Abstract:

In special factories where explosive gases or liquids exist,automated guided vehicles need to be networked in the form of mobile ad hoc networks.When there are obstacles in the factory to block the wireless signal,the vehicle movement will cause the channel conditions between the vehicles to change between line-of-sight and non-line-of-sight,and the wireless transmission range of the vehicle will change accordingly so that the packet loss or delay of the existing data routing methods based on fixed transmission range will increase.Oriented to mobile ad hoc networks in such environments,this paper proposes a low-latency data routing method based on the line-of-sight map.First,a method based on fully-connected neural network is proposed to construct a line-of-sight map,which generates raw data by estimating the signal-to-noise ratio of the routing messages,converts the raw data into training data based on the signal-to-noise ratio magnitude,and trains the fully-connected neural network to obtain the line-of-sight map.Then,a delay-aware data routing protocol is proposed based on the line-of-sight map,which can estimate the expected number of transmissions of wireless links between nodes based on the line-of-sight map,and calculate and update the route evaluation score based on the number of transmissions,with nodes selecting the node with a larger route evaluation score (corresponding to smaller delay) as the next-hop forwarding node.Finally,simulation results show that the proposed method in this paper is able to reduce the communication delay by 66% without decreasing the packet delivery rate compared to the data routing method based on a fixed transmission range.

Key words: mobile ad hoc network, routing protocol, automated guided vehicle, line of sight