Information Technology Reference
In-Depth Information
3 Localization Technologies in WSNs
Currently, localization algorithm in WSNs can be divided into two types: range-based
localization algorithm and range-free localization algorithm. Range-free localization
algorithm using received signal strength or wireless connections between nodes to
determine the distance to neighbor nodes and achieving localization [1]. Range-based
localization algorithm include RSSI [5], TOA [6], TDOA [7], AOA [8], and range-
free localization algorithm include DV-HOP [9], centroid algorithm [10], convex
programming method [11] and so on.
Range-free localization technology which is not need node's absolute distance or
orientation has been proposed. It has reduced the requirements of the node hardware.
Meanwhile, range-free localization algorithm is simple and easy to achieve, distrib-
uted local calculation, small communication load and good at scalability. So it is
widely using in the location algorithm.
3.1 DV-HOP Algorithm Analysis
Distance Vector-Hop algorithm is proposed by D. Niculescu and B. Nath [9], its
location principle is that calculate the minimum number of hop between positioning
nodes and reference nodes at first. Then, estimating the distance of average hop, the
minimum number of hops multiplied by the distance of average hops can get the es-
timating distance between positioning node and reference node. Last, triangular
measurement method or maximum likelihood estimation method is used in estimation
coordinates of the node positioning.
3.2 LDV-HOP Algorithm Description
In DV-HOP algorithm, the sum of hop distance is instead of actual distance. Accumu-
lation error will be increased for the addition of number of hop. If the receiver node is
just accepted the scope of local reference node information, which is in a small
amount near of positioning node. This will reduce the accumulative error and help to
improve positioning accuracy and reduce the whole of network communication load.
Conclusion what have been mentioned, limit DV-HOP(LDV-HOP)algorithm was
proposed. LDV-HOP has divided local processing into three stages.
First, the minimum number of hops has been computed among positioning nodes
and reference nodes with limited number of hops. Reference nodes broadcast its posi-
tion information to the neighbor nodes. Positioning node record what it have received
the minimum number of hops about reference nodes, and the larger number of hops
which come from the same reference node should ignore. Finally, all the nodes have
recorded the minimum number of hops to each reference node. Second, the count of
hops distance has been computed among positioning nodes and reference nodes with
the limited number of hops. According to record information that is location informa-
tion and the number of hops about the other reference nodes, each reference node
estimated the actual distance of average jump following the formula (1), and broad-
cast the result to the network.
Search WWH ::
Custom Search