Energy Efficient Dynamic Sink Multi Level Heterogeneous Extended Distributed Clustering Routing for Scalable WSN: ML-HEDEEC

Abstract

The size of the wireless sensor network (WSN) is extending continually with use of IOT networks. The main difficulty for design wide area WSN is to maintain the higher stability period and energy efficiency (EE) for the routing protocols. The creation of clustering-based routing protocols was applied to the optimization of overall network energy. But, traditional clustering methods were unable to produce improved node heterogeneity, and extended network lifetime. Distributed clustering-based routing protocols are specially designed for enhancing the EE of the networks. In addition, EE can be improved by enhancing the heterogeneity of the node distribution. This paper aims to design the extended distributed clustering-based EE routing protocol. The heterogeneity of nodes is improved by introducing the additional intermediate advanced nodes layer in the network. Therefore, paper proposed to design the Multi-Level Heterogynous EDEEC rousing protocol called ML-HEDEEC by adapting optimum energy enhancing parameters. The notes are divided to normal, advance, advance-interdicted and supper nodes based on the energy allocated to them. The probability of nodes is modified for better clustering and cluster head election by introducing additional energy enhancement parameter. In addition, it is proposed to automatically adopt the network initial energy based on the scaling of network dimensions. This may lead to enhance EE of the network and may improve stability period. Finally, the results are evaluated for a case of WSN routing under the dynamic sink locations. Performance is compared for various distributed clustering protocols and other state of art protocols viz. LEACH, SEP, zonal- SEP, DEC considering the network scaling. Various performances of the network stability, packets sent to base station, and lifetime,.are defined for result evaluation. The network dimensions are scaled up to four times and proposed protocol is tested under scaling consideration. In addition, sink locations are also varied for dynamic sink locations performance evaluation. Overall paper efficiently designed and test heterogeneous improved routing protocol with extended lifetime and stability.

Introduction

The WSN are widely used for accruing the data in real time from fields. The excessive use of the Internet of Thing (IOT) based network for connecting the sensors together is expected to increase the network dimensions in near future. The network scalability may lead to reduce the relative node densities and thus have direct impact over energy required to communicate data within the network. Therefore, it is highly required to design and test the routing performance of scalable sensors networks. Through a base station (BS), data on the WSN is routed via several nodes. Increasing the EE of the routing protocols for the wide area scalable WSN [1] applications are still a challenging field of research.

In the sensor network, nodes gather data and communicate packets via sink nodes to the base stations (BS). Certain researchers [2, 3] have adopted the dynamic sink-based routing to enhance the EE of the networks with scalability. It is proposed to change the location of the sink to adopt the best network lifetime based on the observations. Another way of improving the energy performance is to design heterogeneous network node distributions. This paper proposed to design the multi-level heterogeneous (MLH) architecture of distributed clustering-based routing protocol for wide area scalable sensor network. In addition, a case study of dynamic sinks for MLH architecture is evaluated for optimum routing and EE performance.

The basic network can be homogeneous or heterogeneous in nature. The homogeneous networks primarily contain the similar type of nodes distribution. Since in WSN nodes communicate wirelessly, thus may leads to heterogeneity in terms of network topologies and communication technologies. The improvements in the heterogeneous distributions of the nodes are based on their energy, or nature of traffic and mobility scenarios. The distributed clustering-based protocols performance may be enhanced by improving the election process of sensors nodes. This can be supported by improving the energy and the election probability distributions which is the prime concern of this paper.

Contribution of Work The major contributions of current research work are to design the MLH based distributed clustering extended EE routing protocol based on the energy heterogeneity. Initially the extended review of the clustering based existing routing protocols considering the heterogeneity and scalability of the network is presented. Then an additional intermediate node based on the energy level is proposed to introduce for enhancing the performance of ML- HEDEEC based distributed clustering routing protocol. The CH election probabilities are optimized to achieve better energy heterogeneity. As an experiment a case study of dynamic sink selection-based routing is presented. The sink location is dynamically varied based on the experiences and proposed to enhance the energy performance. The performance of the existing routing protocols is evaluated under the consideration of network scalability. The node initial energy is adapted in respect to scalability of the nodes. The performance of proposed protocol is evaluated based on the higher node density and dimensional scalability to enhance stability period and packet transmissions (Table 1).

Table 1. Nomenclature used for the study

Short form	Abbreviation	Short form	Abbreviation
AES BP	Advance Encryption Standard Blood Pressure	HART	Highway Addressable Remote Transducer
BLE	Bluetooth Low Energy	GSM	Global System for Mobile communication
DEC	Distributed Energy Clustering	LoRa	Long Range
GA	Genetic Algorithm	LIDAR	Light Detection and Ranging
ECG	Electro Cardiograph	WAN	Wide Area Network
EDEEC	Enhanced Distributed Energy Efficient Clustering	LASER	Light Amplification by the Stimulated Emission of Radiation
IOT SEP	Internet of Things Stable Election Protocol	IR FBG	Infra-Red Fiber Brag Gritting Sensor
LEACH	Low Energy Adaptive Clustering Hierarchy	LVDT	Linier Variable Differential Transformer
WSN	Wireless sensor network	RTD	Resistance Temperate detector
PEGASIS	Power Efficient Gathering in sensor information system	GIS Wi-Fi	Geographical Information System Wireless Fidelity
PA MHz	Precision Agriculture Mega Hertz	WPAN GPS	Wireless Personal Area Network Global Position Systems
GHz	Giga Hertz	IP	Internet Protocol
DSRC	Distributed Short Range Communication	AG-IOT	Agriculture IOT
UHF	Ultra-High Frequency	NB-IOT	Narrow Band IOT
IOMT	IOT multimedia technology	MIMO	Connected Disability awareness Network
LPWA	Low Power Wireless Area	CDAN	Multi Input Multi Output
HTTP	Hyper Tests Transfer Protocol	RPL	Routing Protocol in Lossy Network
CoAP	Constrained Application Protocol	EE-LEACH	Energy Efficient LEACH
MQTT	Message Queuing Telemetry Transport	HEED	Hybrid Energy Efficient Distributed Clustering
CH	Cluster Heads	MCH-SEP	Multi Cluster Head SEP
MAC SPT	Medium Access Control Shorter Path Tree	E-SEP DEEC	Enhanced SEP Distributed Energy Efficient Clustering
A-NN	Artificial Neural Network	DDEEC	Developed Distributed Energy Efficient Clustering
HADA	Hybrid Data Aggregation	SWT	Soil Water Temperature
z-SEP MLH	Zonal SEP Multi-Level Heterogeneous	EEZ-SEP	Energy Efficient z-SEP

The WSN's best popular application areas are listed in Fig. 1. Thus, it's really clear that there has been tremendous scope of WSN in different crucial application areas where humans can’t be there for whole time. These applications include target coverage [4], structure health monitoring [5], underwater sensor network (UWSN) [6], monitor environment and disaster management [7] applications like smart city, health care and agriculture [8]. Some programs are so extreme that people can't even stand themselves. Different applications require different sets of sensors, which may also require various communication systems. Recently, some WSN applications have inspired the use of dynamic sink, as investigated by Awais Ahmad, et al. [9]. Dynamic sink has a variety of uses, such as enemy surveillance and monitoring systems, seismic activity monitoring, medical and healthcare monitoring systems, and more. Moisture and pH sensors, gas sensors for monitoring environment, temperature and humidity sensors, pressure sensors, strain gauges, and piezo sensors are among the many sensors utilized in the scalable WSN for various applications.

