In this paper,a sparse graph neural network-aided(SGNN-aided)decoder is proposed for improving the decoding performance of polar codes under bursty interference.Firstly,a sparse factor graph is constructed using the e...In this paper,a sparse graph neural network-aided(SGNN-aided)decoder is proposed for improving the decoding performance of polar codes under bursty interference.Firstly,a sparse factor graph is constructed using the encoding characteristic to achieve high-throughput polar decoding.To further improve the decoding performance,a residual gated bipartite graph neural network is designed for updating embedding vectors of heterogeneous nodes based on a bidirectional message passing neural network.This framework exploits gated recurrent units and residual blocks to address the gradient disappearance in deep graph recurrent neural networks.Finally,predictions are generated by feeding the embedding vectors into a readout module.Simulation results show that the proposed decoder is more robust than the existing ones in the presence of bursty interference and exhibits high universality.展开更多
With the objective of establishing the necessary conditions for 3-D seismic data from a Permian plutonic oilfield in western China, we compared the technology of several multi-parameter seismic inversion methods in id...With the objective of establishing the necessary conditions for 3-D seismic data from a Permian plutonic oilfield in western China, we compared the technology of several multi-parameter seismic inversion methods in identifying igneous rocks. The most often used inversion methods are Constrained Sparse Spike Inversion (CSSI), Artificial Neural Network Inversion (ANN) and GR Pseudo-impedance Inversion. Through the application of a variety of inversion methods with log curves correction, we obtained relatively high-resolution impedance and velocity sections, effectively identifying the lithology of Permian igneous rocks and inferred lateral variation in the lithology of igneous rocks. By means of a comprehensive comparative study, we arrived at the following conclusions: the CSSI inversion has good waveform continuity, and the ANN inversion has lower resolution than the CSSI inversion. The inversion results show that multi-parameter seismic inversion methods are an effective solution to the identification of igneous rocks.展开更多
文摘In this paper,a sparse graph neural network-aided(SGNN-aided)decoder is proposed for improving the decoding performance of polar codes under bursty interference.Firstly,a sparse factor graph is constructed using the encoding characteristic to achieve high-throughput polar decoding.To further improve the decoding performance,a residual gated bipartite graph neural network is designed for updating embedding vectors of heterogeneous nodes based on a bidirectional message passing neural network.This framework exploits gated recurrent units and residual blocks to address the gradient disappearance in deep graph recurrent neural networks.Finally,predictions are generated by feeding the embedding vectors into a readout module.Simulation results show that the proposed decoder is more robust than the existing ones in the presence of bursty interference and exhibits high universality.
文摘With the objective of establishing the necessary conditions for 3-D seismic data from a Permian plutonic oilfield in western China, we compared the technology of several multi-parameter seismic inversion methods in identifying igneous rocks. The most often used inversion methods are Constrained Sparse Spike Inversion (CSSI), Artificial Neural Network Inversion (ANN) and GR Pseudo-impedance Inversion. Through the application of a variety of inversion methods with log curves correction, we obtained relatively high-resolution impedance and velocity sections, effectively identifying the lithology of Permian igneous rocks and inferred lateral variation in the lithology of igneous rocks. By means of a comprehensive comparative study, we arrived at the following conclusions: the CSSI inversion has good waveform continuity, and the ANN inversion has lower resolution than the CSSI inversion. The inversion results show that multi-parameter seismic inversion methods are an effective solution to the identification of igneous rocks.
