لخّصلي

3 | ROUTING PROTOCOLS
3.1 | Classification criteria for routing protocols in MWSNs
In the next sections, we classify the routing protocols for MWSNs into three categories, which are routing protocols for MSN, routing protocols for MS, and routing protocols for the mobility of both sensor and sink nodes together

3.2 | Routing protocols for MSN in MWSNs
3.2.1 | Low Energy Adaptive Clustering Hierarchy Mobile protocol
Low Energy Adaptive Clustering Hierarchy Mobile protocol (LEACH-M)44 is the first proposed protocol that deals with mobile nodes in WSN.

CH node makes a router between the nodes and the sink, gathers and fuses data from nodes on its cluster, and transmits this emerged information to the sink.

The CH that hears this message sends a cluster join ACK message to this mobile node confirming that they are connected.

3.2.4 | Mobility Based Clustering Algorithm

Mobility Based Clustering Algorithm (MBC)47 selects the CHs randomly depending on the mobility and residual energy of nodes every round.

3.2.5 | Enhanced Cluster Based Routing Algorithm Developed for Mobile WSNs
An adaptation of the CBR-Mobile algorithm is Enhanced Cluster Based Routing Algorithm Developed for Mobile WSNs (ECBR-MWSN).48 It contains five phases, initialization, the formation of the cluster, and selection of CHs, the transmis- sion of data, rerouting, and clustering.

3.2.6 | Mobility Adaptive Cross-layer Routing Algorithm The basic idea of the Mobility Adaptive Cross-layer Routing Algorithm (MACRO) algorithm49 depends on a novel cross-layer interaction technique among the physical, Medium Access Control, network, transport, and application layers

3.2.8 | Velocity Energy-Efficient and Link-aware Cluster Tree

A modification of the CIDT algorithm, the Velocity Energy-Efficient and Link-aware Cluster Tree (VELCT) algorithm, was proposed in Velmani and Kaarthick51 to gather the information from MWSN.

VELCT considers the locations of CHs in building the DCT to alleviate the problems in CIDT like delay in tree construction, coverage, and mobility.

3.2.9 | Proactive Highly Ambulatory Sensor Routing Protocol
Proactive Highly Ambulatory Sensor Routing Protocol (PHASeR),52 a multi-hop routing protocol, enables robust and dynamic data routing to the sink.

PHASeR was encouraged with the application of radiation mapping that needs peri- odic and reliable transmission of information to the sink.

To maintain a gradient metric in low and high-speed scenar- ios, a Global-TDMA (GTDMA) MAC layer is used, which allocates fixed timeslots for each sensor.

3.2.12 | Cross-layer Energy Efficiency Protocol An energy-effective cross-layer transmission pattern for MWSN called Cross-layer Energy Efficiency Protocol (CEE) was proposed in Yang et al.55 Three network layers are merged in the transmission model: physical layer, data link layer, and network layer.

In the CEE protocol; each sensor has the two transmission range type: Sensing Trans- mission Range (STR) and Data Transmission Range (DTR).

3.2.13 | Low-energy Adaptive Clustering Hierarchy-Centred Cluster Head Low-energy Adaptive Clustering Hierarchy-Centred Cluster Head (LEACH-CCH) is a new clustering protocol aimed to enhance the MWSN lifetime.56 LEACH-CCH is an enhancement to the LEACH protocol used for static networks.

3.2.14 | Mobility-aware Centralized Clustering Algorithm Mobility-aware Centralized Clustering Algorithm (MCCA) that depends on the three-layer hierarchy aims to reduce data loss and achieve energy-efficient clustering.57 MCCA implements centralized gridding at two tiers.

In the grid head mode, the nodes collect information from other grid heads or ordinary nodes, aggregate it, and send the aggregated information to the upper-tier grid head or sink node.
3.3.1 | Mobile Sink based Routing Protocol
This protocol contains two stages: the setup stage in which the clusters are constructed and the MS informs the CHs about its location via a beacon message and the steady-state stage in which the TDMA schedule is transmitted to the registered CHs.

The CHs transmit their information to the sink which moves to another position to collect information based on the remaining energy of the CH nodes.

3.3.2 | Energy-Efficient Competitive Clustering Algorithm

The moving sink transfers at a particular velocity via a particular line path that is cantered the field and stays at equal distance positions to receive packets from surrounding CHs.

The results show that the EECC algorithm has better network lifetime and energy expenditure performance than LEACH.