[See PDF for image]

Fig. 1

A quick overview of the WSN networks' most popular scalable applications

Routing techniques based on distributed clustering are widely used recently due to a number of advantages it provides over wireless network as:

1. Distributed cluster based on routing network supports scalability in mobile wireless sensors networks.

2. Minimizes wireless networks' energy usage.

3. Increases network lifetime by decreasing energy uses.

4. It is possible to dynamically alter a number of CH using DC-based routing protocols.

5. The performance of DC is unaffected under increasing node densities.

6. The issue of data redundancy has been totally eliminated using distributed CH’s

7. A number of distributed clusters either fixed or variable, of any size can be used (same or different). The size of the cluster won't cause any problems..

Therefore, all these benefits are major motivation behind designing a DC based routing protocol in rest of this paper.

Challenges of Scalable Heterogeneous Sensor Networks

The common challenges may include mobility, data aggregation, limited available DC power of nodes, and connectivity issues etc. In addition to all basic challenges of WSN network there are certain issues to be addressed specific to scalable heterogeneous networks. The scalable sensors networks are classified as shown in Fig. 2. The specific goal of the IOT based networks is to provide global coverage. Thus, there is no limit of the scalability for these networks.

[See PDF for image]

Fig. 2

Various types of the scalable wireless networks

The scalability of the network may be defined in terms of either scalable mechanism or in term of change in network dimensions and workload. The workload is increased with increasing node densities. The challenges of the scalable WSN have been sequentially addressed in this section:

1. The Network Dynamics The scalable networks of sensor are quite dynamic by their very nature. Within the WSN network, the sensor node, the object of interest, and the data observer might all be mobile phones which are dynamic.. Additional, nodes may often connect to the network and this may lead to change network dynamics and leads recurrent nodes to fail. It is because of the limited capability of the routing protocols. Is scalability of network may result, the dynamic changes within WSN network architectures. It’s an open challenge for routing algorithm to perform well under dynamic networks by adopting self-tuning.

2. Limited Power and Energy of WSN The sensor is DC powered by the batteries, so network power is very scarce. As a result of low source energy, a WSN node frequently died. The increasing network dimensions may lead less node densities and therefore may require more energy to communicate and route data.

3. Computing Capability In WSN networks nodes contain a microcomputer and memory. These sensors are equipped with the ability to process data. Nevertheless, the capacity and capabilities of the microcontrollers computing and on-board memory are limited, which also limits the ability to handle sensors. Utilizing the numerous sensors with the limited computer resources for distributed and cooperative data processing is still a difficult task.

4. Latency under Limited Communication Range Even though it's dynamic nature, the WSN in most consumer applications has a restricted communication capacity, which restricts its coverage range. But communication management is a big challenge for large scalable networks. The main problem is that sensor connectivity frequently breaks down when load is higher owing to bandwidth restrictions. The network could go offline for a considerable amount of time if a sensor is disconnected. One open research topic is maintaining connection and connectivity to ensure accurate sensor data transfer even with low bandwidth.

5. Density of Nodes Deployment The number of nodes in a scalable WSN might range from several to many. The deployment density of nodes can be changed to collect high resolution data. In order to get a high and better transmission, this can enable the nodes to connect to other nearby nodes. So, in order to accomplish this, an accurate heterogeneous protocol is required to prevent collisions. A higher throughput is offered by its multiple accessing. The protocols have challenge to readily and effectively maintain dense network performance because they are scalable.

Survey of the Routing Protocols for WSN

There are huge numbers of routing protocols have been already designed in past for data communication amongst the WSN nodes. The first basic challenge of sensors networks is to aggregate the data from nodes. Data aggregation routing's function is to combine data from various sensing devices as well as distribute it across the network [10]12. Data aggregation is also accountable for removing duplication, reducing the number of transmissions, and in turn conserving the network energy [10]. Clustering is a popular technique for raising the WSN network's energy effectiveness. By minimizing the required communication energy, distributed clustering mechanism used in WSNs maximizes overall life of the network. Distributed clustering minimizes direct transmission to the BS's and lowers energy usage by shortening the transmitting distance [11]. The use of scalable WSNs for data collection and processing is widely acknowledged. The study's main focus is on reviewing issues with such clustering-based WSN routing and assessing WSN performance.

Heterogeneous Routing Methodologies

The focus of this paper is to primarily review the clustering based heterogeneous routing protocols designed for WSN’s. The broad classification of the various heterogeneity-based routing methodologies based on network architecture and distributed clustering is given in the Fig. 3. Heterogeneous routing protocols methodologies based on the distributed clustering are considered for the extended review in this paper along with the conventional Flat, Tree, Location and Chain based routing approaches of WSN design. This article briefly discusses hierarchical routing techniques that are too focused on distributed clustering-based methodologies, as indicated by the red marked blocks in Fig. 3. Clustering based methods are considered for review based on the centralized and distributed approaches. Each of the mentioned field in Figure are sequentially reviewed in this paper. More specifically these protocols are based on the EE improvement.

[See PDF for image]

Fig. 3

WSN heterogeneous Distributed clustering routing protocols extended classification

Flat Method All sensor nodes in flat routing methodologies of WSN are connected to one another to accomplish the sensing function. The BS broadcasts the inquiry message to all of the sensors. The information is transmitted to the BS by the nodes involved in the inquiry. The data transmission may involve multi-hop path; thus, these methods may lead to higher network delay. Data aggregation only takes place in a limited area close to a sink node, which is one restriction of this flat network design as given by Sasirekha et al. [12] The overhead in calculations causes a speedier discharge as a result limit life.

Hierarchical Clustering Methodologies In heterogeneous network, the CH is responsible for routing. Clustering reduces energy consumption and improves the life under the scalability of sensor-based networking, helps in conserving energy. In heterogeneous routing nodes are divided based on energy levels and sensor nodes inside a cluster provide their data to the CH, who aggregates it [13]. In clustering-based network routing the CH assigned by the networks performs task of data aggregation. The wide area's WSN energy uses are significantly lowered by this technique. The most frequent clustering based aggregating protocol is LEACH. When a CH fails in the LEACH protocol, the network's highest energy node is chosen to act as the CH. There are several LEACH variations available, such as EE-LEACH, which extend the network's life and use less energy [13]. Clustering based other WSN routing are proposed in LEACH-MAC [14] and are focused on EE cluster-based routing [15]. The extended surveys of the clustering-based routing are being presented in the [16]. A stable election-based protocol (SEP) has been introduced by G. Smaragdakis et al. in [17] for clustering and routing sensors nodes of the WSN. This technique works well for clustering networks. In light of heterogeneous clustering, G. Smaragdakis has suggested the very first stable election protocol, known as SEP. Based on weighted election probability; node energy is distributed in this situation. This technique clusters the networks well. A deterministic heterogeneity-based DEC protocol [18] is a routing protocol based on EE clustering but suffers from the fast dyeing of nodes.

Chain Based Routing Every single sensor node of chain based WSN is connected to its neighboring sensor nodes in the chain. The leader is selected based on the TOKEN allocation technique by lowering the overhead computation of dynamic protocol. The hierarchical chain-based routing is the foundation of the PEGASIS [19] protocol. Haydar et al. [20] have proposed to evaluate the performance of chain-based routing in their research.

