QoS and energy aware cooperative routing protocol for wildfire monitoring wireless sensor networks

Abstract

Wireless sensor networks (WSN) are presented as proper solution for wildfire monitoring. However, this application requires a design of WSN taking into account the network lifetime and the shadowing effect generated by the trees in the forest environment. Cooperative communication is a promising solution for WSN which uses, at each hop, the resources of multiple nodes to transmit its data. Thus, by sharing resources between nodes, the transmission quality is enhanced. In this paper, we use the technique of reinforcement learning by opponent modeling, optimizing a cooperative communication protocol based on RSSI and node energy consumption in a competitive context (RSSI/energy-CC), that is, an energy and quality-of-service aware-based cooperative communication routing protocol. Simulation results show that the proposed algorithm performs well in terms of network lifetime, packet delay, and energy consumption.

Figure 1

Reinforcement learning model.

Figure 2

Multihop mesh cooperative structure for data dissemination in WSNs.

Figure 3

Cooperation between adjacent groups of cooperative nodes.

Figure 4

Sink node (in black) placement for topologies (A), (B), and (C).

Figure 5

Average delay to the sink, percentage of delayed packets, and percentage of lost packets by averaging on the number of nodes being away with the same number of hops from the sink node.

Figure 6

Selected CN groups for data transmission from source node 4 (in red) to sink node 76 (in black).

Figure 7

Energy consumption comparison for each selected CN group between MRL-CC algorithm and RSSI/energy-CC algorithm.

Figure 8

Network energy consumption, comparison between network architectures for MRL-CC and E/RSSI CC algorithm.

Figure 9

Maximal energy consumption in the whole WSN, comparison between MRL-CC and E/RSSI-CC algorithms for different network architectures.

Figure 10

Network energy consumption, comparison between network architectures for MRL-CC and E/RSSI CC algorithms.

Figure 11

Maximal energy consumption in the whole WSN, comparison between MRL-CC and E/RSSI CC algorithms for different network architectures.

Figure 12

WSN topology in circles; sink node is at the center.

Figure 13

Network energy consumption and maximal energy consumption for network in form of circles, for MRL-CC algorithm and RSSI/energy CC algorithm.

Figure 14

Network lifetime (architecture C) for different number of path losses, n, and different shadowing standard deviation, for the MRL-CC and the RSSI/energy algorithms.