DROO-C Data Transfer in a Mobile Edge Computing Offloading Framework.

To solve the problem of transmitting information in wireless-driven Vehicular Networking, Mobile Edge Computing (MEC) becomes an effective method to enhance the data transmission capability in fast fading channel conditions under multi-threaded interactions of different nodes in Vehicular Networking. In order to solve the efficiency problem of data transmission in Vehicular Networking under strong time-varying conditions, this paper proposes a deep neural network-based DROO - C computational offloading framework, which divides the MEC environment with complex network topology problems into two sub-problems of task offloading decision making and task data updating by the Mixed-Integer-Planning (MIP) method, where the deep neural network serves as a continuous iterative decision making scheme to optimize the data transmission in the constant The deep neural network is used as a continuous iterative decision-making scheme to optimize the parameters of data transmission in the continuous update training, and the final result is obtained by weighting and calculating the rate, and finally the lossless compression of entropy reduction transformation compresses the transmission data optimized by the DROO - C framework, which maximally improves the data transmission rate between nodes and reduces the transmission time.