LS and MMSE based Localization Algorithm for WSNs amid obstacles

: In recent years, optimization and Wireless Sensor Networks (WSNs) are extensively used in numerous milieus and hostile topographies. In this paper, we proposed an improved localization algorithm by means of Least square and Minimum mean square error that evolvements with the basic DV-Hop algorithm. The localization error is shortened by our proposed algorithm and improves the localization accuracy of the basic DV-Hop algorithm with minimum number of beacon nodes. The proposed method is simulated and compared with other algorithms such as DV-Hop, ROCRSSI and APIT. The simulation is considered both with obstacle as well as without obstacle. Keywords-: WSN, Distance estimation Method, LS, MMSE, DV-HOP, localization, algorithms I.INTRODUCTION Wireless sensor networks (WSNs) are molding voluminous deeds in our civilization, as they have turn into the quintessence of ubiquitous tools. WSNs obligate a boundless array of latent requests in both military and inhabitant applications, including robotic land-mine recognition, battlefield investigation, target tracking, environmental observing, wildfire detection, and traffic guideline. In these grave applications the liveliness of sensor location is the common feature shared by all. The core function of WSNs is to detect and report events which can be meaningfully assimilated and responded to only if the exact location of the affair is recognized. Also, in any WSNs, the location information of nodes plays an energetic role in understanding the application context. A straightforward solution is to equip each sensor with a GPS receiver that can accurately provide the sensors with their exact position. This is not a realistic solution from an economic perspective since sensors are often deployed in very large numbers and manual configuration is too cumbersome. Therefore, localization in sensor networks is a challenging one. In excess of the years, many protocols have been devised to enable the location discovery process in WSNs. In all these literatures, the focal point of location discovery has been a set of specialty nodes known as beacon nodes, which have been stated as anchor, locator, or seed nodes. These beacon nodes know their location, either through a GPS receiver or through manual configuration, which provide the location information to other sensor nodes. Using this location of beacon nodes, sensor nodes compute their location using various techniques. For the essential operation the elementary idea of WSNs is that, the proficiency of each discrete sensor node is limited and the cumulative power of the whole network is adequate. In several WSNs applications, the placement of sensor nodes is accomplished in an ad hoc manner without vigilant arrangement. Once positioned the sensor nodes, they must be able to solitary establish themselves into a wireless communication system. There are divergent kinds of localization approaches and accuracy necessities [1][2]. Localization can be coarsely alienated into two groups: Range-based approach and range-free approach. Range-based approach uses absolute distance estimate or angle estimate, significance that a node can measure the distances from itself to the beacons in a network. In disparity, range-free approach proceeds that a node to measure the direct distances from itself to beacons is unfeasible. A node can calculate approximately its regions or areas only through connectivity and proximity where it stays. Range-based localization be capable of alienated into a further two kinds. The first kind is distance estimation by one-hop; and the second kind is by multi-hop. Localization in WSNs is a multi-hop approach because a node may not converse directly with beacons. A node can send or receive messages to or from beacon nodes only through multi-hop routing. The sensed data may arrive at the destination by the multi-hop. The stability of a routing path is not guaranteed, so the routing path between the Z.Mary Livinsa et.al / Indian Journal of Computer Science and Engineering (IJCSE) ISSN : 0976-5166 Vol. 4 No.6 Dec 2013-Jan 2014 411 data source and data sink may diverge with time. During the multi-hop routing path error may be occurred which may affect the accuracy of localization in WSNs. Another aspect influencing localization accuracy is the ranging errors. Suchlike kind of ranging approaches is adopted; there will constantly survive some noise in the ranging measurements. Furthermore, since the uniqueness between each transmitter-receiver pair may not the same, this kind of disparity between different motes also pertain negative impact on the accuracy of localization. 1.1 Range based localization: The capability to measure the range of wireless signal transmissions is the key of range-based schemes. This algorithm is based on distance or angle measurements which are used to estimate the position of node in the network. The Rangebased Algorithm usually require expensive hardware (for example: antenna) which is not feasible for WSNs. So the distance estimation is performed using the following techniques RSSI, AOA, TOA and TDOA. Received Signal Strength Indicator (RSSI) is based on the attenuation of received radio signal with distance. Since the nodes are equipped with radios to perform communications, the distance estimation based on received signal strength has attracted enough attention. Thus RSSI can be used to estimate the distance between the two nodes. RSSI is the cheapest one because it does not require any additional device. The functions of RSSI measurements in its protocol are defined by the physical/medium access control layer protocol of IEEE802.15.4.Standard.Range estimation is then completed by path loss model for RF propagation. However, RSSI based range measurements suffer from noise and link reliability. Efforts were expended to obtain the mapping between RSSI measurements and the associated distances to capture the impact of multipath fading and environmental variations on RSSI measurements in the indoor and outdoor space [3]. Probabilistic model of RSSI range measurements was also introduced in [4] to address the uncertainties and irregularities of