A Novel DV-HOP and APIT Localization Algorithm with BAT-SA Algorithm

Abstract

Localization technology is essential for making wireless sensor networks(WSN)’s information processing and information collecting applications actually feasible. The beacon information is made available to the unknown nodes using the route exchange protocol. These data are more useful for determining the coordinates of neighboring nodes. Consequently, it was discovered that the algorithm for localizing nodes always has a flaw. Consequently, a brand-new metaheuristic termed Bat with simulated annealing is proposed to fix the flaw in the WSN standard node localization technique. The overall effectiveness of identifying the nodes is enhanced as a result of the large reduction in localization errors. The most popular localization estimation methods are the distance vector hop (DV-Hop) technique and approximate point-in-triangulation (APIT), which have high node localization accuracy and simple deployment in real-time environments. The primary benefits and their disadvantages, which give it a slight disadvantage in preference, are presented in this work. Both strategies are compared for their conventional performance and efficiency when combined with the Bat-SA algorithm.

1. Introduction

WSN has many uses because it is one of the most efficient and dependable techniques for gathering and processing data. In order to study and monitor environmental changes in locations where people are unable to travel, aircraft are used to deploy the wireless sensor nodes. The WSN nodes collect environmental data and are typically randomly dispersed around the monitoring region. It is found that in these circumstances, the nodes’ deployment position is significant. Due to limitations on cost, power, and stability, it would be very difficult to manually install a GPS receiving device for every sensor node [1].

The effectiveness of the WSN as a whole is significantly impacted by the absence of location data in such environments. A WSN is thus strongly dependent on localization technologies. The nodes convey the data they have collected to the destination node or monitor node across a number of hops. Since the WSN is self-organizing, each node may recognize or learn details about its surrounding nodes independently. It is found that the nodes of the wireless network have greater power than conventional nodes in a communication link. Data derived from nodes are transmitted to the base station using a multi-hop method [2,3,4,5]. Unknown nodes and beacon nodes are the two main categories of nodes in a WSN. In general, the unknown nodes are unaware of their location, as opposed to beacon nodes, which are aware of their whereabouts thanks to the GPS device that is deployed with them. The two types of localization algorithms are as follows:

• Distance-free;
• Distance-based.

Received signal strength indicator (RSSI), the angle of arrivals, time difference of arrivals, and the time of arrival are a few of the distance-based algorithms. Algorithms for localization without regard to distance or range include MAP, approximate triangle point test, amorphous algorithms, and DV-HOP. The perfect distance between the deployed node is identified by the distance-based localization algorithm in order to locate the closest nodes in the network. Therefore, location awareness is essential in highly sensitive applications like military and underwater communication, among others [6].

The availability of the geographic locations of the sensor nodes is thus discovered to be complicated by localization. Triangulation, multilateration, or trilateration are the geometric methodologies used for position determination depending on the characteristic of the communication signals. Although distance-based localization approaches have a high degree of precision, they are not economical [7] and shown in Figure 1.

2. DV-HOP-Based Methodology

The main notion underlying the DV-HOP method is that unknown nodes must first ascertain the beacon node’s maximum hop count. After multiplying the ratio of the average hop to the necessary minimum hops, the hop average distance has been determined. As the last step, when computingthe distance between the unknown node and beacon node, the existing DV-HOP algorithms have a number of problems, some of which are listed below. Beacon localization is less accurate. Lower coverage rate in sparse environments. Figure 2 describes the design of nodes in a WSN with the conventional DV-Hop approach [8,9,10]. At the start, every anchor node broadcasts a signal to the entire WSN. The anchor node is identified by initializing one as count value for the HOP. The minimum count value isconsidered by the beacon. Astale node is detected as a beacon with a higher hop count value than a specific anchor node, and it is disregarded as a result [11,12,13].

The network floods the new nodes that it does not identify as stale nodes. At each intermediate hop, these nodes increase their value for the hop count. Every node in the network finds a path to its anchor node using this technique that requires the fewest hops. Figure 3 depicts DV-HOP implementation.

The widely used APIT range-free localization technique has certain shortcomings. In the main structure of the APIT method, the target nodes detect the place of the anchor nodes first. This is performedusing the beacon signals that the target nodes receive from anchor nodes. These beacon signals are generated by anchor nodes.

