digital signals. The synaptic weights are encoded by the energy of the stimuli, which are then optically transmitted independently. We show that the optical intensity fluctuations and link’s misalignment result in delay in activation of the synapses. For the proposed optical axon, we have demonstrated line of sight transmission over a maximum link length of 190 cm with a delay of 8 μs. Furthermore, we show the axon delay as a function of the illuminance using a fitted model for which the root mean square error (RMS) similarity is 0.95. |
| first_indexed | 2024-03-10T15:17:26Z |
| format | Article |
| id | doaj.art-026b903c18db49e7809f87a8bb730cd5 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T15:17:26Z |
| publishDate | 2020-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-026b903c18db49e7809f87a8bb730cd52023-11-20T18:46:40ZengMDPI AGSensors1424-82202020-10-012021611910.3390/s20216119Optical Axons for Electro-Optical Neural NetworksMircea Hulea0Zabih Ghassemlooy1Sujan Rajbhandari2Othman Isam Younus3Alexandru Barleanu4Faculty of Automatic Control and Computer Engineering at Gheorghe Asachi Technical University of Iasi, 700050 Iasi, RomaniaOptical Communications Research Group, Faculty of Engineering and Environment at Northumbria University, Newcastle upon Tyne NE7 7XA, UKHuawei Technologies Sweden AB, 412 50 Gothenburg, SwedenOptical Communications Research Group, Faculty of Engineering and Environment at Northumbria University, Newcastle upon Tyne NE7 7XA, UKFaculty of Automatic Control and Computer Engineering at Gheorghe Asachi Technical University of Iasi, 700050 Iasi, RomaniaRecently, neuromorphic sensors, which convert analogue signals to spiking frequencies, have been reported for neurorobotics. In bio-inspired systems these sensors are connected to the main neural unit to perform post-processing of the sensor data. The performance of spiking neural networks has been improved using optical synapses, which offer parallel communications between the distanced neural areas but are sensitive to the intensity variations of the optical signal. For systems with several neuromorphic sensors, which are connected optically to the main unit, the use of optical synapses is not an advantage. To address this, in this paper we propose and experimentally verify optical axons with synapses activated optically using digital signals. The synaptic weights are encoded by the energy of the stimuli, which are then optically transmitted independently. We show that the optical intensity fluctuations and link’s misalignment result in delay in activation of the synapses. For the proposed optical axon, we have demonstrated line of sight transmission over a maximum link length of 190 cm with a delay of 8 μs. Furthermore, we show the axon delay as a function of the illuminance using a fitted model for which the root mean square error (RMS) similarity is 0.95.https://www.mdpi.com/1424-8220/20/21/6119optical neural networksoptical axonsoptical signal fadingVLC |
| spellingShingle | Mircea Hulea Zabih Ghassemlooy Sujan Rajbhandari Othman Isam Younus Alexandru Barleanu Optical Axons for Electro-Optical Neural Networks Sensors optical neural networks optical axons optical signal fading VLC |
| title | Optical Axons for Electro-Optical Neural Networks |
| title_full | Optical Axons for Electro-Optical Neural Networks |
| title_fullStr | Optical Axons for Electro-Optical Neural Networks |
| title_full_unstemmed | Optical Axons for Electro-Optical Neural Networks |
| title_short | Optical Axons for Electro-Optical Neural Networks |
| title_sort | optical axons for electro optical neural networks |
| topic | optical neural networks optical axons optical signal fading VLC |
| url | https://www.mdpi.com/1424-8220/20/21/6119 |
| work_keys_str_mv | AT mirceahulea opticalaxonsforelectroopticalneuralnetworks AT zabihghassemlooy opticalaxonsforelectroopticalneuralnetworks AT sujanrajbhandari opticalaxonsforelectroopticalneuralnetworks AT othmanisamyounus opticalaxonsforelectroopticalneuralnetworks AT alexandrubarleanu opticalaxonsforelectroopticalneuralnetworks |