Location Based SPAN Protocols In this type of network method the global positioning system (GPS) is used to identify the location of wireless sensor nodes, based on that the task of aggregation is performed [21]. To reduce the energy consumption the non- active nodes always be in sleep mode. SPAN protocol belongs to in this category.

There are certain hybrid approaches (HA) uses a combination of various data aggregation (DA) and clustering approaches. For instance, I-LEACH protocol [22], is an upgraded version of LEACH, is used in HADA, a hybrid data aggregation method, which reduces energy consumption [22].

Review of WSN Routing Protocols

The concern of paper is to discuss various WSN routing protocols that to clustering based routing. An energy optimal solution for the WSN's was presented by HE Bin, èt al. [23]. They have demonstrated how well the energy inside the WSNs can be optimized. When figuring out the best order to transmit data from source nodes and sink nodes, they have used Dijkstra shortage path algorithm. However, the approach seems a little slow and traditional. Bhakti Parmar et al. [24] have conducted a thorough analysis of the clustering-based protocol known as the hierarchical Low energy dependent adaptive clustering hierarchy, or LEACH. (LEACH). They have provided a variety of enhancements to the basic LEACH that make it possible to use WSN, namely LEACH-SM, S-LEACH, and R-LEACH, which have been discussed in [24]. The LEACH routing algorithm is frequently used in WSN. As a result, creating an effective protocol algorithm using the concept of minimum energy is necessary.

Genetic Algorithms (GA) with A-NN has been suggested by Stephan et al. [25] as a technique to localize a plan for sensor networks. The method was too time-consuming and impractical for real-time applications like soil monitoring since they presented a WSNs design employing machine learning. Three-tier heterogeneous design was used in the improved clustering approach Aderohunmu et al. [26] suggested utilising extended-SEP protocol. By adding the intermediate nodes layer, they have improved the effectiveness of the fundamental SEP protocol. Jagadeesh et al. [27] has a goal to design a version of SEP by using MCH-SEP approach for multiple CH selection. The method greatly outperforms the standard SEP. The effectiveness of the SEP and LEACH routing methods were contrasted by Sharma et al. [28]. For SEP protocol expansion, Singh et al. [29] have developed a modified technique. The approach was basically modified by making use of k-means clustering for choosing cluster heads. Performance of the method is greatly improved over SEP and E-SEP protocols.

For first time, DEC, a deterministic clustering technique for improving EE, was created by Aderohunmu et al. [30] According to certain reports, DEC outperforms over the SEP and LEACH protocols in terms of latency. However, the main difficulty in scaling WSN design is energy efficiency. DEC died sooner in terms of overall network life. For heterogeneous WSN networks, Qing et al. [31] almost two decades back have designed the first version of the distributed clustering (DC) based EE heterogeneous routing protocol abbreviated as DEEC. The uses of DC allowed enhancing the network life.

Prasada et al. [32] presented the parametric modification-based design of the DEC distributed routing protocol and achieve life of 1830 as FND. Dinesh et al. [33] recently have presented the extension of the DEC protocol using the multi tire heterogeneous network design to enhance the lifetime. The probability of CH election was modified and parametric optimization-based performances are compared by Dinesh. Although the fast dying of nodes once first node died is still a challenge. Javaid et al. [34] created an enhanced distributed EE clustering EDEEC protocol. They presented a brand-new clustering-based routing strategy in this study for heterogeneous WSNs. The process dynamically alters the chance of choosing a cluster head, making it more effective [34]. Sharma et al. [35] have designed the hybrid data aggregation and DC based routing protocol. Redjimi et al. [36] have contrasted the performance of DEEC with EDEEC routing protocol and reported the heterogeneity improvement for extended lifetime. Heterogeneous networks are illustrated by Brahim et al. [37]'s presentation of the DEEC, a DC-based EE protocol for WSN design. In this protocol, the network model is described. It is assumed that there are N sensor nodes, that are evenly dispersed throughout a M*M square area and are designed to be scalable. The similar research is reported by the Priya et al. [38] but for EDEEC protocol.

There are two type of clustering approaches centralized and distributed as compared by the Zanjireh et al. [39]. They have reported the edge of distributed clustering over centralized clustering for WSN routing. Qureshi et al. [40] have designed an improved energy performance-based BEENISH protocol to be for the WSN networks. The reported lifetime is extended beyond the 2000 but at the cost of initial energy. A zone-based routing protocols extension is designed by the Naresh et al. [41] called the EE-z-SEP protocol. Behera et al. [42] have designed another improved version of heterogeneous SEP protocol abbreviated as I-SEP. they extended the level of nodes energy distribution. Certain review of clustering-based routing protocols has been presented in the [43, 44], 45. The overall it can be stated and concluded from the discussion that improving the level of heterogeneity and using DC based methodologies may enhance the EE of the routing protocols.

Review of Scalable WSN

The development of scalable routing protocols for WSNs seems to be another area of active research. A scalable design of WSN for a down-ward protocol has been developed by X. Zhong et al. [46] for use in the IOT networks. They developed an algorithm for the down-ward routing for WWAN a scalable WSNs using opportunistic source-based routing (OSR). They utilized a Blooms filtering to increase the OSR method's scalability. Salim et al. [48] improved.'s clustering-based technique for WSN routing, which uses K-means based fuzzy clustering, has been proposed and assessed [47]. Nandan et al. [48] have devised a multiple -route based routing technique that's been used for the aforementioned scalable WSN networks. The efficiency of DC-based routing significantly enhanced thanks to its design. Because battery power is limited for WSN nodes for routing-specific to WSN routing scalability difficulties were solved. Scalable routing requires optimal use of node batteries. The congestion of the nodes has emerged as a challenge that must be considered while routing in scalable WSNs with high node density. The network's capacity limit of coverage is defined by the nodes' limited ability to send data due to bandwidth limits.

The scalable structure of the WSN routing protocol design has been published by Sandhya et al. [49], as well as the design of the WSN utilizing a cross-layer architecture oriented parametric assessment has been provided by Kumar et al. [50]. The overview of the numerous WSN attack strategies has also been offered. Scalability issues in WSN routing protocols has been addressed by Singh et al. [51]. They have assessed the effectiveness of node counts versus CH counts for the various routing methods underneath the scalability of sensor networks. Dead nodes, CH counts, and FND were taken into consideration as the assessment criteria for up to 1500 rounds and a variety of nodes up to 1000. The paper provides useful insight into WSN scaling.

For scalable EE-WSN routing, Rani et al. [52] have given the analysis of the linear scalable protocol version for the various routing protocols, Anurag et al. [53] have offered the 3 scenarios of WSN performance taking into account static and dynamic nodes, for scalable EE protocol (SEEP) design. They suggested scenario 3 with SEEP, in which the node is free to roam between different zones, outperforms. However, SEEP method performance for such DEEC protocol is shown to be poor and needs to be enhanced.

