Research And Applications Of Differentiated Quality Of Service Guarantee For Heterogeneous Random Access Networks

Wireless communication can be divided into centralized access networks represented by the cellular system and random access networks represented by wireless local area networks Wi Fi,wireless body area networks,etc.The latter has achieved great success in the past ten years due to the advantages of easy networking,low cost,and high robustness,but it has always had problems of unstable performance and inability to provide quality of service guarantee.However,with the diversification of services and the high demands of users,the network is gradually showing a trend of heterogeneous development,and increasingly diverse quality of service requirements are emerging,such as online games and other real-time services that require low latency and AR / VR and other data services that require high throughput also pose more stringent challenges to random access networks.Although the IEEE 802.11 standard proposes improvements to the Enhanced Distributed Channel Access(EDCA)protocol,which sets multiple priority types to provide differentiated services,it still cannot meet the user's delay requirements,and the network throughput still fell sharply with the number of users increases.Therefore,this thesis takes the diversified service quality assurance of heterogeneous wireless random access networks as the core issue,and conducts theoretical and application optimization research.First,based on the unified analysis model in the previous homogeneous network,this thesis proposes a theoretical model of heterogeneous IEEE 802.11 wireless LAN.The state transition process of head-ofline packet of the node transmission queue in the network is modeled with an embedded Markov chain.Considering the general heterogeneous wireless local area network with M groups of different types of nodes,each group has different nodes parameters of quantity,input rate,initial backoff window and maximum backoff phase.Based on this model,explicit expressions of throughput and delay performance were obtained,which intuitively showed the relationship between performance and the above parameters,and verified its accuracy through simulation.In view of the current wireless LAN's difficulty in meeting the service quality requirements of diverse applications,this thesis combines the actual traffic characteristics of real-time and data services to study how to meet the needs of real-time services in a partially saturated multi-service coexistence heterogeneous wireless LAN scenario under the premise of end-to-end delay constraints,maximize the throughput of data services.Based on the analysis of the theoretical model,the explicit expressions of the maximum achievable throughput of the data service and the optimal initial backoff window sizes of the corresponding node under the constraints of the real-time service delay are obtained for the first time.The theoretical analysis is verified through simulation.Under standard parameters,as the network size increases,the delay of real-time services rises and the throughput of data services decreases.Under the optimized parameters,the delay of real-time services can always meet the requirements below Given the delay constraints,while the throughput of data services can be maintained at the maximum,the optimized throughput performance can be improved by up to 20.7%.Finally,this thesis extends the theoretical research to the anti-fallen network of the elderly in one of the important scenarios in the body area network,to solve the problem that the current IEEE 802.15.6wireless body area network standard parameter settings cannot guarantee the communication delay in the network problem.Consider a unsaturated time-sensitive anti-fallen body area network.Through theoretical analysis,the optimal initial backoff window sizes solution of fall-detection signals and daily signals that minimizes delay of fall-detection signals are obtained.Verification by simulation tools,compared with the current standard parameter configuration,the optimized settings can ensure that the communication delay performance of fall-detection signals is stable and lower than the fall response delay constraint.In summary,this thesis takes the quality of service guarantee of heterogeneous random access networks as the core research problem,starting from a simple and accurate general theoretical model,and applying it to both heterogeneous wireless local area networks and anti-fallen body area networks which are highly practical application scenarios,corresponding optimization theories and application schemes are proposed respectively,which provide theoretical guidance for the future heterogeneous wireless random access network that can provide quality of service guarantee.