Optimization of Age of Information for Energy Harvesting Two-hop Networks Based on Partial Relay Selection

‍ ‍To solve the problem of high information transmission delay caused by energy limitation of sensor nodes in traditional relaying cooperative networks， a two-hop relaying wireless sensor network model is constructed by combining short packet communication （SPC） and energy harvesting. In this network， both the source and the relay need to collect RF signal energy from the dedicated energy base station to ensure the continuous transmission of information. The relay uses half duplex decoding and forwarding protocol to transmit the status update information sent by the source node to the destination in the form of short packets. Using relay selection strategy and maximum ratio combining technology， the earliest partial relay selection scheme based on timeliness is proposed. Considering the freshness of the network information， firstly， the SPC transmission performance is analyzed， and the average packet error probability at the relay and destination is derived. Secondly， the order statistics are utilized to describe the number of packet retransmission attempts of the packets from the source to the relay， the expression of the average age of information （AoI） of the network is derived based on the update process theory， and the gradient descent method is used to optimize some parameters affecting the average AoI of the network. Finally， the simulation results show that optimizing the number of relays and the length of packets can effectively improve the freshness of network information. Meanwhile， comparing the optimal value obtained by optimizing some parameters with the theoretical value of minimum average AoI obtained by greedy method， which can verify the feasibility of partial relay cooperation scheme in two-hop state update network.