Yadav et al. [54] have proposed an enhancement of the I-DEEC protocol's scalability. For the comparison research of the WSN networks, Maitra et al. [55] have published a survey of numerous MAC protocols. The innovative method for heterogeneous EE routing protocol employing DEEC has been introduced by Samayveer Singh [56]. A protocol based on threshold balancing-TBSDEEC has been developed by Sercan et al. [56] to increase the network's heterogeneity. They demonstrated that the suggested approach is more EE, in comparison. However, the final node to die is comparatively low because the other nodes all perished more quickly after the initial one. For scalable WSN networks to perform better in terms of energy efficiency.

Khan et al. [57] have proposed the hybrid approach of DEEC routing protocol. To improve Sheenam et al. [58] developed a gateway-based multi-hop DEEC protocol known as G-DEEC for routing WSN's. For WSN, Singh et al. [59] developed a scalable routing mechanism to extend network life to improve energy efficiency. A comparable contribution is shown in the [60].

Summary of Routing Protocols

This section provides a comparison of the network longevity, latency, and heterogeneity performances of the current protocols for routing WSN. For comparing the clustering-based routing protocols, summary of performance comparison is presented in Table 2.

Table 2. A summary of a routing protocols applicable to WSN design

Author/ Protocol	Methodology	FND Latency	Clustering	Energy Efficiency	Heterogeneity	Protocol Organization
Sasirekha et al. [12]	Energy minimization using hierarchical data clustering and aggregation	NA	Data aggregation	Yes	No	Hierarchical
G. Smaragdakis et al. [17]; SEP	CH's are determined using stable election probabilities	1385	Probability based	Yes Based on stable election	2 level	Hierarchical
Gurjit Kaur et al. (19] I –DEC	Have created a more effective distributed clustering algorithm by distributing energy more effectively	1979	Distributed	Yes Residual Energy	Yes	Hierarchical
Lindsey, S et al. [19] PEGASIS	Developed an effective power-gathering technique for the information management. Distance is used to pick CH	1491	NA	Good	NA	Chain based
Bhakti Parmar et al. [24]	An analysis of the energy-saving LEACH procedure variations	995	Adaptive clustering	Yes	No Homogenous	Hierarchical
F. A. Aderohunmu et al. [26] SEP-E	Cluster heads are selected based upon stable election probability. However, the idea of an intermediate node increases the number of tires to two	1449	Probability based	Yes Intermediate layer based	3 level	Hierarchical
L Jagadeesh et al. [27] MCHSEP	Modified SEP employing k means clustering for CH selection	1496	Multi cluster	Yes Multi-level CH	Multi-level	Hierarchical
Praveen Kumar et al. [29] M- SEP	Reworked on SEP approach using the advanced nodes energy to lengthen life of network	1532	Probability based	Good Multi-level	3 level	Hierarchical
M. M. Prasada et al. [32] DEEC	Have suggested a routing system that uses clustering that is effective, distributed, and low-energy.	1231	Distributed	Yes Good	2 level	Hierarchical
Dinesh et al. [33] DEC	An multi tire heterogamous improved DEC routing with 4 level architecture	2023	Deterministic	Good	Yes 4 level	Hierarchical
Brahim Elbhiri et al. [37] DDEEC	Has suggested using the improved DEEC methodology for EE design	1263	Distributed	Yes Very Good	3 levels	Hierarchical
Priya,Rashmi et al. [38] EDEEC	Suggested for employing clustering a multi-level effective, distributed, and EE routing algorithm	1324	Distributed	Yes Very Good	Multi- level	Hierarchical
T. M. Behera et al. [42]	I-SEP routing protocol clustering using threshold energy for CH election	2837	Probability based	Good	3 level	Hierarchical
Dongyao Jia et al. [43]	Protocol for dynamic cluster-based routing	1500	Dynamic	Better	Single level	Hierarchical
G. Karthik et al. [44]	Reviewed the WSN protocols for energy heterogeneity for clustering	NA	EE	Good	NA	Hierarchical
Sandhya et al. [49]	EE clustering-based routing for scalable WSN design	NA	Distributed	Good	NA	Hierarchical

It can be closely observed from Table that distributed E-DEEC based protocols performs better than other state-of-the-art protocols, despite the fact that there is still room for further performance enhancements as demonstrated by recent SEP protocol extensions. Increased awareness of multi-level node distributions is anticipated to improve the performance of E-DEEC method.

Thus, based on the review it can be summarized that E-DEEC protocol can perform better than similar heterogeneous routing. Thus, paper first defied the mathematical modeling for EDEEC protocol and then extended the work to our proposed multi-level heterogeneous (MLH) routing version.

Modeling of E-DEEC Protocol

For modeling the routing issue, Enhanced-Distributed EE Clustering (DEEC) routing protocol is suggested in this paper. The protocol is designed based on residual energy (RE) idea and three levels of the node’s probabilities corresponding to normal, advance, and supper nodes were presented by the model. An additional third advance tire was added by the EDEEC as top nodes, hence establishing the respective probability distribution. The probability counts for advance nodes, super, and normal nodes are as follows:

1

$$N_{\mathit{no}}=n\ast \left(1,-,m\right);$$

2

$$N_{\mathit{ad}}=n\ast m\ast \left(1,-,m,o\right)$$

3

$$N_{\mathit{sup}}=n\ast m\ast mo$$

The values of the number of nodes and respective scale parameters used by EDEEC [26] are given in the Table 3. It can be observed that there are 3 levels of nodes and thus their energy levels are also required to adopt accordingly in hierarchy.

Table 3. List of the design parameters for EDEEC routing

Parameter Variable	Description	Values
n	Number of total nodes in WSN network	100
m	Scaling factor for normal nodes	0.5
mo	Scaling factor for advanced nodes	0.4
$N_{\mathit{no}}$	Number of the normal nodes in network	50
$N_{\mathit{ad}}$	Number of the advance nodes in network	30
$N_{\mathit{Sup}}$	Number of the advance nodes in network	20

The probabilities of the cluster head CH selection corresponding to the 3 level Normal, advance and supper nodes are;

4

$${p(i)}_{\mathit{Norm}}=P\ast E_o\left(1,+,m,,\ast ,\left(a,+,m,o,,\ast ,b\right)\right)\ast Ea$$

5

$${p(i)}_{\mathit{ad}}=P\ast (1+a)\ast E_o(i)/(1+m\ast (a+mo\ast b))\ast Ea$$

6

$${p(i)}_{ad-int}=P\ast (1+b)\ast E_o(i)/(1+m\ast (a+mo\ast b))\ast Ea$$

The sink positions are placed static as;

7

$$S_{i,j}=Sink(x_i,y_j)$$where variables $i$ and $j$ are the coordinate of sink positions basically placed at the center at 50, 50 locations. The CH election energy is defined as; for $E_{\mathrm{elec}}$ = 50 nJ/bit energy model as referenced by [26], for transferring k bits;

8

$$E_{T_{x}x}\left(K,D\right)=E_{T_{x}x-elec\left(k\right)}+E_{T_{x}x-amp\left(k,d\right)}$$

9

$$=\left\{\begin{array}{cc}K{E}_{\mathit{elec}}+k€f_s{d}^2& ifd<d_0,\\ K{E}_{\mathit{elec}}+k{€}_{\mathit{mp}}{d}^2& ifd\ge d_0,\end{array}\right)$$where $E_{\mathrm{elec}}$=$E_{\mathrm{o},}$ and d is defined as the measure for Euclidian distance (Fig. 4).

