Moussa, N.[14] presents a novel energy-efficient and reliable routing protocol for wildfire monitoring in bandwidth-constrained wireless sensor networks (WSNs). The latter prioritizes energy efficiency and network reliability, it outperforms existing protocols with a 30.55% increase in network lifetime and 14.71% faster response times for critical applications such as forest fire detection.
[15] Seyed Reza Nabavi: focuses on improving the lifetime of wireless sensor networks (WSNs), composed of sensor nodes monitoring various applications, introducing a particle swarm optimization (PSO) method. for clustering and proposing an energy-efficient routing protocol that takes into account factors such as node density, energy and communication quality to efficiently transmit data to the base station. The protocol uses parallel computation of fitness functions for faster convergence to the best solution and uses PSO-based clustering inspired by flocking behavior. By combining the concept of fuzzy logic with PSO, it can improve network performance in terms of packet delivery, energy efficiency, throughput, load balancing and network lifetime.
Mohammed Zaid Ghawy [16] addresses the emerging challenge posed by the rapid proliferation of smartphones and Internet of Things (IoT) devices in wireless sensor networks (WSNs). They can impact quality of service (QoS) requirements, particularly end-to-end latency, energy consumption, and packet loss during transmission. To improve WSN performance and meet QoS metrics, the paper presents a multi-path routing protocol based on particle swarm optimization (PSO), designed for IoT applications with heavy traffic loads and flow imbalances. network. Using the NS-2 simulator, experiments are conducted to evaluate the protocol against the AODV and DSDV routing protocols. The results demonstrate the benefits of the proposed approach, including energy savings, low end-to-end delay, high packet delivery rate, increased throughput, and reduced normalization overhead.
R. Samadi: [17] This paper presents the energy-aware intelligent routing protocol in mobile IoT networks based on SDN, a novel routing approach designed to optimize network lifetime and minimize energy dissipation in case of dynamic topology changes. result of mobile nodes. This protocol uses clustering and an intelligent scalable algorithm to determine the necessary number of clusters and their distribution within the dynamic network environment. Additionally, the approach includes mechanisms to reduce packet control and routing overhead. They can have a significant impact on the energy consumption of nodes. The simulation results highlight the effectiveness of this protocol compared to other approaches in terms of packet delivery rate, average energy consumption, network lifetime, number of active nodes, coverage and routing overhead.
Roberts, MK[18] This research presents an advanced cluster-based secure routing protocol designed for high performance, reliability, and energy efficiency. The protocol addresses various critical aspects including energy optimization, packet management, congestion control, encrypted data transmission, and monitoring against attacking nodes to improve the quality of data management. It addresses network isolation and segmentation issues that can prevent wireless sensor nodes from communicating with the base station or lead to node failures. Several metrics, such as ransomware attack detection rate, ergodic residual energy, clone attack detection, throughput maximization, delay, capacity optimization, and network lifetime, are used to demonstrate the effectiveness of the proposed technique.
A. K. Rao[19] Wireless sensor networks (WSNs), initially developed for research purposes, have found a wide range of applications in various fields including communication, weather forecasting, agriculture, industries, intelligent monitoring and monitoring. In the agricultural sector, IoT-enabled WSNs are essential for monitoring environmental and climate conditions using a wide range of sensors. Used in agricultural environments, these sensors contribut from smart agriculture, decision-making and the collection of vital data on factors such as temperature, humidity and irrigation patterns, with the aim of improving production yields. Since agriculture employs almost 70% of India's population and contributes 27% to GDP, it is a critical area for research and development. This study revolves around investigating IoT-based methods to achieve better agricultural results.
Gupta, A[20]This paper presents a novel multi-layer clustering and routing model aimed at achieving energy-efficient and scalable long-distance data communication in wireless sensor network-assisted Internet of Things (IoT) systems ( WSN), particularly for smart agriculture. It starts with fuzzy clustering of k-medoids to optimize energy consumption by forming clusters. Next, an improved sparrow search algorithm, which combines SSA and chameleon swarm algorithm are proposed for optimal cluster head selection to address the challenges of energy holes in WSNs. A multi-layer approach is used to improve data transmission, taking into account the parameters of sensor nodes of the physical layer, network layer and media access control (MAC) for routing. Finally, a bio-inspired algorithm, the sandpiper optimization algorithm, in combination with cosine similarity, is used to determine the optimal route for efficient data transmission and retransmission. The proposed protocol is evaluated by simulation using the Network Simulator (NS2), with performance metrics including end-to-end delay, packet delivery rate, communication overhead, communication cost, average energy consumption and network lifetime.
