[1]
The wireless sensor networks need to be energy efficient in order to improve the lifetime of the sensor nodes involved in the network as well minimizing the redundant packets transmitted to the sink. To ensure such a situation, there is a need to ensure that only those sensors which are in proximity to the source of useful and desirable information (also termed as event of information) should be kept active in the sensor network. In this paper, the authors proposed a novel enhanced energy efficient leach algorithm based on energy and distance which can be used to ensure that useful packets get transmitted to the sink and simultaneously minimizing power consumption of network to ensure longer lifetime of the sensor nodes in the network.
In Wireless Sensor networks, the power resource of every sensor node is limited. There is a serious requirement to develop smart sensor networks which are capable of not only detecting useful information, but also ensuring that the power consumption of the network is minimized without compromising on the degree of accuracy of the wireless network and less loss of useful information. In LEACH algorithm, the network is divided into several fixed size clusters. All these fixed size clusters form a cluster head to which all the nodes within the same cluster transmit data and CH transmits entire data to Sink. Nodes that have been cluster heads cannot become cluster heads again for P fixed rounds, where P is the desired percentage of cluster heads. Thereafter, each node has a 1/P probability of becoming a cluster head again. At the end of each round, each node that is not a cluster head selects the closest cluster head and joins that cluster. The cluster head, then creates a schedule for each node in its cluster to transmit its data and forwards the collective data received from all the sensor nodes within the cluster to the sink. Data of all the clusters transmitted by all the cluster heads to the sink gets aggregated at the sink which processes the information.
The remained of this paper is organized as follows. In Section 1, the authors introduce the existing literature along with the improvements of proposed Algorithm. In Section 2, a brief overview about the problem statement and in Section 3, a brief overview of the LEACH algorithm along with the protocols and the assumptions in WSN is presented. In Section 4, a step-wise procedure for the proposed algorithm is presented. In Section 5, simulation results are shown, and the final section includes the conclusion part.
Minimization of power consumption can be carried out using Low Energy Adaptive Clustering Hierarchy (LEACH) Algorithm [1]. The LEACH Algorithm was further improved by accommodating Multi-Hop Routing [2], Variable Round Time [3], Balanced LEACH to modify the number of cluster heads after each round LEACH-B [4] and utilizing the slots of the nodes which don't have data to send at its scheduled time slot [12]. Two new protocols, multi-level leach and DD- TL-Leach protocol have been presented in [13] and WSN is developed for machine monitoring and data processing [6]. After LEACH-B, another algorithm LEACH-R [5] was proposed to further improve the selection of cluster head by considering residual energy during the selection of cluster-head, thus reducing the possibility of low-energy nodes being selected as a cluster-head. HEED (Hybrid- Energy Efficient Clustering algorithm) was proposed in [7] in which cluster head selection is based on residual energy and node degree. A modified version of LEACH was developed for water regime monitoring system in [8]. A new algorithm UDACH is proposed in [9], which involves a three stage protocol and each node is assigned its cluster head based on the cluster head tree, where the tree is created based on base station distance from the node. A master cluster head approach is introduced in [10] along with cluster head, which also tries to minimize communication distance between nodes. An analysis of LEACH protocols is presented in [11]. The authors propose a further modification of these algorithms by considering that portion of the scenario to introspect, where large amount of useful information is available at the information event.
The sensors at a fixed radius from these events are considered active and form a cluster and select the node within the cluster with maximum residual energy as the cluster head. The position of the cluster is variable and changes with the location of the information event. The remaining sensor nodes away from the event are kept passive, i.e. in sleep mode till the position of the event does not change to a location near that sensor node. The main objective of this paper is to develop an algorithm based on event location, which not only decreases the energy consumption of the sensor nodes, but also eliminates redundant packets sent to the sink resulting in less memory usage and increases energy efficiency of the network.
The Sensor Nodes in a WSN have to be deployed randomly with high density so that, information is not lost and each sensor node transmitting information to sink via Cluster Head. The problem with this algorithm is that, a lot of redundant data gets transmitted to the Sink as most of sensor nodes don't have any useful information at a particular instant of time, which not only needs a lot of memory but also increases the energy consumption of network and decreasing the lifetime of the Sensors.
The authors developed an algorithm based on event location. The proposed method not only ensures that, the cluster head is the node with maximum energy within the cluster, but also ensure that only those nodes are active which are placed near the event of desirable information. Thus, nodes in the wireless sensor network capture the most crucial information and save energy by not capturing data of less significance, which not only decreases the number of packets sent to the sink, but also makes sure that the whole network is energy efficient resulting in an increase in lifetime of individual nodes.