3.3.3 | Optimizing LEACH Clustering Algorithm

An Optimizing LEACH Clustering Algorithm (O-LEACH)63 is a new clustering protocol that combines the LEACH pro- tocol with MS and rendezvous points (RPs) for reducing the expenditure energy.

The O-LEACH depends on dividing the network lifetime into a round, and each one contains two phases: setup phase and steady-state phase.

3.3.4 | Ring routing protocol
To resolve the hotspot problem and balance the load among the sensors, the ring routing protocol replaces ring sensors with standard sensors and changes the ring layout.

The ring routing protocol's dis- advantages include requiring each node to know its position as well as the positions of its neighbors, as well as repeat- ing the ring construction method until a closed loop is obtained, which consumes more energy in processing and overheads.

3.3.5 | Energy Management Algorithm with Multiple Sinks

Second, EMMS is divided into rounds for energy management, with each round containing a routing tree construction stage, a sojourn time calculation stage, and a data collection stage.

A routing tree at each temporary position of every moving sink in the closed tour is established depending on the remaining energy of all sensors in the routing tree construction stage.

The temporary time at each temporary location on the closed tour of each sink is computed in the sojourn time- calculation stage based on the MNs of the established tree.

3.3.6 | Particle Swarm Optimization -Based Routing Protocol with Mobile Base Station
A Particle Swarm Optimization -Based Routing Protocol with Mobile Base Station (PSO-MBS) routing protocol that depends on PSO utilizing MBS was proposed in Latiff et al.66 to achieve efficient utilization of the energy and amelio- rate the lifetime and PDR of WSNs.

As with many protocols, this protocol starts with the setup stage in which MS utilizes the PSO to choose the optimum temporary MS positions based on the cluster numbers and the sensor locations.
3.3.13 | An Energy-Efficient Routing Schema

This schema is based on splitting the whole network into several sections of the same size, while each section selects a CHs based on the weight of its MNs.

The simulation results demonstrated that this schema has better performance in aspects of energy expen- diture and network lifetime comparing with CCMAR and ECDRA.

3.3.14 | Improved Mobile Sinks Based Energy-Efficient Clustering Algorithm

An Improved Mobile Sinks Based Energy-Efficient Clustering Algorithm (IMECA) is a cluster-based routing protocol that depends on different predictable movement paths for the MSs.74 The main objective of IMECA is to decrease the distance between the CHs and the expected routes of the MS and increase network longevity.

3.3.15 | Particle Swarm Optimization Based Selection

The PSOBS protocol is utilized and enabled the sink for selecting its RP according to the data position of all sensors to reduce the EED, ensure net- work coverage, and preserve the sink path cost.

3.3.16 | Mobile Sink Improved Energy-Efficient PEGASIS-based Routing Protocol with Direct Transmission
Mobile Sink Improved Energy-Efficient PEGASIS-based Routing Protocol with Direct Transmission (MIEEPB-DT) improved energy utilization and network lifetime by switching the sink node instead of keeping it static.76 The basic principle of multi-chain, which divides the area into close and far nodes, is to reduce the distance between connected nodes in the chain by using fewer sensors and reducing the overhead.

3.4 | Protocols for the mobility of both sensor and sink nodes together

3.4.1 | Anycast Tree-Based Routing Protocol
This protocol orga- nizes and preserves the routing tree between the MSNs and the MS. All sensors on the constructed route utilize the unicast mode instead of broadcast mode for requesting transmission to transmit the sensory information from a sensor to the sink.

3.4.2 | Energy-efficient and reliable routing protocol (E2R2)
E2R2 is a protocol that deals with sink and node mobility as a whole divides the network into clusters by choosing one CH and two deputies CH in each cluster.78 To save energy and minimize re-clustering time, the sink nodeselects a collection of likely CH sensors and forms the CH panel in this protocol.

3.4.3 | Joint Nodes and Sink Mobility Based Immune Routing-Clustering Protocol for Wireless Sensor Networks

The Joint Nodes and Sink Mobility Based Immune Routing-Clustering Protocol for Wireless Sensor Networks (JNSMIC) protocol79 allows the mobility of the sink and the sensor nodes simultaneously.

During the clustering phase, the JNSMIC protocol analyses several objectives, including network coverage, consumption energy, residual energy, link connection time (LCT), and mobility.

3.4.4 | Comparison between protocols for the mobility of both sensor and sink nodes together
A comparison of the routing protocols for the mobility of both sink and sensor nodes together in MWSNs based on mobility pattern, several sinks used, control manner, mobile element, network construction, cluster density, cluster size, intra/inter-cluster routing, protocol goals, and applications