[See PDF for image]

Fig. 4

Energy model of the DEEC protocol Tx-Rx

Even though of using 3 levels heterogeneity in EDEEC, still there is lot of scope of energy enhancement for use in scalable WSN. Thus, the prime goal of this paper is to enhance the performance of existing distributed clustering protocol considering scalability by improving the heterogeneity.

Proposed ML-HEDEEC Protocol Design

To achieve the goal of enhanced latency and life of the WSN routing paper proposed to design the multi-level (ML) energy distribution based heterogeneous (H) protocol. The energy distributions of the nodes are improved based on adding the extra 4th intermediate level among the advance and super nodes in the network. The node distribution is modified based on the ML election probabilities of the CH in the network. The trans-receive energy model of the network is kept constant. The mathematical modeling of election probabilities, nodes ratio and the energy scaling parameters optimization is used for improvement. The modified node densities and probabilities are mathematically defined as;

10

$$N_{\mathit{no}}=n\ast \left(1,-,m\right),N_{\mathit{ad}}=n\ast m\ast \left(1,-,m,o\right),$$

The additional advance – intermediate node layer is introduced for improving the heterogeneity level and number of nodes is defined as.

11

$$N_{ad-int}=n\ast m\ast imo,$$

12

$$N_{\mathit{sup}}=n\ast m\ast \frac{\mathit{imo}}{2},$$where variable $\mathit{imo}$ is introduced as the new scaling parameter for the selection of the ad-int nodes. The modified election probabilities are defined as;

13

$${p(i)}_{\mathit{Norm}}^{\mathit{ML}}=P\ast E_o\left(1,+,m,,\ast ,\left(a,+,m,o,,\ast ,b,+,c,,\ast ,i,m,o\right)\right)\ast Ea$$

14

$${p(i)}_{\mathit{ad}}^{\mathit{ML}}=P\ast (1+a)\ast E_o(i)/(1+m\ast (a+mo\ast b+c\ast imo))\ast Ea$$

15

$${p(i)}_{ad-int}^{\mathit{ML}}=P\ast (1+c)\ast E_o(i)/(1+m\ast (a+mo\ast b+c\ast imo))\ast Ea$$

16

$${p(i)}_{\mathit{Sup}}^{\mathit{ML}}=P\ast (1+b)\ast E_o(i)/(1+m\ast (a+mo\ast b+c\ast imo))\ast Ea$$

Apart from the $a$ and $b$ extra parameter for election probability scaling for the advance – intermediate nodes are introduced as $c$ and is optimized for enhanced network performance.

The design methodology is presented in the Algorithm 1 for the proposed ML-HEDEEC protocol for the WAN routing under scalability.

Based on the statistical analysis in terms of number of alive nodes the sink location may change dynamically. The specific case study is given in the result section.

Results and Evaluation

The network design parameters for scalable WSN design are optimized experimentally. The optimum modified parameter values are shown in the Table 4.

Table 4. optimally selected values of design parameters for ML-HEDEEC

Parameter Variable	Description	Values
n	Number of total nodes in WSN network	100–400
m	Scaling factor for normal nodes	0.5
mo	Scaling factor for advanced nodes	0.6
$N_{\mathit{no}}$	Number of the normal nodes in network	50
$N_{\mathit{ad}}$	Number of the advance nodes in network	20
$N_{\mathit{Sup}}$	Number of the super nodes in network	10
$N_{\mathit{ad}}$-int	Number of the advance—intermediate nodes in network	20
Eo	Network total Initial energy	F * 0.5 J
$E_{\mathit{tx}}$	Transmission energy	50 $\mu \mathrm{J}$
$E_{\mathit{rx}}$	Receive Energy	50 $\mu \mathrm{J}$
p	CH probability	0.1

According to the initial assumption of 100 nodes and 5000 rounds being static, the performance of the current routing protocols. First Fig. 5 contrasted the results of the number of alive nodes displayed against the number of rounds.

[See PDF for image]

Fig. 5

Performances of current routing methods tested on 100 nodes for max 5000 rounds for the percentage of alive nodes

Figure 5 makes it very evident that when the first node failed, the DEC protocol network collapsed more quickly. Additionally, the E-SEP procedure performs well under 2000 rounds in comparison. The EDEEC protocol performs significantly better in terms of the network's extended life. It is so because, using the EDEEC protocol, about 50% of nodes are still functioning after more than 3000 rounds. Figure validated the performance of existing state of art routing protocols and that to with our proposed multi-level heterogeneous ML-HEDEEC routing protocol. The improvement in terms of the prorogated network lifetime offered by proposed method can be clearly observed. It is because of the enhanced energy distribution amongst the ML nodes. The stability time or latency of EDEEC is enhanced more than of 15% by using proposed ML-HEDEEC protocol.

Based on the study of existing routing protocols in the Fig. 5 the network lifetime performance of these protocols are summarized and compared in the Table 5. The enhancement of network lifetime as best time is marked by red color in the table. The performance improvement offered by the proposed routing protocol is clearly observed by graphical plot shown in Fig. 6 as the around 25% improvement is noted in terms of time of Half Node Died (HND) and 50% improvement over last node died (LND) by ML-HEDEEC.

Table 5. The life time comparisons of state of art routing protocols

Parameters	LEACH	DEC	E-SEP	EDEEC	ML-HEDEEC
FND	1019	1849	1106	1202	1323
HND	1126	1897	1236	1497	2092
LND	4338	1920	5020	4093	8000

[See PDF for image]

Fig. 6

Graphical performance improvement reported by the proposed method for lifetime

Comparison of the number of alive nodes for the 100 nodes deployed in the 100 m × 100 m region for EDEEC and the proposed ML-HEDEEC are shown in the Fig. 7. The stability time performance improvement is highlighted by the orange box in the Figure on the top corner. This stability period represents the FND or latency of the network.

[See PDF for image]

Fig. 7

Result comparison of the number of alive nodes for EDEEC and proposed ML-HEDEEC for the 100 nodes and 100 m × 100 m area

Packets Sent BS The network throughput or the data transmission performance is accessed in terms of the number of packets sent to the BS by the protein each round. The results improvements in the quantity of packets transmitted to the BS by the proposed MLH protocol version and basic EDEEC are presented in the Fig. 8. The maximum packets transmuted in the network are almost doubled for considering the 100 nodes and for the basic 100 m × 100 m network size.

[See PDF for image]

Fig. 8

The results improvement in terms of number of packets sent to BS

Experimentation for Optimal Parameter Selection

The series of trials with the different energy scaling parameters and the percentage of ML nodes are performed to set the optimal values achieving the maximum stability period.

In order to enhance the life of the network the performance of proposed optimum ML-HEDEEC routing is compared with that of existing EDEEC protocol. Based on the energy and node scaling parameters two versions of the ML-HEDEEC protocols are compared. The first version (1) has selected the values as a = 1, b = 3, c = 2, while second version selected it as optimum to a = 1, b = 2.5 and c = 2. Thus, it is clear from the Fig. 9 that the proposed ML-HEDEEC has outperform over existing EDEEC. And in order to achieve the better stability period the parametric optimization is must.

[See PDF for image]

Fig. 9

The result comparison of proposed ML-HEDEEC with different network nodes energy scaling parameters. a, b, c for E0 = 0.5 J and n = 100

It can be observed from the Fig. 9 that ML-HEDEEC (Red) seems to have mode alive nodes at the end of rounds. But we need a closure look at the latency or stability period at the top where 100% nodes are alive. Thus, in terms of latency it is concluded that ML-HEDEEC [O] represents the optimal network performance.