Beacon signals carry a variety of data, including the IDs of anchor nodes andthe power used for transmission and for the position in the transmission. Generally, anchor nodes consume more power and are equipped for the localization of the devices. The target nodes select from the list three of the ‘n’ anchor nodes. During APIT process, target nodes attempt to identifytheir own locations by using the anchor. The target node searches for a position within the designated triangulation area. The target nodes would generate C3 triangles by testing each combination. The target node finds its COG inorder to determine its estimated position by locating the intersection of all probable triangles. Figure 4 shows the generic triangle area generation using the APIT method. The theoretical idea of point-in-triangulation (PIT) demonstrates whether the localization triangle zone is present on the exterior or interior of the zone. A TABC with threeanchor nodes andonetarget node is constructed while keepingin mind the target node T. T can travel in a way that either keeps it a long way from the triangle’s vertices or brings it fairlyclose to them. The ‘T’target node is present at the exterior of the triangle, and if this is the case, theninside the triangle if the opposite is true [14].

3. Principle of APIT Algorithm

Due to the fixed nature of sensor nodes in WSN, which prevents random mobility, PIT studies have been demonstrated to be inapplicable in a variety of practical applications. As a result, the problems in PIT are generally overcome by the APIT algorithm, in which there is no change in the network sensor point. It can be observed that anchor nodes are received from the RSST, and every final node action is described with a link tothe sensor nodes and is compared with the standard initial algorithm of the APIT. All the nodes are present together in the interior of the design, which is compared with the target node to obtain the range of communication. Figure 2 demonstrates the position of the target node andwhether the node is present on the interior or exterior of the triangle zone. No neighbor of the target node is located nearer or further away from any of the triangle’s three anchor nodes. The target nodes are always present in the interior of the triangle, and the rest of the nodes, like the anchor, are locatedcloseto the nodes. Target nodes and anchor nodes are also included in the neighbor nodes. The target node moves to the location of a nearby node, let us say node 4, and we can assume that since it is equallyclose to the anchor nodes, it is outside the triangulation area [15,16,17].

4. Simulation Results

The network region is bounded by a 1000 m radius. The position of nearby nodes can be determined viaa normal DVHOP algorithm; however, it does so with increased latency and inaccuracy. The proposed Bat with SA methods DVHOP and APIT canbe used to acquire the performance metrics parameters and outcomes. In terms of latency, the anchor shift illustrates randomization [18]. The performance of the DV-HOP and APIT algorithms with and without the Bat-SA localization approach is compared based on the results of the simulation. When assessing performance, the following variables are considered:

Changing Communication Radius

• Step size 100 and changing the area from 600 to 1200;
• Node shifting.

Changing radius from 120 to 240 with step size 20. Figure 5 and Figure 6 show the comparison of the simulation results of APIT and DV-hop without Bat-SA methodology.

Simulation results comparison between DV-Hop and APIT using the Bat-SA method areshown Figure 7 and Figure 8.

Figure 9 and Figure 10 compare APIT with DV-Hop without the Bat-SA method for the parameter areasregardingposition errors and latency [19]. The comparisons of APIT with DV-Hop without the Bat-SA method between the parameter areas with position errors and latency areshown in Figure 11 and Figure 12.

Table 1. Error tolerance of proposed and existing methods.

Algorithm	Max. Error	Min. Error	Average Error
Standard DV-Hop	0.374R	0.0382R	0.3011R
DV-HOP Anchor	0.877R	0.0222R	0.2414R
DV HOP (PSO)	0.789R	0.0200R	0.2311R
Proposed DV HOP	0.647R	0.0055R	0.1101R

illustratesthe error tolerance of the proposed and existing methods. The error tolerance is low in the proposed method.

Nodes are horizontally moved by one position: The comparison of the APIT and DV-Hop algorithms without the Bat-SA method is shown in Figure 13. Nodes move horizontally by one position. The nodes move to one position ten times [20]. The comparison of the APIT and DV-Hop algorithms with the Bat-SA method is shown in Figure 14 and Figure 15. It is evident from the aforementioned simulation findings that the DV-Hop and APIT algorithms surpass the Bat-SA method in terms of efficiency.

5. Conclusions

Compared to APIT, the conventional Bat method is more accurate. The suggested Bat with SA approach achieves better results than the traditional APIT and DV-HOP methods in terms of convergence, calculation rate, and success rate. The simulations’ findings show that when network area scalability rises, computation time falls. Additionally, compared to the conventional localization process, position accuracy and latency are enhanced. The association between the nodes is more firmly located than in the APIT algorithm. The simulation also shows the consistency of the proposed approach. Thus, the results of the entire system are enhanced, and the WSN is considerably optimizedwiththe help of DV-HOP and APIT with Bat-SA method. The APIT with Bat-SA localization approach outperforms the DV-HOP algorithms in a comparison of their efficacy. Future work on this subject may concentrate on multi-objective programming and the creation of novel optimization methods for localization challenges.