IEEE TRANSACTIONS ON FUZZY SYSTEMS 1 A Two-Tier Distributed Fuzzy Logic Based Protocol for Efficient Data Aggregation in Multi-Hop Wireless Sensor Networks Seyyit Alper Sert, Member, IEEE, Abdullah Alchihabi, Student Member, IEEE, and Adnan Yazici, Senior Member, IEEE Abstract—This study proposes a two-tier distributed fuzzy logic based protocol in order to improve efficiency of data aggregation operations in multi-hop wireless sensor networks (WSNs). Clustering is utilized for efficient aggregation require- ments in terms of consumed energy. In a clustered network, member (leaf) nodes transmit obtained data to cluster-heads (CHs) and CHs relay received packets to the base station. In multi-hop wireless networks, this CH-generated transmission occurs over other CHs. Due to the adoption of a multi-hop topology, hotspots and/or energy-hole problems may arise. In this paper, we propose a Two-Tier Distributed Fuzzy Logic Based Protocol (TTDFP) to extend the lifespan of multi-hop WSNs by taking the efficiency of clustering and routing phases jointly into account. The proposed protocol, TTDFP, is a distribution- adaptive protocol that runs and scales efficiently for sensor network applications. Additionally, along with the two-tier fuzzy logic based protocol, we utilize an optimization framework to tune the parameters used in the fuzzy clustering tier in order to optimize the performance of a given WSN. This paper also includes performance comparisons and experimental evaluations with the selected state-of-the-art algorithms. The experimental results reveal that TTDFP performs better than any of the other protocols under the same network setup considering metrics used for comparing energy-efficiency and network lifespan of the protocols. Index Terms—fuzzy clustering, fuzzy routing, parameter opti- mization, wireless sensor networks. I. I NTRODUCTION T ECHNOLOGICAL developments have made the gen- eration and usage of wireless sensor nodes possible. Although an individual node is capable of gathering data alone, these nodes nearly always cooperate to extract high level se- mantic information from the sensed region. Networks of such nodes are named as Wireless Sensor Networks (WSNs). These nodes are powered with batteries which cannot be recharged easily in most cases. For this reason, energy requirements of nodes have attracted attention among researchers that drove the tendency to gear them with energy-harvesting techniques [1]. However, most of the nodes in market may not utilize S.A. Sert, A. Alchihabi and A. Yazici are with the Department of Computer Engineering, Middle East Technical University, Ankara, 06800 Turkey, e-mail: (see www.ceng.metu.edu.tr). A. Yazici is currently with the Nazarbayev University, Department of Computer Science at School of Science and Technology, Astana, Kazakhstan, e-mail: adnan.yazici@nu.edu.kz. This work is supported in part by a research Grant from TUBITAK with Grant No. 114R082 and in part by Social Policy Grants (SPG) from Nazarbayev University, Astana, Kazakhstan. Manuscript received August 23, 2017; revised March 4, 2018. this technology because of its current cost or the intended usage area, which hinders the implementation of such available methods. For this reason, derogating consumed energy through energy-efficient methodologies is among major goals [2] and contriving an energy-efficient algorithm is still of great impor- tance to extend the lifespan of the operating network. Lifespan of the network is especially crucial in mission critical systems such as outdoor environment monitoring, military surveillance, and disaster area monitoring. In a network, nodes may be partitioned into clusters. Each cluster consists of a single cluster-head (CH) and one or more member nodes. CHs are also called as leaders since they organize data gathering from their member nodes then transfer produced data to intended locations. Clustering in WSNs en- sures great performance requirements [3][4] and also increases the scalability of the network [5]. Bandwidth conservation, preservation of energy, and topology stabilization are other clustering advantages [6]. In clustering, efficiently selecting these leaders can signifi- cantly decrease consumed energy. For this reason, there is an ongoing thorough study concerning selection mechanisms in the literature. The common point of the proposed solutions in literature is the utilization of a two-phase process: in the first phase a CH with more energy is selected, and then in the second phase leadership is transferred among member nodes with the aim of balancing consumed energy. This common point actually retains two crucial hidden pieces of knowledge: a) The first hidden knowledge: Some of these method- ologies like [7] and [8], which we discuss in detail in the next section, lack of other necessary relevant information such as the location or connectivity of a node that can be utilized as input parameters in the clustering process or possess a central- ized operation architecture. For this reason, the final clusters cannot meet the expected efficiency demand. Additionally, because of not taking the position of each node into account, the hotspots or the energy hole problems may arise in WSNs. The hotspots problem is related with the premature death of the leaders that are around the base station or on busy routes because of intense inter-cluster relay. Similarly, the energy hole problem is related with the early energy depletion of some close nodes that are located in an area which degrades or sometimes completely impedes the transmission of the relayed data to the intended location. Besides, the energy hole problem may also occur in evolving networks since initial deployment location of nodes may change drastically. This variable node This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/TFUZZ.2018.2841369 Copyright (c) 2018 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.