Throughput improvement in multihop ad hoc network using adaptive carrier sensing range and contention window

Demand for decentralized, wireless, ad-hoc systems, where hosts are free to leave or join, to replace wired communication systems has seen a phenomenal growth. Such networks need little or no infrastructure support to operate. Deploying these networks such as in wireless sensor networks (WSN) enables new frontiers in developing opportunities to collect and process data from remote locations. The large number of nodes in these wireless networks invariably results in higher node densities and increased levels of network interference. Interference mitigation is therefore crucial in ensuring these networks operate efficiently. Often the lack of network planning and regulations for such networks require the targeted access strategy to be adaptive to network conditions and distributed.;The goal of this research is to design an algorithm employing mathematical tools in optimizing spatial reuse among nodes in the ad hoc network so that multiple communications between nodes can proceed simultaneously thereby maximizing the network throughput. To maximize spatial reuse, the IEEE 802.11 Medium Access Control (MAC) protocol would be modified so that each transmitting node can fine-tune its data rate and carrier sense range adaptively depending on minimal receiver response local data. All nodes must be able to detect and communicate with their neighbors in order to determine the network structure, to execute network functions and transmit collated information back to the remote node. The network topology will be discovered using clustering schemes such as the K-means technique that minimizes the Euclidean distance between random nodes. Each cluster will have a cluster head that would keep track of local information about nodes in its cluster. A further goal of this research would be to demonstrate that the physical carrier sensing incorporated in the 802.11 MAC protocol can adaptively optimize the sensing threshold of the nodes and minimize interference within the network without the benefit of the request-to-send and clear-to-send handshake of the virtual carrier sensing. Considerable nodal energy and packet overhead would be saved by turning off the RTS/CTS handshake process. An analytic design will be presented for acquiring the optimal sensing threshold given a network topology; data rate and transmit/receive power of the nodes.;Two major issues to be addressed in improving spatial reuse are: (1) The optimal range of transmit data rate/ carrier sense threshold for maximum network capacity. (2) The relationship between the carrier sense threshold and contention window.;Furthermore, results from this research will show that tuning the carrier sense threshold and contention window offers several advantages including delivering considerable aggregate throughput more than that obtained from a static carrier sense threshold network with no previous knowledge of the network topology. This will enable nodes sustain a high data rate, while maintaining the adverse effect of collision on other neighboring simultaneous communications at minimum. In the end, the communication protocol will be improved to achieve better utilization of the scarce wireless spectrum. The simulation and performance evaluation tools required for this work would be Network Simulator-2 (NS-2) simulator, AWK and PERL programming languages.