Results under Network Nodes Scalability

In order to evaluate the performance of the proposed MLH routing approach under the network scalability in ms of number of nodes and the network dimensions this section compares the number of alive nodes for different scalability ranges for optimal ML-HEDEEC parameters as already discussed.

Theorem 1

Based on the experience it is observed what as the network node density or the dimensions are scaled up it is advantageous to scale the network initial energy in same proportion. The relation amongst the initial energy and network dimensions or node density are mathematically related as;

$${\mathit{Eo}}_Q^S=Eo\ast \sqrt{Q}$$where${\mathit{Eo}}_Q^S$ is the scaled energy, if the network dimensions are expected to scale Q m².or the node density is scaled only$\sqrt{Q}$ times. For example, if the network dimensions are scaled fr100 × 100 m² to 200 × 200 m², then Q = 4 the scaling factor. Thus, the Eo will becomes doubled. Although, the practicality of the Theorem in real-time is big challenge in near future.

Proof and Discussion To justify the statement of Theorem experimentations have been performed and the Results of the alive nodes are plotted with the different node density as scaling rage. The node density are scaled from 100 to 200 with Q = 2. The results are plotted in the Fig. 10.

[See PDF for image]

Fig. 10

Node density vs adaptive energy scaling analysis

It can be noted in the Figure that the network dimensions were kept to 100 × 100 m² constant in the experiment and the nodes densities are also kept same to 100 nodes. But network initial energy is doubled from 0.5 to 1 J. and results are compared as shown in the Red color in Fig. 10. It was observed that enhancing the Eo may significantly improve the lifetime. But this does not seems practical to increase the energy for small network size as it leads hardware cost.

Thus, in second part the Theorem 1 is proposed where node density is doubled keeping same dementias and the energy Eo is scaled accordingly as shown by green curves in Fig. 10. Thus, it is found that it is beneficial to enhance the energy in proportion to expected scalability of network nodes or dimensions. Although increasing network dementias may reduce the latency but improve life with energy scaling.

Dynamic Sink Case Study

As an experiment the three sink locations are varied as (25, 75), (50, 50), (75, 25)] as per the Grid based routing. It is proposed to evaluate the network life with these three sink locations which performs differently under different set of rounds. It is proposed to adopt the sink location corresponding to the best performance of the sink location in terms of alive nodes. It is observed that around 1400 nodes the sink location at center with (50, 50) coordinates perform better then sink is changed to (25, 75) for around 600 rounds and again for next 1500 rounds up to 2000 to 3400 rounds sink proposed to get back (50, 50). But for last phase of lower network energy the sink location of (75, 25) is proposed to adapt. The results of sink dynamic adoption are plotted in the Fig. 11 for the EDEEC protocol. The detailed experimental analysis of dynamic sink placement is done by Gupta et al. [61].

[See PDF for image]

Fig. 11

Number of Alive nodes for the adaptive dynamic sink location for 100 nodes and 5000 rounds

Conclusions

The major contribution of the paper is to design the multi-level heterogeneous extension of EDEEC routing called ML-HEDEEC for energy enhancement and prolonging the life. It is also proposed presented the case study of the optimum adaptive dynamic sink locations for lifetime and Alive nodes enhancement. The major conclusions observed are as follows,

• Paper presents detailed survey of existing research methodologies specific to design routing protocols for IOT based scalable wide area WSN. The survey is done based on various applications of WSN, and various routing protocols for communication. Paper contributed to design the new ML-HEDEEC routing protocol by modifying the energy distribution probabilities and the four layers of heterogeneity. It is concluded that using multi-level node structure can enhance the network life significantly by saving energy uses.

• IOT-WSN is now frequently used and have gained popularity for their worldwide reach. For the WSN, a lot of study has gone into establishing an effective routing protocol. As the EE of the nodes affects the network's energy performance. It has been noted that the energy efficiency may be increased if networks are heterogeneously distributed. Data transmissions or routing use up the majority of the energy. Therefore, it is necessary to enhance these networks' EE.

• The major challenges for scalable distributed clustering based WSN design are addressed based on the heterogeneity and energy used of network. In heterogeneous WSN networks, efficient selection of the cluster heads (CH) is a challenging field of research.

• Performances are compared for the most recent routing techniques tested on 100 nodes for up to 5000 rounds in terms of latency, the proportion of living nodes, and the number of packets transmitted.

• According to the proposed ML-HEDEEC, the ideal energy scaling factors are a = 1, b = 2.5, and c = 2. By using the proposed ML-HEDEEC protocol, the stability time or latency of EDEEC is increased by more than 15%.

• The period of Half Node Died (HND) has improved by around 25%, and the last node died (LND) has improved by about 50%, according to ML-HEDEEC.

• When 100 nodes are taken into account together with the standard network size of 100 m × 100 m, the maximum number of packets that may be transmuted increases by roughly twofold.

Overall, it is concluded that heterogeneity and scaling energy in terms if scaling factor of network load and dimensions may enhance network performance.

Authors Contribution

All authors have participated in (a) Conception and design, or analysis and interpretation of the data, (b) Drafting the article or revising it critically for important intellectual content, and (c) Approval of the final version.

Data Availability

The data query is generated through MATLAB simulator for evaluation purpose.

Code Availability

The code for different existing protocol is available at MATLAB repository. By changing different parameter related to sink position and number of nodes in a particular sensor area the comparative study and statistics presented in the research work.

Declarations

Conflict of interest

The authors have NO affiliations with or involvement in any organization or entity with any financial interest, or non-financial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) in the subject matter or materials discussed in this manuscript. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Ethics Approval

We declare that this manuscript is original, has not been published before and is not currently being considered for publication elsewhere.

Consent to Participate

We confirm that the manuscript has been read and approved by all named authors and that there are no other persons who satisfied the criteria for authorship but are not listed. We further confirm that the order of authors listed in the manuscript has been approved by all of us. We understand that the Corresponding Author is the sole contact for the Editorial process. He/she is responsible for communicating with the other authors about progress, submissions of revisions and final approval of proofs.

Consent for Publication

This manuscript has not been submitted to, nor is under review at, another journal or other publishing venue. We are ready to publish this original work with this journal.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1. Ahmed Elsmany, EF; Omar, MA; Wan, T-C; Altahir, AA. EESRA: Energy efficient scalable routing algorithm for wireless sensor networks. IEEE Access; 2019; 7, pp. 96974-96983. [DOI: https://dx.doi.org/10.1109/ACCESS.2019.2929578]

2. Daas, MS; Chikhi, S; Bourennane, E-B. A dynamic multi-sink routing protocol for static and mobile self-organizing wireless networks: A routing protocol for Internet of Things. Ad Hoc Networks; 2021; 117, 102495 pp. 1-16.

