Function simulation,which is called virtual reality too,is popularly applied to solve uncertain problems.Good performance of hidden layers and perfect capability of function simulation make artificial neural networks ...Function simulation,which is called virtual reality too,is popularly applied to solve uncertain problems.Good performance of hidden layers and perfect capability of function simulation make artificial neural networks one of the best choices to simulate functions with form unknown.Inputs and outputs were used to train the structure of the ar- tificial neural network to make the outputs of network vary with the given inputs and keep consistent with the original data within tolerance.However,we couldn't get expected re- sults by using samples of a simple two-variable-model for the cause of dimensional differ- ence.The way of artificial neural networks to fit functions,which uses 'multi-dimensional surface' of high dimension to fit 'multi-dimensional line' of low dimension,was proved;the conclusion that good effects of fitting don't mean good function modeling when a dimen- sional difference exists was provided,and a suggestion of 'surface collecting' in practical engineering application was proposed when collecting useful data.展开更多
An optical transceiver with a novel optical subassembly structure is proposed in this paper, which achieves high coupling efficiency and low assembly difficulty. The proposed optical transceiver consumes 0.9 W power a...An optical transceiver with a novel optical subassembly structure is proposed in this paper, which achieves high coupling efficiency and low assembly difficulty. The proposed optical transceiver consumes 0.9 W power and retains a small size of 28 mmx16 mmx3 mm. The fabrication process of the silicon substrate and the assembly process of the optical transceiver are demonstrated in details. Moreover, the optical transceiver is measured in order to verify its transmission performance. The clear eye diagrams and the low bit error rate （BER） less than 1013 at 10 Gbit/s per channel show good transmission characteristics of the designed optical transceiver.展开更多
文摘Function simulation,which is called virtual reality too,is popularly applied to solve uncertain problems.Good performance of hidden layers and perfect capability of function simulation make artificial neural networks one of the best choices to simulate functions with form unknown.Inputs and outputs were used to train the structure of the ar- tificial neural network to make the outputs of network vary with the given inputs and keep consistent with the original data within tolerance.However,we couldn't get expected re- sults by using samples of a simple two-variable-model for the cause of dimensional differ- ence.The way of artificial neural networks to fit functions,which uses 'multi-dimensional surface' of high dimension to fit 'multi-dimensional line' of low dimension,was proved;the conclusion that good effects of fitting don't mean good function modeling when a dimen- sional difference exists was provided,and a suggestion of 'surface collecting' in practical engineering application was proposed when collecting useful data.
基金supported by the National High Technology Research and Development Program of China(No.2015AA016904)
文摘An optical transceiver with a novel optical subassembly structure is proposed in this paper, which achieves high coupling efficiency and low assembly difficulty. The proposed optical transceiver consumes 0.9 W power and retains a small size of 28 mmx16 mmx3 mm. The fabrication process of the silicon substrate and the assembly process of the optical transceiver are demonstrated in details. Moreover, the optical transceiver is measured in order to verify its transmission performance. The clear eye diagrams and the low bit error rate （BER） less than 1013 at 10 Gbit/s per channel show good transmission characteristics of the designed optical transceiver.