S. Bhatia[21] This article focuses on intelligent deployment of sensor nodes for IoT-WSN in smart agriculture using analytical algorithms such as genetic algorithm (GA) and swarm optimization of particles (PSO). The objective of this method is to minimize the energy exhaustion of sensor nodes and extend the network lifetime by using direct and multi-hop routing protocols. Experimental results demonstrate an average increase in network lifetime of 140–150%. The paper also includes a comparison between GA and PSO approaches. In addition to this, it explores the advantages of heterogeneous networks through experimental results.
Although this work offers valuable information or an optimal answer in a particular scenario, it is important to note that their applicability to different agricultural contexts may vary and that they may present limitations in terms of scalability, adaptability and feasibility in all agricultural scenarios. Implementing and managing these results in practical, real-world settings can also be complex.
Research in the area of IoT (Internet of Things) and WSNs (wireless sensor networks) for agriculture is promising. However, to fully realize this potential, it requires further real-world validation, increased adaptability to various agricultural applications, and balanced consideration of performance parameters such as energy efficiency, reliability, and safety. Future research efforts should prioritize addressing these limitations to advance the field.
Our approach
Our routing algorithm is intended for agricultural IoT networks. It is based on the MQTT protocol and relies on several key mechanisms to optimize communication between sensors, actuators and brokers. This section presents the proposed algorithm in detail, highlighting the steps and strategies used.
Step 1: Advertisment from brokers to the other brokers
-
Network brokers send an advertising message to the network to indicate that they are brokers.
-
The brokers in the network handle this message and add it to their local list of brokers.
-
At the end of this step, the brokers order their list geographically according to the ordinate, then the abscissa, Therefore all the lists in the network have the same order.
Step 2: Round-robin for brokers
-
Since the network brokers are organized in a list, in each round, the active broker is selected according to the round-robin strategy and becomes the central broker.
-
The other brokers become bridge brokers during this round.
Step 3: Advertisment from the central broker to the other brokers
-
The central broker sends an advertisment to the other brokers to establish communication.
-
If the central broker does not send an advertizing message during a specific period, the next broker will be selected in his place.
-
bridge brokers connect and subscribe to the central broker in a common topic called CB (Central Broker).
Step 4: Advertisment from the bridge brokers to the IoT nodes
Step 5: Selection of the closest bridge broker by the IoT nodes
-
Each IoT node selects the closest bridge broker based on its geographic position.
-
This selection is based on criteria such as distance, signal quality and availability of the bridge broker.
-
The IoT nodes connect to their respective brokers, and the actuator nodes subscribe to their respective topics.
Step 6 : Communication via the MQTT protocol
-
In this step, each sensor node publish on his bridge broker in a specific topic.
-
The bridge broker aggregates the payloads of its sensors connected to it and publishes the aggregated message in the CB topic of the central broker.
-
Then, the central broker forwards the message to all bridge brokers.
-
The bridge brokers disaggregate the message and forward to their subscriber actuators the payloads that concern their topic.
-
Finally, the actuators take action according to the received message.
-
The sensors only publish if their value changes in order to save their energy as much as possible.
We can see the protocol diagram in the Fig. 1 below, the sensors and actuator nodes haven't been represented to sumerize and facilitate the representation of protocol.
The proposed algorithm offers several advantages for agricultural IoT networks. The load distribution is balanced between brokers thanks to the round-robin strategy, thus avoiding overloads on certain nodes. The communication between IoT nodes and brokers is facilitated by bridge brokers, reducing long-distance transmissions and energy consumption. Additionally, the selection of the closest bridge broker by IoT nodes ensures efficient communication with reduced transmission delays.