A Study Of Energy Efficient Routing Protocols

Published Date: 02 Nov 2017

Abstract

With the improvement of networking field, wireless sensor network application is on the rise. Energy usage is the main issue of any protocols used in Wireless sensor network, because sensors are self powered. If one node fails in a network its might mislead the whole application. Since energy consumption is a big concern for any modern technology, it became a great concern for routing protocols designs. There are various routing schemes like query based routing, cluster based routing, in which best possible routing can be achieved in perspective of energy. Some routing protocols of these schemas can be defined in a new category "Data Aggregation". With the help of some routing protocols significant amount of power can be save from the sensors. These routing protocols also help to increase in the lifetimes of any sensor network. The main focus of this paper was surveyed on some energy-aware routing protocols in sensor network based on their category and also discussed best routing protocol based on their attribute.

Keywords: Wireless sensor network, energy-efficient, hierarchal, cluster, routing protocols.

1. Introduction

Wireless sensor network is one of the most important technologies for this era. In the past few decades it’s received a huge concentration from the both academic in industry aspects. According to Business Week, WSN is considered as one of the most important technologies for the 21st century [2]. Sensor nodes of sensor network can provide remarkable collective help. They can help to avoid calamitous transporter malfunction, preserve valuable natural property, boost up the output, improve security, and facilitate new applications such as context-aware systems and smart home technologies. Recent improvement in MEMS (Micro Electronic Mechanical Systems) and wireless communication technologies have allowed the development of tiny, low-cost, low-power and multifunctional smart nodes in a wireless sensor network [1]. Military applications motivated improvement in this technology. Though it’s improved for military applications, still its exercise is not limited to military application. Focus on implementing sensor network (e.g. Vibration sensors) is diverted to more public infrastructure (bridges and tunnel) maintenance. Apart from military applications there are many applications of WSN, such as: environment monitoring, health monitoring, traffic control, security surveillance, home intelligence. Wireless sensor network is most widely researched topic in the communication field over the last few decades.

2. Objectives

The main aim of this research paper is to study some energy-efficient routing protocols proposed by researchers and discuss about their merits and demerits. This paper will contain few sections, in which, the first part will be a brief taxonomy of the sensor network. The next section will consist, discussion of different energy-aware routing protocol scheme in WSN based on their category. In the discussion section, I will discuss the difference and drawback of different routing protocols.

3. Wireless Sensor Networks

A sensor network is an infrastructure comprised of sensing, computing, and communication elements that give an administrator the ability to instrument, observe, and react to events and phenomena in a specified environment [3]. On the other hand, "A wireless sensor network consists of nodes organized into a cooperative network" [4]. In a sensor networked environment nodes are used to observe physical or environmental conditions, for example temperature, sound, pressure etc. These days it’s also does the control of sensing. In addition to sensing, one is often also interested in control and activation [3]. In short, "While many sensors connect to controllers and processing stations directly (e.g., Using local area networks), an increasing numbers of sensors communicate the collected data wirelessly to a centralized processing station" [20]. In a sensor network, a network may consist of few connected nodes (sensors) to thousands. Sensors perform the sensing task in WSN. It’s also known as node in WSN terminology.

A sensor node is mainly made of several parts like battery, radio transceiver, antenna and a microcontroller. The size of a sensor node is differing from normal box size to the size of a single grain of rice.

Figure: 1.1 some typical sensors are in use:

Smart Dust, TelosB, Tmote Sky.

4. Energy-aware Routing protocols in Wireless Sensor Networks

In WSN, there so many limitations of applying traditional routing protocols. For the reason of no infrastructure, vulnerable links, sensor nodes may fail and high demanding of energy saving requirements routing protocols should have a solution for these problems. Routing protocols in WSN are divided into different categories according to their type, but here I will discuss five categories out of them. I will study a few of them in this article. In Table 1.1, some major routing protocols are mentioned by their specification.

Categories

Protocols

Location-based Protocols

MECN, SMECN, GAF, GEAR

Data Centric Protocols

Directed Diffusion, SPIN, EAD

Hierarchal Protocols

LEACH, HEED, PEGASIS, TEEN, APTEEN

Mobility-Based Protocols

SEAD

QoS-Based Protocols

Energy-Aware QoS Routing Protocol

Table 1.1: Routing Protocols in WSN

5. Location-Based Protocols

Location based routing is also known as position-based routing protocol. It was first introduced by FINN [7] as "Cartesian routing".

MECN (Minimum Energy communication Network): - MECN is a location-based protocol for achieving minimum energy for randomly deployed ad hoc networks, which attempts to set up and maintain a minimum energy network with mobile sensors [6]. This theory is planned to use with low power GPS. The main concept behind MECN is which requires less power for transmission between any two particular nodes. It also adapts failed node and newly available nodes dynamically. This protocol is mainly designed for mobile network, but it works equally well for WSN [8]. This protocol was first proposed by Rodoplu and Meng, which computes an energy-efficient subnetwork [6] [22]. MECN maintains network connectivity in spite of sensor mobility. Its computes an optimal spanning tree rooted in the sink, known as minimum power topology. It contains the minimum power paths to sink from each sensor. This idea is mainly for plane and consists of main two segments, specifically enclosure graph construction and cost distribution. For stable network, in enclosure graph construction, MECN creates a meagre graph known as enclosure graph. This is based on instant neighbourhood of sensors. In this idea, a relay region is determined by the sensor, so that it can communicate with each sensor directly. This process makes a region known as enclosure region. In cost distribution, non-optimal links of the enclosure graph are basically omitted and the consequential graph is minimum power topology.

