Research On Strategies To Resource Allocation In Wireless Networks

With the diversification of wireless communication services and the proliferation of different wireless network technologies,the demand of corresponding resources in wireless networks is increasing dramatically for users.How to maintain a stable and efficient service while ensuring the fairness of network users as well as rational revenue at the same time is one of the most important research areas for wireless communication networks.And in the meanwhile,more and more people use wireless networks in recent years and they intend to enjoy wireless services in their daily time.This leads to the heterogeneity,diversity and temporal variations of users' requirements,which raise a significant influence on the functionality and quality of service in wireless networks.Based on the networking environment in their surroundings,users can choose whether to contact with the center of wireless service or not.When users are located within the same area of service center(i.e.,the WLAN case),one may interact with the service center directly and compete others for the access to spectrum resource.However,when the service center is far away from network users,the distributed computing can be introduced to serve these users' requests.Concluding above,this thesis studies the issue of resource allocation with different networking environments and users' strategies in wireless communications.Specifically,this thesis combines the wireless communication and auction theory,and uses economic models to capture the complex interactions between different users.Furthermore,considering the flexibility and computational complexity,this thesis designs resource allocation scheme in various networking environments.The goal of this thesis is to improve the network performance and meanwhile,guarantee the truthfulness for network users.This thesis investigates issues of resource allocation with different cases.On the one hand,for the WLAN case,this thesis tackle both the single-side and double-side spectrum auction in cognitive radio networks.On the other hand,for the distributed case,a hierarchical approach for computing resource allocation is discussed in hybrid cloud environments.Moreover,equipped with energy harvesting modules,a wireless network with energy constraint can prolong its lifetime and optimize its social welfare by an appropriate spectrum allocation scheme.The main work and contributions of this thesis are shown as follows.(1)On single-side spectrum auction in cognitive radios: To address the heterogeneity and flexibility of users' requirements,we propose a spectrum-division model in time and frequency domains,and build an optimization problem of winner determination to maximize the social welfare.Afterwards,we give the computational complexity of above problem and propose approximation algorithms with approximation ratio,respectively.We also design a payment strategy for secondary users,which is proven to be truthful.(2)On double-side spectrum auction in cognitive radios: Due to the egoism of primary users,they intend to participate the spectrum auction to guarantee their own profits.In this part,we represent an efficient double spectrum auction model with a knowledge of spectrum reuse,and employee a Markovian prediction based algorithm to generate guidelines for primary and secondary users,respectively.We also show that proposed auction model satisfies the economic properties.(3)On resource allocation in hybrid cloud environments: In this chapter,we propose a hierarchical approach for resource management in hybrid clouds,and use a game model to capture the complex interactions between the online service provider and cloud service provider.Thereafter,we design efficient online algorithms to distribute users' requests within above two providers.Finally,we determine the cloud service provider's and online service provider's optimal decisions.(4)On spectrum allocation in energy harvesting networks: In energy harvesting networks,nodes are capable of transmitting information with energy harvested from surrounding environments.When such a network is exposed,especially in rough environmental conditions,the issue of resource allocation is made rather challenging.To tackle this problem,we formulate this particular network as a combinatorial auction model,based on which we demonstrate that the Qo S of access point and network performance can be guaranteed.Moreover,we propose an approximation algorithm to obtain efficient and effective results,and derive the approximation ratio in different situations.