An Improved Time Feedforward Connections Recurrent Neural Networks

Recurrent Neural Networks(RNNs)have been widely applied to deal with temporal problems,such as flood forecasting and financial data processing.On the one hand,traditional RNNs models amplify the gradient issue due to the strict time serial dependency,making it difficult to realize a long-term memory function.On the other hand,RNNs cells are highly complex,which will signifi-cantly increase computational complexity and cause waste of computational resources during model training.In this paper,an improved Time Feedforward Connections Recurrent Neural Networks(TFC-RNNs)model was first proposed to address the gradient issue.A parallel branch was introduced for the hidden state at time t−2 to be directly transferred to time t without the nonlinear transforma-tion at time t−1.This is effective in improving the long-term dependence of RNNs.Then,a novel cell structure named Single Gate Recurrent Unit(SGRU)was presented.This cell structure can reduce the number of parameters for RNNs cell,consequently reducing the computational complexity.Next,applying SGRU to TFC-RNNs as a new TFC-SGRU model solves the above two difficulties.Finally,the performance of our proposed TFC-SGRU was verified through sev-eral experiments in terms of long-term memory and anti-interference capabilities.Experimental results demonstrated that our proposed TFC-SGRU model can cap-ture helpful information with time step 1500 and effectively filter out the noise.The TFC-SGRU model accuracy is better than the LSTM and GRU models regarding language processing ability.

Key Dolerite Dyke Swarms of Amazonia:U-Pb Constraints on Supercontinent Cycles and Geodynamic Connections with Global LIP Events Through Time

High-resolution U–Pb(ID-TIMS,baddeleyite)ages are presented for mafic dykes from selected swarms in two important Amazonian regions:the Carajás Province in the east,and the Rio Apa block in the southwest–areas

Research and Optimization of GPSR Routing Protocol for Vehicular Ad-Hoc Network

Security is one of the most critical issues to Vehicular Ad-hoc Networks (VANETs) since the information transmitted is asynchronous and distributed. Vulnerability and instability are two of the challenges remain to be addressed by the research community and the industry. In this paper, we first proposed a trust reliability based model and extended the GPSR protocol to TM-GPSR protocol. Then, we improved the LET-GPSR protocol based on the link connection time prediction. On this basis, combined the decision index of the TM-GPSR and LET-GPSR protocols, we proposed the RC-GPSR routing protocol. We built the standard testing platform on the NS2 and SUMO, the average end-to-end delay and packet delivery rate of GPSR protocol and the three updates protocols under different node density, node speed, and malicious node ratio are simulated and evaluated. The results showed that under the same conditions, compared with GPSR protocol, RC-GPSR protocol has a lower average end-to-end delay and a higher packet delivery rate, which effectively improves the link stability and security.

Intermittent Geocast Routing in Urban Vehicular Delay Tolerant Networks

Nowadays, both vehicular active safety service and user infotainment service have become two core applications for urban Vehicular Delay Tolerant Networks（u VDTNs）. Both core applications require a high data transmission capacity over u VDTNs. In addition, the connection between any two vehicles in u VDTNs is intermittent and opportunistic. Intermittent data dissemination over u VDTNs is a stringent and challenging issue. In this paper,we propose Intermittent Geocast Routing（IGR）. For the first step, IGR has to estimate the active connection time interval via the moving directions and velocities between any two vehicles. Second, the throughput function for u VDTNs is fitted by building a wavelet neural network traffic model. Third, the throughput function within the effective connection time interval is integrated to obtain the forwarding capability estimation of the node. Fourth, a high-efficiency geocast routing algorithm using the node forwarding capability for u VDTNs is designed. Finally, IGR is simulated on the opportunistic Network Environment simulator. Experimental results show that IGR can greatly improve the packet delivery ratio, transmission delay, delay jitter, and packet loss rate compared with the state of the art.