The authors start their assumption by taking the identical radio model as proposed by W.R. Heinzelman [1], which is the first order radio model. The radios can control their power to enhance the least power required to reach the anticipated recipients. The radios can also be turned off to avoid receiving undesired transmissions. An r2 energy loss occurs due to channel transmission, 'r' refers to the path-loss exponent, usually (2<=r<=4) depending upon the propagation model used and in this simulation, r=2 is utilized. Hence, considering the radio transmission problem, the number of transmissions and receptions ought to be as low as possible, since it is a very energy consuming process.
It has also been assumed that, all the nodes have the same energy at the starting of observation, i.e. all the sensors are equivalent at the beginning of the observation of the process and all the sensors have an equal probability of becoming the Cluster Head (CH) on the basis of energy model at the starting of the observation. The base station is static and is positioned distant from the sensor nodes. Each node periodically senses its neighboring environment and is keen to send its data to base station.
In the LEACH algorithm [1], each round consists of two phases- The initialization phase and the stability phase. In order to avoid the extra processing cost, the stabilization phase lasting a relatively long time. In the initialization phase, a cluster head election process occurs in the wireless sensor network. The election of cluster head involves the following procedure. The selection of CH depends on the decisions made by the node by choosing a random number between 0 and 1. If the number is less than a particular threshold value, the node becomes a cluster-head for the current round. The following equation is utilized to set the threshold,
In this equation, P refers to the desired fraction of cluster heads, r refers to the current round, and G is the set of nodes that have not been cluster-heads in the last 1/P rounds. Using this threshold, each node will be a cluster-head at some point within 1/P rounds. Nodes that have been cluster heads cannot become cluster heads for a second time for P rounds. After that, each node has a P probability of becoming a cluster head in every round. At the end of every round, each node that is not a cluster head select the nearest cluster head and joins that cluster to transmit data. The cluster heads combine and compress the data and forward it to the base station, therefore it extends the life span of major nodes.
After the node is selected as cluster head, it will broadcast the information about its selection as CH to the rest of the nodes in the same cluster. The remaining nodes decide to join the cluster according to the size of the received signal, and return the join signal. When the cluster head receives all join messages, it will allocate TDMA (Time Division Multiple Access) time slot information to all the nodes in the same cluster. Thus the nodes within the same cluster send a TDMA message to the cluster head in its own time slot. In order to avoid signal interference near the cluster, cluster head can determine the CDMA codes, which all nodes used. The CDMA (Code Division Multiple Access) codes which are used in the current phase and TDMA timing information will be sent together. When nodes within the cluster receive the message, they will send data to the cluster head in their own time slot. The Algorithm will enter a Stable phase.
In the stable phase of LEACH algorithm, member nodes continuously monitor the collected data, and send data to the cluster head node in their own time slots. While the other time, it can turn off the radio module, into hibernation, and it is one of the main ways to save energy for LEACH. After the cluster head node receives the data from its member, it will do the necessary processing of data fusion. Then the information is sent to the sink node. After a period of time of data transmission, nodes enter into a new work round, to re- select a new cluster head, and constantly circulating.
LEACH algorithm ensures that, each node does not have to individually transfer the sensed information to the sink. It just needs to transmit the data to the cluster head. The cluster head can then transmit collective data corresponding to all the nodes within the cluster to the sink thus saving transmission energy. Since the cluster heads in LEACH protocol are randomly generated, energy consumption can be evenly distributed in the network; however, it ignores residual energy of nodes, geographic location and other information in the election of cluster head node. So it can easily lead to exhaust the energy quickly in cluster head nodes.
Moreover, the LEACH algorithm does not ensure that the nodes sense the maximum useful and relevant information and saves energy by not transmitting information, which is of less importance. In order to overcome such limitations, the authors have carried out certain modifications in the LEACH algorithm. They have introduced the concept of events in their proposed work. Events are those locations in the scenario, which the user wishes to sense and analyze in a particular round. Locations of events can be varied after each round, because the user might be interested in carrying out analysis of different portions of the scenario at different intervals of time. In the proposed extension of LEACH algorithm, after the location of events for the particular round are selected, the nodes which are at a located at a distance (R) and less from the event form a cluster. Thus the number of clusters is equal to the number of events in the scenario. The remaining node has the ability to capture less resourceful information for that round and remain in dormant mode for that particular round, thus saving on the total remaining energy of the sensor network. Moreover, in the proposed algorithm, certain modifications have been proposed for the selection of the cluster head. In the cluster head election procedure corresponding to the proposed work, the nodes within the cluster communicate to elect the node with maximum remaining energy as the cluster head and at a minimum distance from the Sink. This modification to the LEACH algorithm not only ensures lower energy consumption and higher network lifetime, but also allows maximum useful information to be transferred to the sink by filtering out redundant information, which is not required by the user for analysis at that particular interval of time.
