Performance optimization of cluster head selection in heterogenous wireless sensor netowrk using clustering techniques
Budget: $30 – $250 USD
This research will use computer software including either to design and simulate the results:
• MATLAB.
• NS2
The Plan for Data Management Analysis
Stage A:
1. In NS2, wireless sensor network (WNS) nodes are initiated in known topology size (width and length).
2. Nodes number are isolated in three groups i.e. 100, 200 and 500 nodes at the time. Traffic to be initiated in known data rate in each node.
3. Using the number of nodes decided in point (2), nodes to be distributed randomly throughout the topology area.
4. When all nodes are transmitting the data into defined base station (BS), network performance is measured using metrics alike throughput and time delay.
Stage B: in NS2 , algorithm for node clustering is to be designed as follow
1. setting fixed/static node objects with variable locations (to be set randomly) and constant nodes IDs.
2. setting the thresholds of cluster e.g. cluster size, threshold location for measuring the node belongness.
3. Iterating of nodes number i.e. 100, 200 and 500 nodes at the time.
4. Apply search algorithm for sensing location and form the cluster, this algorithm can be Feed forward neural network.
5. As clusters are made, cluster-head (CH) to be assigned where all nodes will be reporting to it.
6. Locations of clusters and their nodes to be taken and the same to be regenerated in NS2 understandable coordination.
7. To apply the clusters into NS2 and to determine network performance in every iteration.
• MATLAB.
• NS2
The Plan for Data Management Analysis
Stage A:
1. In NS2, wireless sensor network (WNS) nodes are initiated in known topology size (width and length).
2. Nodes number are isolated in three groups i.e. 100, 200 and 500 nodes at the time. Traffic to be initiated in known data rate in each node.
3. Using the number of nodes decided in point (2), nodes to be distributed randomly throughout the topology area.
4. When all nodes are transmitting the data into defined base station (BS), network performance is measured using metrics alike throughput and time delay.
Stage B: in NS2 , algorithm for node clustering is to be designed as follow
1. setting fixed/static node objects with variable locations (to be set randomly) and constant nodes IDs.
2. setting the thresholds of cluster e.g. cluster size, threshold location for measuring the node belongness.
3. Iterating of nodes number i.e. 100, 200 and 500 nodes at the time.
4. Apply search algorithm for sensing location and form the cluster, this algorithm can be Feed forward neural network.
5. As clusters are made, cluster-head (CH) to be assigned where all nodes will be reporting to it.
6. Locations of clusters and their nodes to be taken and the same to be regenerated in NS2 understandable coordination.
7. To apply the clusters into NS2 and to determine network performance in every iteration.