Performance Enhancement Of MAC And Routing Protocols In Cognitive Radio Ad Hoc Networks

Due to limited natural available spectrum and rapid increase in communication technology,there is tremendous research attention that is being started in the spectrum scarcity area,to make use of the available spectrum efficiently.Currently,the spectrum is allocated in a static manner where it is under-utilized in terms of time and geographical location(space).To solve this issue,the federal communication commission(FCC)made a provision to opportunistically use of licensed spectrum with the introduction of dynamic spectrum access(DSA)using software defined radio(SDR).Cognitive radio(CR),built on SDR enables dynamic spectrum exploitation by changing the operating parameters like frequency,power,modulation,etc.complying with environmental conditions.With this,existing TCP/IP stack has to be amended to work along with dynamic spectrum access,cognitive radio technology.Common control channel(CCC)performs a critical role in a cognitive radio medium access layer(CR-MAC)and cognitive radio network layer in the simultaneous path selection across the spectrum holes(unused spectrum or white space)for an end to end transmission of application data in the CR network.Thus,efficient route selection in cognitive radio ad hoc networks(CRAHNs)depends on the reliable control channel.In a multi-channel ad hoc network,a channel is dedicated for exchange of control messages.All nodes are synchronized through control channel for the exchange of control messages to contend for data channel in ad hoc traffic indication message window(ATIM window).However,there is no dedicated control channel for node synchronization in CRAHNs,which opportunistically access licensed channels.This results in decreased network throughput and increases energy utilization due to interference caused by multi-channel hidden terminals.Thus,the performance of CRAHNs relies on the efficient and updated design on medium access control and routing algorithms which complies with rules of dynamic spectrum access(DSA).Current CR-MAC and Routing algorithm designs are based on in-band and out-of-band which have some inherent problems to be addressed.First,in-band information exchange is based on sequence and group CCC suffers from multi-channel hidden terminal problems.Also,throughput is severely affected by excess control overhead due to increased channel rendezvous time when there are more number of primary user(PU)channels.Similarly,intruders attack out-of-band information exchanges,causing single point network failures.Also,with wireless tech-nologies like Bluetooth,Wi-Fi etc.out-of-band CCC is getting saturated causing interference to one another.Henceforth,there is a need to amend the existing CR-MAC and routing protocol standards to address issues like node synchronization,power control,interference,enhancing throughput,and reducing delays.To address the current obstacles in CR-MAC and routing algo-rithm's established on common control channel,we have designed"performance enhancement of MAC and routing protocols in cognitive radio ad hoc networks".This PhD dissertation fo-cuses on improving end to end connectivity which enhances throughput and reducing delays by minimizing control overhead at the MAC layer and routing layer.This dissertation gives so-lutions to CCC design problems which propose two directional CR-MAC protocols at the data link layer and one directional routing protocol.First,a hybrid directional CCC based CR-MAC is proposed to resolve dynamic control channel availability issues in CR networks and enhance the networks performance.Many of the existing CR-node communications use in-band CCC or out-of-band CCC.Channel rendezvous time is high in case of in-band CCCs,and out-of-band CCCs are prone to saturation and interfer-ence.Thus,resulting in frequent link failures and control overhead hinders the performance of the MAC layer protocol.In the proposed approach,we employ directional antennas to enhance the spectrum utilization,conserve node transmit power,and reduce interference,and the hybrid control channel along with GPS is used to share cognitive control data.In this stage,node sorts out common control channel with the exchange of control messages.Nodes communicate with their neighbour with the help of directional antenna table which provides the antenna indices to be used for communication.Experimental results explain that designed directional antenna based hybrid CCC based software defined CR-MAC protocol has increased link layer through-put and reduced link layer delays with minimized node energy consumption when measured with omni-directional antenna based software defined CR-MAC protocols.Second,a hybrid directional CR-MAC based on q-learning with directional power control is proposed to assist nodes in switching to non-overlapping heterogeneous channels opportunisti-cally with improved quality of service.The common control channel(CCC)is used for network initialization,control message exchange,and synchronization.In CR ad hoc networks,nodes need to be vigilant as channel availability keeps changing.With a large number of vacant or unused PU channels,network setup and discovery time will be great in case of in-band CCC.Alternatively,out-of-band CCC's which have dedicated channel is relatively faster in synchro-nization compared to in-band CCC's.However,the static or fixed control channel is subject to congestion and attack from intruders,consequently degrading the CR-networks performance.Even though dynamic channel availability issues are addressed in"a hybrid directional CCC based CR-MAC",nodes use random channel selection policy resulting in the selection of a non-optimal channel for forwarding data.With the use of q-learning algorithm,nodes find the best available channel for data transmission.Also,to allow neighboring nodes to share channel and improve throughput,we use directional transmission power control scheme which allows neigh-boring nodes to reuse the channels.Node calculates optimal power to be used to communicate with its neighbour based on the received signal strength.The simulation results show that nodes using the proposed algorithm which select channels adaptively and uses optimal transmission power,helps to achieve higher throughput with minimized power consumption.Finally,cognitive radio assisted wireless sensor network(WSN)with interference-aware routing protocol is proposed that allows increasing throughput and reducing delays in WSN n-odes.In CR based networks common control channel(CCC)dictates the throughput,energy consumption,and message routes.Hence,efficient CCC based routing algorithm is necessary for end-to-end(sensor nodes to sink)application for selection of the shortest path with fewer delays and control overhead.The existing state of the art routing protocols are overlay based either in-band or out-of-band.Both in-band and out-of-band techniques have unsolved issues;former suffers from high synchronization time problem whereas the later suffers from satura-tion and interference.Wireless sensor nodes operate in heavily crowded industrial scientific medical(ISM)bands(902 MHz/2.4GHz),where severe performance degradation occurs due to channel saturation and high collision rate.In this proposed work,at first clustering is done with sensor nodes as the cluster members and cognitive nodes as cluster heads and gateways.Clus-ter heads and gateway nodes help to forward sensor data to sink and finally to end-user.Later,CR nodes(cluster head)route sensor nodes data in opportunistic licensed channels using direc-tional hybrid CCC based AODV routing with power control.Simulation out-comes show that the designed solution with dynamic spectrum access enhances the throughput,minimizes energy consumption,and reduces the delay compared with existing solutions.