Research on secure Internet of things gateway technology based on multi-communication methods

Abstract The Internet of Things, as an important part data aggregation, forwarding and control, often leads to objectivity errors due the huge complex received data. Based on this, this paper introduces GRU, LSTM, SRU deep learning optimize by selects most suitable communication mode optimization algorithm. experimental results show that accuracy algorithms a downward trend, from 0.024 0.010%; training time is reduced 254 minutes, speed increased 86%, indicating excellent performance in IoT gateways.