针对矿井涌水量预测中存在的深度学习模型预测精度不高和适用性不强的问题,提出了一种基于深度残差网络(Deep Residual Network,DRN)和双向长短记忆网络(Bidirectional short and long memory network,BiLSTM)的矿井涌水量预测方法。首...针对矿井涌水量预测中存在的深度学习模型预测精度不高和适用性不强的问题,提出了一种基于深度残差网络(Deep Residual Network,DRN)和双向长短记忆网络(Bidirectional short and long memory network,BiLSTM)的矿井涌水量预测方法。首先,将矿井涌水量数据进行小波分解和归一化处理,得到趋势项数据和细节项数据;其次,采用DRN网络方法对趋势项数据进行预测,采用BiLSTM网络方法对细节项数据进行预测;最后,将2部分预测结果进行重构得到矿井涌水量预测结果。研究结果表明:DRN-BiLSTM模型相比于单一模型预测精度更高,说明该模型具有更好的泛化性。展开更多
车载网(vehicular Ad hoc networks,VANETs)是由车辆和路由设施(road side units,RSU)构成的网络,在VANETs中需要有效的路由协议实现源节点的数据向目的节点传输。在可用协议中,直接路由节点选择方案DRNS无法辨别在路由过程中节点移动...车载网(vehicular Ad hoc networks,VANETs)是由车辆和路由设施(road side units,RSU)构成的网络,在VANETs中需要有效的路由协议实现源节点的数据向目的节点传输。在可用协议中,直接路由节点选择方案DRNS无法辨别在路由过程中节点移动的方向,而按需多路径距离矢量路由(Ad hoc on-demand multipath distance vector,AOMDV)因节点快速移动,路由不稳定。由于不能检测节点在其他方向上移动,DRNS仅适合节点移动方向单一的高速场景(highway scenarios)。针对城市环境,提出敏捷方向节点选择多路径路由(acute direction route node selection multipath,ADRNSM)。ADRNSM属于多路径路由,能适用16个不同的移动方向;ADRNSM敏捷地检测节点的移动方向,并促使节点仅在目的节点方向上移动,同时提供多条路径向目的节点传输数据。仿真结果表明,与AOMDV相比,提出的ADRNSM提高了分组传递率、吞吐量、端到端传输时延以及归一化路由开销的性能。展开更多
Higher-order statistics based approaches and signal sparseness based approaches have emerged in recent decades to resolve the underdetermined direction-of-arrival(DOA)estimation problem.These model-based methods face ...Higher-order statistics based approaches and signal sparseness based approaches have emerged in recent decades to resolve the underdetermined direction-of-arrival(DOA)estimation problem.These model-based methods face great challenges in practical applications due to high computational complexity and dependence on ideal assumptions.This paper presents an effective DOA estimation approach based on a deep residual network(DRN)for the underdetermined case.We first extract an input feature from a new matrix calculated by stacking several covariance matrices corresponding to different time delays.We then provide the input feature to the trained DRN to construct the super resolution spectrum.The DRN learns the mapping relationship between the input feature and the spatial spectrum by training.The proposed approach is superior to existing model-based estimation methods in terms of calculation efficiency,independence of source sparseness and adaptive capacity to non-ideal conditions(e.g.,low signal to noise ratio,short bit sequence).Simulations demonstrate the validity and strong performance of the proposed algorithm on both overdetermined and underdetermined cases.展开更多
文摘针对矿井涌水量预测中存在的深度学习模型预测精度不高和适用性不强的问题,提出了一种基于深度残差网络(Deep Residual Network,DRN)和双向长短记忆网络(Bidirectional short and long memory network,BiLSTM)的矿井涌水量预测方法。首先,将矿井涌水量数据进行小波分解和归一化处理,得到趋势项数据和细节项数据;其次,采用DRN网络方法对趋势项数据进行预测,采用BiLSTM网络方法对细节项数据进行预测;最后,将2部分预测结果进行重构得到矿井涌水量预测结果。研究结果表明:DRN-BiLSTM模型相比于单一模型预测精度更高,说明该模型具有更好的泛化性。
文摘车载网(vehicular Ad hoc networks,VANETs)是由车辆和路由设施(road side units,RSU)构成的网络,在VANETs中需要有效的路由协议实现源节点的数据向目的节点传输。在可用协议中,直接路由节点选择方案DRNS无法辨别在路由过程中节点移动的方向,而按需多路径距离矢量路由(Ad hoc on-demand multipath distance vector,AOMDV)因节点快速移动,路由不稳定。由于不能检测节点在其他方向上移动,DRNS仅适合节点移动方向单一的高速场景(highway scenarios)。针对城市环境,提出敏捷方向节点选择多路径路由(acute direction route node selection multipath,ADRNSM)。ADRNSM属于多路径路由,能适用16个不同的移动方向;ADRNSM敏捷地检测节点的移动方向,并促使节点仅在目的节点方向上移动,同时提供多条路径向目的节点传输数据。仿真结果表明,与AOMDV相比,提出的ADRNSM提高了分组传递率、吞吐量、端到端传输时延以及归一化路由开销的性能。
基金supported by the Program for Innovative Research Groups of the Hunan Provincial Natural Science Foundation of China(2019JJ10004)。
文摘Higher-order statistics based approaches and signal sparseness based approaches have emerged in recent decades to resolve the underdetermined direction-of-arrival(DOA)estimation problem.These model-based methods face great challenges in practical applications due to high computational complexity and dependence on ideal assumptions.This paper presents an effective DOA estimation approach based on a deep residual network(DRN)for the underdetermined case.We first extract an input feature from a new matrix calculated by stacking several covariance matrices corresponding to different time delays.We then provide the input feature to the trained DRN to construct the super resolution spectrum.The DRN learns the mapping relationship between the input feature and the spatial spectrum by training.The proposed approach is superior to existing model-based estimation methods in terms of calculation efficiency,independence of source sparseness and adaptive capacity to non-ideal conditions(e.g.,low signal to noise ratio,short bit sequence).Simulations demonstrate the validity and strong performance of the proposed algorithm on both overdetermined and underdetermined cases.