SMECN (Small Minimum Energy Communication Network): - SMECN proposed to improve MECN, in which a minimal graph is characterized with regard to the minimal power energy [22]. SMECN creates a sub graph of the sensor network that contains the minimum energy path [21]. It also adapts failed node dynamically. According to Li Li et al, SMECN has lower link maintenance cost than MECN and can achieve a significant saving in energy usage [22]. In this protocol, sensors discover their immediate neighbours by broadcasting a neighbour discovery message using some preliminary power that is updated incrementally. In details, a sensor starts broadcasting a neighbour discovery message with a number of preliminary power p and checks whether the conjectural set of immediate neighbours is a subset of the set of sensors that replied to that neighbour discovery message. If this is the case, the sensor will use the resulting power up to correspond with its immediate neighbours. Otherwise, it increments p and rebroadcasts its neighbour discovery message.

6. Data Centric Protocols

SPIN (Sensor Protocol for Information via Negotiation): - SPIN is one of the first data-centric protocols [9]. This protocol is developed to improve common flooding protocols. There is another main reason behind this is to overcome the problems caused by overlap and implosion. In a SPIN protocol environment, sensors are able to calculate energy consumption needed to send and receive data over the network. The SPIN protocols are based on two key mechanisms namely negotiation and resource adaptation [10]. To avoid repetitive and non-useful data in the network, SPIN enables a mechanism in sensors to negotiate with each other. There are four protocols in SPIN – SPIN-1or SPIN-PP, SPIN-2 or SPIN-EC, SPIN-BC and SPIN-RL [11] [12]. To reduce sensor consumption SPIN-1 uses a negotiation mechanism, on the other hand SPIN-2 use resource-aware mechanism for energy savings. SPIN-PP and SPIN-EC is mainly optimised in point-to-point network. For broadcast network, SPIN-BC and SPIN-RL are optimised. SPIN-BC improves SPIN-PP by using one-to-many Communication instead of many one-to-one communications [10].

Directed Diffusion: - Directed diffusion consists of elements such as: interests, data messages, gradients and reinforcements [14]. The main requirements like energy efficiency, scalability, and robustness of WSN meets by Directed Diffusion [10]. In this protocol data broadcasts interest message at a regular interval. This is known as interest message on the entire sensor network. For the reason of interest cache every sensor node aware of in which required message they can reply. A gradient is like a reply to originate interest message, from where it was received. A sink helps to establish multiple paths during the gradient process. These multiple paths used by sink for high quality events by increasing its data rate. This is done by a process known as reinforcement.

7. Hierarchal Protocols

LEACH (Low-Energy Adaptive Clustering Hierarchy): - This power efficient protocol is the most popular protocol on sensor network. LEACH protocol presumes to be a dense sensor network of homogenous, energy-constrained nodes, which shall report their data to a sink node [15]. As we know LEACH is a cluster based network, so during communication only communicating clusters responds with data. Because of that it saves a good amount of power. LEACH is absolutely circulated and awareness of the global network is not required. It decreases the energy consumption by minimizing the communication cost between sensors and their cluster heads and its turns off non-head nodes as much as possible [23].

LEACH achieves over a factor of 7 reduction in energy dissipation compared to direct communication and a factor of 4-8 compared to the minimum transmission energy routing protocol. The nodes die randomly and dynamic clustering increases lifetime of the system [16]. As we know, LEACH is a cluster based routing protocol. Based on duration clustering task is rotates among the nodes. To ensure proper meaningful data from sensors, LEACH uses a technique called aggregation [10]. This technique merges the important data into small chunk and carries into every sensor.

PEGASIS (Power-Efficient Gathering in Sensor Information Systems):- PEGASIS is an extension of LEACH, which creates a chain of sensor nodes so that each node transmits and receives from neighbour node and also one node is selected from that chain to transmit data to base station [17]. PEGASIS allows only one cluster head to transmit to the sink in each round [17]. In this protocol, sensors are organized to form a chain. This formation can be complete with a sink or by the sensor PEGASIS is able to increase the lifetime of the network as much the lifetime of the network under LEACH protocol [10].

TEEN (Threshold Sensitive Energy Efficient Sensor Network Protocol): - TEEN is a protocol designed for conditions like sudden changes in sensed attributes [18]. In this protocol sensor nodes sense environment continuously. The main characteristics of TEEN is, its sends a hard threshold and soft threshold value to cluster heads (CH). TEEN is a clustering communication protocol that targets a reactive network and enables CHs to impose a constraint on when the sensor should report their sensed data [19].

