Desktop version

Home arrow Computer Science

  • Increase font
  • Decrease font

<<   CONTENTS   >>

An Energy-Efficient Fuzzy Logic-Based Clustering with Data Aggregation Protocol for WSN-Assisted IoT System


The progressive development of IoT enhances the lifestyle and performance of a human being [1]. IoT is one of the innovative patterns of accumulating different objectives that is composed of two semantics. Initially, the base of IoT is a network that is assumed to be the enhancement of the Internet. Second, the end users have exchanged the data, which is referred to as connectivity over objects. Because of the sensing performance and wireless interaction ability, the usage of IoT provides extensive benefits. As it is a pervasive method, it can also apply in vehicle observation, medicare, urban transportation, space searching, and some other challenging applications [2]. The significant member of IoT is named as WSN, and it is treated as common environment for several fields like estimation of diverse ecological variables such as temperature, pressure, humidity, light, and so on.

This network is applied in massive domains like environmental observation, medical sector, electricity grid, surveillance, and persistent patient observation, respectively. Under the reduction of wiring expenses and activating new kinds of evaluation domains, the remote system enhances the wired structures. Thus, remote monitoring has numerous benefits in several other applications. It is one of the three system topologies gathered in WSN. Each hub interfaces specifically with a path in star topologies [3]. In case of collective network, the tree is organized in such a way that all hub interfaces with a hub maximum in a tree to door. A data has been steered from lower hub on a tree to portal. Consequently, it represented that one hub can interface with diverse hubs for providing upgraded unwavering quality in network.

The WSN is composed of different connection units that is additive with wireless network. A link network is changed in the network node, which develops various modules, for instance, battery, analog circuit, sensor interfaces radio, microcontroller, etc. [4]. In last decades, the Internet is replaced for simulating new ideas regarding IoT from interfacing individuals for connecting the objects. Hence, it has resulted in developing the novel pattern into the Internet and provides best applications and business [5, 6]. Figure 4.1 shows the layers in WSN-assisted IoT systems.

In WSN, the base station (BS) should generate the collected value for end users. Under the reduction of transmission load and significant application, the data collection is forwarded [7]. These hubs are applicable in minimum collections, which support data collection named as clusters. For data accumulation, clustering is assumed to be isolation of hubs [8-10]. Thus, clustering is mainly applied for enhancing the network lifetime. Some important measures are taken for determining the implementation of sensor nodes. In order to accomplish better elasticity of well-organized system, possible productivity is carried out [11, 12]. According to the borders enhancement of group, additional portions are allocated to hubs. The cluster

Layers in WSN-assisted IoT systems

FIGURE 4.1 Layers in WSN-assisted IoT systems.

head (CH) is termed as an incharge of management and sending data to BS present in group. By the replacement of hubs that are applicable for predicting and transmitting, the collected data to CH is termed as cluster member (CM) nodes [13-15].

The sensors are capable of producing massive data and contain various features like processing energy, storage, and communication potentials [16-18]. WSNs are said to be identical when the nodes are symmetric. A WSN that is nonidentical is said to be heterogeneous. Basically, the machines are battery-powered and data collection is highly effective and significant [19-21]. Clustering is a well-known and power-effective solutions developed by the research community for gathering information from the WSN. Consequently, a group of clusters are deployed. A cluster is composed of a member node and CH. It is helpful in gathering data from members called as intracluster communication. CHs support to address the centralized BS named as intercluster communication [23].

Feedback mechanism-based unequal clustering (FMUC) [24] is defined as the feedback approach that depends upon unequal heterogeneous protocol. FMUC is mainly applied for eliminating energy hole problem or hot spot issue, while the energy load is managed in application-based WSNs. First, FMUC separates the system as layers and processed systematically. An arithmetic method has been employed for making equal power application and overall initial energy of all layers. The cluster would be one among these layers. The cluster size can be determined by the amount of power utilization of a layer. Clusters forward the sizes as a feedback to BS where it publishes the gathered values to a network. Every nodes of WSN derives an opinion measure, although only CH modifies the processing requirements on the basis of obtained values.

An unequal HEED (UHEED) [25] is defined as uneven-sized cluster-dependent procedure to WSNs. UHEED applies a strategy of EEUC procedure to HEED for developing unequal size clusters. A cluster size of CH is based on a distance measure. On the other hand, a cluster that is placed distant from BS contains maximum range in terms of clusters closer to sink. UHEED minimizes the hot spot issue and enhances network duration than other models. Rotated UHEED (RUHEED) applies the unequal-sized cluster-dependent model that maximizes the hot spot problem and increases the network survival rate. RUHEED contains three phases: CH selection, clusters development, and CH rotation. HEED is mainly employed for selecting CHs, which depends upon the residual energy (RE) and processing expense. EEUC is relied on the distance from sensor node to BS, which has been utilized for developing unequal-sized clusters. Reclustering of network is carried out if sensor nodes exhaust the complete energy. RUHEED saves power and reduces the count of cluster selection as well as cluster development stages.

In Ref. [26], the diverse energy-yielding models are presented: (i) energy harvesting integrated with concurrent data decoding, which is referred to as a trade-off from energy used for future and the amount of energy that is consumed in signal decoding; (ii) energy-effective task of WSN exploits better routing models or lists the process of sensor nodes; (iii) mobile chargers terminates the best positions for performing charge; and (iv) energy distribution. The objectives (iii) and (iv) are selected in a two-step procedure integrating mobile charger that shifts within the system to discharged CH that alters an energy operating from overcharged CHs and alternate nodes. An energy operating is carried out for all clusters. A CH is selected from nodes with maximum number of neighbors within the inclusion circle. Initially, the mobile charger applies a better path to obtain consecutive discharged CH, and it is terminated and overcharges. The overcharged CHs give energy to cluster members (CM) without any competition, where the count of seller nodes is maximum when compared to count of buyer nodes. Reference [27] applies a cross-layer cooperative TDMA model instead of existing methods for optimizing the CHs transmitting function. A CH function is changed among the nodes with the help of duty cycling for individual energy harvesting abilities. It determines the best number of clusters on the basis of intensity of power source. The protocol depends upon LEACH. The CH option relies on possibility application, which employs duty cycle approach in which a node is not capable to become CH prior to passing duty cycles.

This chapter develops an energy-efficient fuzzy-based clustering and data aggregation protocol called FC-DR for WSN-assisted IoT system. The proposed FC-DR method operates on three major stages: FC-based node clustering, data collection, and error-bounded lossy compression (EBLC) technique-based data aggregation. The EBLC technique aggregates the data before sending it to the cloud server, which is treated as a maximum energy consumption stage among the IoT devices. The efficiency of the presented model has been examined under different aspects. The attained simulation results ensured the goodness of the FC-DR model in terms of several evaluation parameters.

<<   CONTENTS   >>

Related topics