1) Store the information related to the network, including the position of the nodes, location of the sink, dimensions of the scenario etc.
2) Round= 1.
3) The user defines the locations of the events (regions where sensing action is required for that round).
4) For every node follow step 5.
5) If the location of the node is within a fixed radius (R) from an event, classify it into the cluster corresponding to the event.
6) Select the node in each cluster with maximum remaining energy within the cluster as the cluster head and minimum distance from the sink, since the power dissipation of cluster head of the sensor network is maximized.
7) For every cluster, follow step 8 and step 9
8) Nodes within the cluster send the sensed information to cluster head. Nodes not group under any cluster, save energy by operating in sleep mode.
9) Cluster heads collect the information from all nodes within the cluster and forward's it to the sink.
10) Round= Round+1.
11) If round <= maximum number of rounds, then proceed to Step 3.
12) Else go to the next step.
13) End.
Figure 1 shows the proposed algorithm and Figure 2 shows the flowchart of the proposed algorithm with all relevent steps.
Figure 1. Proposed Algorithm
Figure 2. Flowchart of the Proposed Algorithm
The results of the existing and the proposed algorithm have been compared and carried out in MATLAB corresponding to a scenario with around 100 nodes with a random deployment and a fixed sink in the Scenario with coordinates (50,178). On comparing the Scenario corresponding to LEACH Algorithm (Figure 3 and the proposed Algorithm (Figure 4), the authors observe that the proposed algorithm also includes event (marked red in Figure 4) along with the nodes (marked blue in Figure 4) and the sink (marked green in Figure 4). Note the Scenario shown in Figure 4 is adaptive in nature and gets modified in each round. However the scenario originally used in LEACH consisted of nodes (marked blue in Figure 3) and the sink only (marked green in Figure 3). It did not include the concept of the event. The performance of the LEACH algorithm is compared with the proposed modification for the first 400 rounds and the response of the both the networks is compared. After 400 rounds, LEACH algorithm involves more than 10000 information packets being sent. The proposed modification filters out redundant information and only sends the required desirable information as shown in Figure 5. Thus around 400 useful packets are sent thereby, decreasing the traffic and reducing the increased memory requirement. In LEACH algorithm, 40 nodes die out due to insufficient residual energy after 400 rounds. However, in the proposed modification, only 4 nodes die out after 400 rounds depicted in Figure 7 which shows a higher lifetime of sensor nodes in our algorithm. In LEACH algorithm, sum of remaining energy reduces from 50 units to around 12 units after 400 rounds. However, in the proposed modification, it is reduced to 47.8 units after 400 rounds as shown in Figure 6 which indicates better energy efficiency in the proposed network.
Figure 3. The scenario of the Existing Algorithm with Sensors marked with ‘blue’ color and Sink marked with ‘green’ the location of each Sensor is Random with X and Y Coordinate Specified in the Scenario
Figure 4. The Scenario of the Existing Algorithm with Sensors marked with ‘blue’ color and Sink marked with ‘green’ and the Agent is marked with red color the Location of Each
Figure 5. Number of Packets Sent to the Sink in the Proposed Algorithm Vs Rounds shown that Redundant Packets and Eliminated in Proposed Algorithm
Figure 6. Proposed Algorithm is Energy Efficient if Total Remaining Energy of nodes is compared in both Algorithm.
Figure 7. Comparison of Number of Dead Nodes Vs Rounds for Proposed and Existing Algorithm
In this paper, modification of the WSN Low Energy Adaptive Clustering Algorithm has been conducted by carrying out clustering based on the location of event and assigning node with maximum residual energy and at a minimum distance from the sink as the cluster head and results have been compared. A drastic improvement in the network lifetime because of less energy consumption by each sensor has been observed, while ensuring transmission of desirable and resourceful data packets only. It only ensures a greater lifetime of the sensors, which is a crucial aspect in the Wireless Sensor Network. In future work, the authors would like to work on Energy Harvesting Wireless Sensor Network which incorporates evolutionary algorithms for design of event based clustering, which will further increase he efficiency of the network and also the lifetime of the sensors.