3. Khelifi, N; Nataf, E; Oteafy, S; Youssef, H. RESCUE-SINK: Dynamic sink augmentation for RPL in the Internet of Things. Transactions on Emerging Telecommunications Technologies; 2018; 29, 2 1. [DOI: https://dx.doi.org/10.1002/ett.3278]

4. Njoya, AN; Thron, C; Barry, J; Abdou, W; Tonye, E; Konje, NSL; Dipanda, A. Efficient scalable sensor node placement algorithm for fixed target coverage applications of wireless sensor networks. IET Wireless Sensor Systems; 2017; 7, 2 pp. 44-54. [DOI: https://dx.doi.org/10.1049/iet-wss.2016.0076]

5. Pakzad, SN; Fenves, GL; Kim, S; Culler, DE. Design and implementation of scalable wireless sensor network for structural monitoring. Journal of Infrastructure Systems; 2008; 1089, pp. 1-14.

6. Mirza, MA; Shakir, MZ; Alouini, M-S. A scalable global positioning system-free localization scheme for underwater wireless sensor networks. EURASIP Journal on Wireless Communications and Networking; 2013; 2013, pp. 1-10. [DOI: https://dx.doi.org/10.1186/1687-1499-2013-122]

7. Abutu, IM; Imeh, UJ; Abdoulie, TMS; Adewale, AE; Bashir, MM. Real time universal scalable wireless sensor network for environmental monitoring application. I. J. Computer Network and Information Security; 2018; 6, pp. 68-75. [DOI: https://dx.doi.org/10.5815/ijcnis.2018.06.07]

8. Kandris, D; Nakas, C; Vomvas, D; Koulouras, G. Applications of wireless sensor networks: An up-to-date survey. Applied System Innovation; 2020; 3, 14 1.

9. Ahmad, A; Rathore, MM; Paul, A; Chen, B-W. Data Transmission Scheme Using Mobile Sink in Static Wireless Sensor Network. Journal of Sensor; 2015; 2015, 1. [DOI: https://dx.doi.org/10.1155/2015/279304]

10. Randhawa, S., & Jain, S. (2017). Data aggregation in wireless sensor networks: Previous research, current status and future directions wireless personal communication. Springer.

11. Zhang, J; Hu, P; Xie, F; Long, J; He, A. An energy efficient and reliable in-network data aggregation scheme for WSN. IEEE Access; 2018; 6, pp. 71857-71870. [DOI: https://dx.doi.org/10.1109/ACCESS.2018.2882210]

12. Sasirekha, S; Swamynathan, S. A comparative study and analysis of data aggregation techniques in WSN. Indian Journal of Science and Technology; 2017; 8, 26 pp. 1-10.

13. Arumugam, GS; Ponnuchamy, T. EE-LEACH: Development of energy efficient LEACH protocol for data gathering in WSN. EURASIP Journal on Wireless Communications and Networking; 2015; 76, 1.

14. Khurana, B; P., & Kant, K. ,. LEACH-MAC: A new cluster head selection algorithm for wireless sensor networks. Journal of Wireless Networks; 2016; 22, pp. 49-60. [DOI: https://dx.doi.org/10.1007/s11276-015-0951-y]

15. Shah, T., Javaid, N., & Qureshi, T. N. (2012). Energy efficient sleep awake aware (EESAA) intelligent sensor network routing protocol. In 2012 15th International Multitopic Conference (INMIC), Islamabad (pp. 317–322).

16. Afsar, MM; Tayarani-N, MH. Clustering in sensor networks: A literature survey. Journal of Network and Computer Application; 2014; 46, pp. 198-226. [DOI: https://dx.doi.org/10.1016/j.jnca.2014.09.005]

17. Smaragdakis, G., Matta, I., Bestavros, A. (2004). SEP: A Stable Election Protocol for clustered heterogeneous wireless sensor networks. Second International Workshop, on Sensor and Actor Network Protocols and Applications (SANPA 2004).

18. Kaur, G; Rani, S; Kakkar, S. Design of an improved DEC protocol for wireless sensor networks. International Journal of Advanced Research in Electrical, Electronics and Instrumentation Engineering; 2016; 5, 4 1.

19. Lindsey, S; Raghavendra, CS. PEGASIS: Power efficient gathering in sensor information system. IEEE Aerospace Conference Proceedings; 2012; 3, pp. 1125-1130.

20. Marhoon, HA; Mahmuddin, M; Nor, SA. Chain based routing protocols in wireless sensor networks. ARPN Journal of Engineering and Applied Sciences; 2015; 10, 3 1.

21. Vaidya, R., & Dandekar, D. R. (2013). Comparison of SPAN and LEACH protocol for topology control in wireless sensor networks. IEEE International conference on signal processing, image processing & pattern recognition.

22. Kaur, S; Vashisht, R. Hybrid approach of data aggregation (HADA) based on iLEACH in WSNs. American Journal of Advanced Computing; 2014; I, 2 pp. 24-30.

23. Bin, H. E., Hongtao, Z. (2013). An Energy Optimization Method For Wireless Sensor Network” 27th International Conference On Advanced Information Networking And Applications Workshops (WAINA).

24. Parmar, B; Munjani, J; Meisuria, J; Singh, A. A Survey of routing protocol LEACH for WSN. International Journal of Scientific and Research Publications; 2014; 4, 1 1.

25. Chagas, S. H., Martins, J. B., & de Oliveira, L. L. (2012) An approach to localization scheme of wireless sensor networks based on artificial neural networks and Genetic Algorithms. IEEE 10th International conference on New Circuits and Systems Conference (NEWCAS).

26. Aderohunmu, F. A., & Deng, J. D. (2009) An Enhanced Stable Election Protocol (SEP) for Clustered Heterogeneous WSN. IEEE workshop available at research Gate.

27. Jagadeesh Naik, L; Ramanaiah, KV; Soundara Rajan, K. Performance evaluation of MCHSEP and SEP protocol for wireless sensor networks. International Journal of Recent Technology and Engineering (IJRTE); 2019; 7, 1.

28. Sharma, U; Tiwari, S. Performance analysis of SEP and LEACH for heterogeneous wireless sensor networks. International Journal of Computer Trends and Technology (IJCTT); 2014; 10, 4 1.

29. Singh, PK; Yadav, DK; Dixit, S. Modified stable election protocol (M-SEP) for wireless sensor network. International Journal of Advanced Research in Electronics and Communication Engineering (IJARECE); 2015; 4, 4 1.

30. Aderohunmu, FA; Deng, JD; Purvis, MK. A deterministic energy-efficient clustering protocol for wireless sensor networks. ISSNIP an IEEE International Symposium.; 2011; 1, pp. 341-346.

31. Qing, L; Zhu, Q; Wang, M. Design of a distributed energy-efficient clustering algorithm for heterogeneous wireless sensor networks. Computer Communications; 2006; 29, 22302237. [DOI: https://dx.doi.org/10.1016/j.comcom.2006.02.017]

32. Prasada Reddy, MM; Varada Rajan, S. DEEC protocol for WSNs. Advances in Wireless and Mobile Communications.; 2017; 10, 1 pp. 51-63.

33. Sharma, D; Tomar, GS. Energy efficient multitier random DEC routing protocols for WSN. Agricultural Wireless Personal Communications; 2021; 120, pp. 727-747. [DOI: https://dx.doi.org/10.1007/s11277-021-08486-0]

34. Javaid, N., Qureshi, T. N., Khan, A. H., & Iqbal, A., Akhtar, E., & Ishfaq, M. (2013). EDDEEC: Enhanced developed distributed energy-efficient clustering for heterogeneous wireless sensor networks. arXiv preprint arXiv:1303.5274.

