Multistage fuzzy logic congestion-aware routing using dual-stage notification and the relative barring distance in wireless sensor networks

Congestion management in a wireless sensor network (WSN) is a key determinant of the quality of service. Congestion in a network causes data loss, a reduced transmission rate, increased delays, and excess energy consumption. The latter has a direct impact on tiny sensor devices with limited resources and processing, buffering, and transmitting capabilities. In addition, a WSN relies on multiple packet relays between nodes, which inevitably results in network congestion near the base station, whose neighboring nodes incur crowded traffic from multisource deliveries. Thus, this paper proposes a novel routing method that minimizes congestion. The adaptive routing strategy consists of 3 main modules. First, an optimal notification level for queue control is specified by using multistage fuzzy logic (MFL). The resulting weights evaluated from congestion-related parameters are then passed onto the subsequent modules. The second module adjusts the congestion notification, which makes the module more flexible to improve its routing discovery efficiency and to reduce the chance of loss during the rerouting stage. Finally, we propose a routing adjustment and control mechanism by using a novel navigation technique based on linear and angular distances and MFL to create weights for path assessment. Simulation results demonstrate that the proposed method outperforms the state-of-the-art methods in terms of the packet loss ratio, average hop count, network lifetime, and energy consumption metrics.