Research Of Media Access Control Protocol & Cross-layer Optimization For Mobile Wireless Sensor Networks

With developments of hardware and network technologies,wireless sensor networks have gained much more attention than ever before.In the military area,ecological environment monitoring area,smart agriculture area,logistics monitoring area and many other circumstances,wireless sensor networks have been wildly used and are still expected to deploy wilder.In ocean and river monitoring,biocenose monitoring and atmosphere monitoring applications,sensor nodes are usually in the moving state.So wireless sensor networks technology must be adapted to the moving nodes.However,the traditional wireless sensor network technologies are aimed to using in the immobilized situation and fail to accommodate the mobile network.Therefore,it is necessary to do research about mobile wireless sensor networks technologies.Among the technologies,protocols are always the popular and difficult area of researching wireless sensor networks.Inside the protocols,because the MAC(Media Access Control)protocol has important functions of wireless channel and physical layer state management,it is crucial to the network and QoS(Quality of Service)performance.Meanwhile,research of the individual layer of network cannot adapt to the increasing demands of applications,the cross layer optimization which can improve the network performance by multi-layer information interaction and cooperation have became the focus.It aims to make the network server the applications in better way.Therefore,main contributions of this paper are as follow:1.First of all,this paper proposed one analytical probabilistic model to investigate delay and stability performance of different MAC mechanisms which including 802.11DCF-like(without sleeping)mechanism,sleeping mechanism,adaptive listening mechanism and different dutyCycle values.Through the analysis,we find some interesting results: In the without-sleeping network,performance of delay and stability keep in good and is directly relevant to contention window size.Sleeping mechanism does have negative impact on delay and stability;moreover,adaptive listening mechanism can optimize them to some extent.Furthermore,through controlling dutyCycle value,delay and stability performance can be improved.Finally,the simulation exhibits consistent results which obtained from analytical probabilistic model.2.This paper proposed one Load-Speculation based Throughput Adaptive Protocol(LSTAP).LSTAP is organized by two part: Load-Speculation based Contention-window Adaptive mechanism(LSCA)and Hop-Differentiate based Adaptive Interruption mechanism(HDIA).LSCA put forward one conditionally interrupt continuous exponent control back off algorithm.This algorithm divides network into two different states:Frequent Collision State(FCS)and Infrequent Collision State(ISC).In each state,the algorithm has individual strategy to adjust the contention window size.Also this algorithm provide one sensitive method which is controlled by the continuous exponent to make the transformation process between two states be smooth when network load is changed.The results of simulation confirm that LSCA is better than constant contention window size and binary exponent back off mechanism in the performances of delay,collision ratio,throughout and energy efficiency.The HDAI mechanism can interrupt working period based on AIT(Activity Interrupt Time)which is calculated with hops value from node to sink and neighbor nodes number.When node have movement event,the circumstance around the node was changed.But the HDAI mechanism can detect the real time difference about hops value and neighbor number and recalculate the AIT to make sure network perform well.By simulating HDAI and SMAC,results show that HDAI does negative impact on delay and throughout,but it can improve the energy efficiency performance effectively.At last,this paper make a detail comparison between LSTAP and SMAC about the performance of delay,throughout and energy efficiency,results confirm that LSTAP dose have positive improvement.3.In mobile wireless sensor networks,nodes have apparent movement character.Based on this character,first of all,this paper propose one ORPB(On-demand Routing Protocol with Backup mechanism).Compare with AODV,ORPB not only found a primary route,but also found the most independent and minimum hops backup route using one single route discovery process.When the primary route was fail,the backup route can keep the normal transmission.Meanwhile,ORPB can maintain both the primary and backup routes in best condition by using route control packet which are useful.Simulation results show that OPRB does improve the performances of route overhead,delay,packet delivery ratio and energy efficiency for mobile wireless sensor networks.Secondary,based on analysis of cross-layer optimization methods,RA-ORPB(RSSI-Assisted based ORPB)is proposed.RA-ORPB provide one mechanism that can predict moving tendency of neighbor node by using RSSI information interaction among physical layer,MAC layer and network layer,so that it can select the more reliable routes to improve communication dependability.Simulation results show that RA-ORPB can make apparent improvement about route overhead,delay,packet delivery ratio and energy efficiency.The cross-layer method is convinced to be effective.