Mobility-Based Protocols

SEAD (Scalable Energy-Efficient Asynchronous Dissemination): - SEAD is a distributed self-organizing protocol, was planned to operate between minimizing the forwarding delay to a mobile sink and energy savings [24]. In SEAD a source sensor reports its sensed data to multiple mobile sink. SEAD protocol consists of three major elements namely dissemination tree (d-tree) construction, data dissemination and maintaining a relationship to mobile sinks.

In SEAD, every mobile sink is linked with one of their neighbours, which is known as access node. On behalf of a mobile sink Access node sends a link request to the d-tree source. For that reason, sensed data in source received from the access point and sends them to mobile sink. In this framework, there are four key segments. These segments are (i) subscription query (ii) gate replica search (iii) replica placement (iv) d-tree management. In the first segment, access node is selected from nearest neighbour by a mobile sink and these access nodes generate a join query to a source node. In the second segment, a gate replica is determined so that it can act as an embedded node on the data dissemination tree. In the next segment d-tree is newly adjusted in the neighbourhood of gate replica. This process is mainly to save more communication energy. The fourth and final segment is d-tree management, which is to sustain connectivity between the mobile sink and their access nodes.

QoS-Based Protocols

Energy-Aware QoS Routing Protocol: - This protocol is proposed by Akkya and Younis [26], where real time traffic is produced by imaging sensors. This proposed protocol expands the routing approach in [27] and come across with low cost and energy-efficient path. This path meets end-to-end delay during connection. This protocol uses Dijkstra’s algorithm to make a list of low cost path and choose a path from that list which meets the requirements of end-to-end delay.

Discussion

Wireless sensor network is one of the most researched fields in communication sector. This network always had to face a different kind of challenges over the year. In this paper, I concentrated on energy-aware routing protocols which are widely researched by the researchers. Based on the research I’ve done data routing in WSN can be categorise into three main categories – location-based, data-centric and hierarchal. But some protocols can be categorized into more than one category. Since this paper mainly focuses on energy efficiency of routing protocols of WSN, so many routing protocols are skipped mentioning.

Hierarchal protocols are also known as a cluster based protocol. In this cluster based approach, consumes less energy because of its cluster mechanism. In this approach sensor nodes are divided into cluster and a node gets the role of cluster head. During the operational period all the focus remains with cluster head. As a result cluster head consumes more energy compare to other nodes in the network. Among the mentioned protocols LEACH is most popular in WSN. It’s really a good approach for dense network. Sometimes LEACH is not beneficial for small network. The reason behind that, the cost of the cluster based network can be made up in dense network, it doesn’t make up in small networks. Still Hierarchal protocols (like LEACH, PEGASIS) are good to create a sensor network.

In data centric approach, sink sends queries and waits for the data from the destined queried region. Since this approach uses queries to request data, attribute based naming is required to specify data properties. Sometimes complex queries can’t be handled by this approach. This complex query depends on the application. In data centric approach, naming is also a matter of a huge concentration. An efficient way of standard naming is can be future research direction of data centric approach.

On location-based protocols mainly focused on deployment of sensor nodes and location information of sensor nodes is exploiting. To accommodate energy efficient routing location based protocols needs intelligent exploitation of local information. If this utilization fails then energy consumption in routing will increase. This process is one of the future researches for this kind of protocols.

One of the interesting issues for routing protocols of WSN is that, nodes and sinks in sensor network are considered as constant. It might be a problem in situations where mobility is important. In some situation constant update from all the sensor nodes is needed with mobility. But these stable nodes may not able provide the feature called mobility. This can be a future research field for WSN.

During this research I found that some energy efficient protocols can be classified as fusion routing protocol. Some of them can be fixed in more than one sub-category. Another interesting fact is that sensor node generates a significant amount of redundant data, similar packets from multiple nodes can be aggregated. As a result of that number of transmissions is reduced. The data aggregation technique is used to achieve energy efficiency and data transfer optimization in a number of routing protocols. This table also shows that which protocols using data aggregation. The below table will show this classification

.

Routing Protocol

Data-centric

Location-based

Hierarchal

Data Aggregation

SPIN

Directed Diffusion

LEACH

TEEN

PEGASIS

Table 1.2: New classification of routing protocols in sensor network.

Conclusion

Most of the sensor node of a sensor network is working with battery powered energy as a power source. The nodes are really tiny. But still this is the main area where research is like pausing. That’s why this holds narrow gap in utilization of battery power nodes. Despite the fact, efficient energy utilization is the main issue of the sensor network. This field is really wide and huge to research. In this paper, I mainly discussed and explained different energy-efficient routing protocols based on their category.

Due to the shortage energy resources for sensor, energy efficiency is the main confront of designing routing protocols in WSN. Consequently the efficient energy utilization is a key factor in the sensor network. As this is a huge field, in this paper I only cover up only a few of them. These protocols have their own drawbacks. Still topology holds the key of choosing routing protocols and plan.