the radio communication patterns. In which RSSI value can be mapped to a log-normal distribution of the distance between the two nodes. Angle of Arrival (AOA): In which a node position is calculated by estimating the angles between neighbour nodes. AOA measures the angle at which the signals are received from anchor nodes. Generally AOA provide more accurate localization. But they are not applicable to WSNs as they require very expensive hardware. Time of Arrival (TOA): It measures the distance between nodes using signal propagation time or signal’s time of flight from source to destination. To estimate TOA, this technique requires precise synchronization between the sender and receiver clocks and high speed sampling of receiving signal. The foremost challenge encrustation TOA based techniques is the difficulty of accurately measuring the time of flight, in the meantime the propagation speed could be tremendously high compared to the distance to be measured. Even though computing distances between each pair of locations is insignificant, the inverse problem of finding the node locations given the pairwise Euclidean distances is far from trivial. Another common method for range measurement is based on the time difference of arrivals (TDOA). The signal could be radio frequency (RF), acoustic or ultrasound. In this technique the distance between nodes are calculated by sending two different signals which travel at different speeds to their neighbors, and then uses their arrival time difference in propagation times of radio and acoustic signals originated at the same point. Given the time difference of the arrivals, the distance between the sender and the receiver can be obtained by multiplying the time difference by the speed of the ultrasound signal. 1.2 Range free localization: The range-free localization is being considered as a cost-effective alternative to range-based methods because of hardware limitation of deployment of WSNs devices. Irregularity transmission propagation as well as stringent restriction on cost of hardware has rendered localization a very challenging. The range free localization is more capable and promising to achieve higher localization accuracy without introducing any extra hardware in comparison to range-based technique of localization which depends on received signal strength to calculate absolute point-to-point distance. Range free localization technique deploys information related to network topology as well as connectivity status for valuating location. Low cost, no extra hardware, little communication traffic as well as flexible precision in position estimation is some of the advantageous features of range-free methods. Therefore range-free technique is considered to be most effective solution for the localization issues in wireless sensor network. In comparison to range-based approach, the range-free techniques facilitate sensor nodes to evaluate their position without depending on parameters like distance or angles. Such methodology normally requires various anchor nodes, that enable position unknown sensor nodes to estimate their position by using the radio. DV-Hop localization, APIT, Centroid localization, amorphous positioning etc. is some typical algorithms of Range free techniques. Based on the algorithm of DV-HOP, sensor nodes estimate their position based on the anchor positions, number of hops from anchor, and also the average distance per hop [5][6]. Amorphous Z.Mary Livinsa et.al / Indian Journal of Computer Science and Engineering (IJCSE) ISSN : 0976-5166 Vol. 4 No.6 Dec 2013-Jan 2014 412 positioning algorithm uses offline hop-distance estimations, improving location through a neighbor-information exchange [7]. Low density of anchors poses a challenge to the multi-lateration approach. In order to apply multilateration, DV-distance follows the approach similar to DV-hop. The distances of each hop are summed up to approximate the distance between a non-anchor node and an anchor node that is multiple hops away. The approximated distance is then used in the localization process. One possible variation could be to use the Euclidean distance instead of the multi-hop distance. The Euclidean distance can be computed from geometric relationships and the single hop distances. Therefore, additional neighbors and the corresponding range measurements are needed to eliminate the false estimation. On based on DV-Hop localization algorithm we instruct a new technique which guarantees to reduce the localization error to a massive level. Techniques such as Least Square (LS) and Minimum Mean Square error (MMSE) are linked together with basic DV-Hop algorithm. The proposed method is simulated and compared with other algorithms such as DV-Hop, ROCRSSI and APIT. The organization of the paper will be as follows: Section 2 presents the DV-Hop Algorithm and improved DV-Hop Algorithm based on Euclidean distance estimation method. Section 3 presents proposed method using LS and MMSE. Section 4 explains Simulation scenarios and Simulation Results. Finally, section 5 concludes the proposed work. II. DV-Hop Algorithm The algorithm is divided into two portions. In the first portion, each and every beacon node broadcasts a beacon signal at extreme power so that the signal stretches throughout the network. Initially all the unknown node and the anchor node location with a hop-count value are set to one. Every receiving node withstands the minimum hop-count value per beacon of all beacons it gathers. Beacons with greater hop-count values are defined as stained information and will be ignored. Then the nonstained beacons are flooded outward with hop-count values at every intermediary hop. Through this mechanism, all nodes in the network get the minimal hop-count to every beacon node. In the second segment, once a beacon node gets hop-count value to other beacons, it calculates an average size for one hop, which is then flooded to the intact network. After receiving hop-size, visor folded nodes multiply the hop-size by the hop-count value to derive the physical distance to the beacon. The average hop-size is predicted by beacon node using the succeeding formula: 2 2 ( ) ( ) j k j k k j jk k j x x y y HopeSize h ≠