35. Sharma, AK; Kourtney, H. Hybrid energy efficient distributed protocol for heterogeneous wireless sensor network. International Journal of Computer Applications; 2010; 4, 6 pp. 1-5. [DOI: https://dx.doi.org/10.5120/828-1173]

36. Redjimi, K; Redjim, M. The DEEC and EDEEC Heterogeneous WSN Routing Protocols. International Journal on Advanced Networking and Applications; 2022; 13, 4 pp. 5045-5051. [DOI: https://dx.doi.org/10.35444/IJANA.2022.13405]

37. Elbhiri, B., Rachid, S., Elfkihi, S., Aboutajdine, D. (2010). Developed Distributed Energy-Efficient clustering (DDEEC) for heterogeneous wireless sensor networks. In 5th International Symposium on I/V Communications and Mobile Networks (ISVC). ISBN 978-1-4244-5996-4.

38. Priya, R. (2018). EDEEC-enhanced distributed energy efficient clustering protocol for heterogeneous wireless sensor network (WSN) scheme for WSN. IEEE Access.

39. Zanjireh, M. M., & Larijani, H. (2015). A Survey on Centralised and Distributed Clustering Routing Algorithms for WSNs. 2015 IEEE 81st Vehicular Technology Conference (VTC Spring) (pp. 1–6).

40. Qureshi, TN; Javaid, N; Khan, AH; Iqbal, A; Akhtar, E; Ishfaq, M. BEENISH: Balanced energy efficient network integrated super heterogenous protocol for wireless sensor networks. Elsevier Procedia Computer Science; 2018; 1, 1.

41. Uniyal, N., Thakkar, V. M., & Bahuguna, A. (2017) Enhanced Energy Zonal Stable Election Protocol for Clustered Heterogeneous Wireless Sensor Network. IEEE 3rd International Conference on Advances in Computing, Communication & Automation (ICACCA) 2017.

42. Behera, TM; Mohapatra, SK; Samal, UC; Khan, MS; Daneshmand, M; Gandomi, AH. I-SEP: An Improved Routing Protocol for Heterogeneous WSN for IoT-Based Environmental Monitoring. IEEE Internet of Things Journal; 2020; 7, 1 pp. 710-717. [DOI: https://dx.doi.org/10.1109/JIOT.2019.2940988]

43. Jia, D; Zhu, H; Zou, S; Hu, P. Dynamic Cluster Head Selection Method for Wireless Sensor Network. IEEE Sensors Journal; 2016; 16, 8 1. [DOI: https://dx.doi.org/10.1109/JSEN.2015.2512322]

44. Karthik Reddy, G., & Nirmala Devi, L. (2018). A Review on Clustering Protocols with Energy heterogeneity in Wireless Sensor Networks. IEEE International Conference on Communication, Computing and Internet of Things (IC3IoT Nov 2018).

45. Bhola, J; Shabaz, M; Dhiman, G et al. Performance evaluation of multilayer clustering network using distributed energy efficient clustering with enhanced threshold protocol. Wireless Personal Communication; 2021; 1, pp. 1-15. [DOI: https://dx.doi.org/10.1007/s11277-021-08780-x]

46. Zhong, X., & Liang, Y. (2018). Scalable Downward Routing for Wireless Sensor Networks and Internet of Things Actuation. 2018 IEEE 43rd Conference on Local Computer Networks (LCN), Chicago, IL, USA (pp. 275–278).

47. El Khediri, S; Fakhet, W; Moulahid, T; Khand, R; Thaljaouie, A; Kachouric, A. Improved node localization using K-means clustering for Wireless Sensor Networks. Computer Science Review; 2020; 37, 1.4118835 [DOI: https://dx.doi.org/10.1016/j.cosrev.2020.100284]

48. Mishra, S. N., Elappila, M., & Chinara, S. (2018). Development of Survival Path Routing Protocol for Scalable Wireless Sensor Networks. In 2018 International Conference on Information Technology (ICIT), Bhubaneswar, India (pp. 210–215).

49. Sandhya, R., & Sengottaiyan, N. (2016) S-SEECH secured - Scalable Energy Efficient Clustering Hierarchy Protocol for Wireless Sensor Network. In 2016 International Conference on Data Mining and Advanced Computing (SAPIENCE), Ernakulam, 2016 (pp. 306–309).

50. Sah, DK; Amgoth, T. Parametric survey on cross-layer designs for wireless sensor networks. Computer Science Review; 2018; 27, pp. 112-134.3766561 [DOI: https://dx.doi.org/10.1016/j.cosrev.2017.12.002]

51. Singh, O., & Rishiwal, V. (2017). On the scalability of routing protocols in WSN. In 2017 3rd International Conference on Advances in Computing, Communication & Automation (ICACCA) (Fall), Dehradun (pp. 1–6).

52. Rani, P., & Sharma, A. (2021) Linear scalable routing protocol for wireless sensor network. IOP Conference Series: Materials Science and Engineering, 1057. 012094.

53. Shukla, A; Tripathi, S. A multi-tier based clustering framework for scalable and energy efficient WSN-assisted IoT network. Journal of Wireless Network Springer; 2020; 9, 1.

54. Yadav, A; Kumar, S. An Enhanced Distributed Energy-Efficient Clustering (DEEC) Protocol for Wireless Sensor Networks. International Journal of Future Generation Communication and Networking; 2016; 9, 11 pp. 49-58. [DOI: https://dx.doi.org/10.14257/ijfgcn.2016.9.11.05]

55. Maitra, T; Roy, S. A comparative study on popular MAC protocols for mixed Wireless Sensor Networks: From implementation viewpoint. Computer Science Review; 2016; 22, pp. 107-134.3579166 [DOI: https://dx.doi.org/10.1016/j.cosrev.2016.09.004]

56. Singh, S; Malik, A; Kumar, R. Energy efficient heterogeneous DEEC protocol for enhancing lifetime in WSNs. Engineering Science and Technology, an International Journal; 2017; 20, 1 pp. 345-353. [DOI: https://dx.doi.org/10.1016/j.jestch.2016.08.009]

57. Vançinand, S; Erdem, E. Threshold Balanced Sampled DEEC Model for Heterogeneous Wireless Sensor Networks. Wireless Communications and Mobile Computing; 2018; 2018, 1. [DOI: https://dx.doi.org/10.1155/2018/4618056]

58. Khan, M. Y., Javaid, N., Khan, M., Javaid, A., Khan, Z., Qasim, U. (2013). Hybrid DEEC: Towards Efficient Energy Utilization in Wireless Sensor Networks arXiv preprint arXiv: 1303.4679

59. Sheenam,. G-DEEC: Gateway based multi-hop distributed energy efficient clustering protocol for wireless sensor networks. International Journal on Cybernetics & Informatics (IJCI); 2015; 4, 5 1.

60. Huang, J; Zhao, Z; Yuan, Y et al. Multi-factor and distributed clustering routing protocol in wireless sensor networks. Wireless Personal Communication; 2017; 95, pp. 2127-2142. [DOI: https://dx.doi.org/10.1007/s11277-017-4045-2]

61. Gupta, SK; Singh, S. Survey on energy efficient dynamic sink optimum routing for wireless sensor network and communication. International Journal of Communication Systems; 2022; 35, 11 [DOI: https://dx.doi.org/10.1002/dac.5194] e5194.