Radio Resource Management For Heterogeneous Networks Based On Auction Models

Existing heterogeneous wireless networks are often composed of multiple high-power base stations and low-power base stations. In the heterogeneous wireless network, some coverage areas of macro cells and micro cells are overlapped, where are usually hotspots or blind spots. Channel assignments in heterogeneous wireless network are usually unfair and cheatings in the allocation exist, so methods that allocating channel resources truthfully has been research hotspots. Hence we conducted the following studies.Firstly, channels of wireless networks are heterogeneous and have differences in costs, prices, transmission rates and so on, in that a novel multi-attribute auction mechanism is proposed in this thesis aiming to allocating heterogeneous channels, which is improved on the basis of Mc Afee double auction mechanism. First, we find all the interference-free users by finding all the independent subsets with graph theory method. Secondly, we group multiple interference-free users as large virtual buyers, take high-power base stations and low-power base stations that hold multiple channels as sellers, and an auction agent independent of buyers and sellers host the transactions between them. There are a variety of matching results for the buyers and sellers in the transactions. Comparing several optimization matching methods, we select the matching method resulting in a maximum transmission rate as the optimal matching. Proofs show that the proposed method is a truthful auction mechanism encouraging bidders reporting bid price truthfully in the auction. Simulation results show that, compared to other mechanisms, the proposed mechanism has higher total transmission rates and channel reusability, as well as higher seller revenues and buyer utilities. The proposed mechanism is also a polynomial time complexity of the mechanism. Thus, the proposed mechanism is a truthful, efficient multi-attribute auction mechanism for heterogeneous channels.Secondly, for the users moving randomly, users' channel requirements updating in time, and users preferring continuous time slots, we proposed an online combination auction mechanism aiming to deal with them. Based on the traditional auction mechanism, this mechanism is added with pre-emption factor, which promotes the buyer who has already won continuous time slots to win more continuous time slots channels; the mechanism is also introduced with adjacent factor, thus users who have less interferences to others also have advantages to bid. Meanwhile, theoretical analysis proved that this proposed mechanism is able to encourage users biding truthfully in price and time. Simulation results show that, compared to other mechanisms, the proposed mechanism has a higher buyer satisfaction and channel utilization, as well as higher seller revenues and buyer utilities. The proposed mechanism is also a polynomial time complexity of the mechanism. Thus, the proposed mechanism is a truthful, efficient online combination auction mechanism.In summary, the two auction mechanisms proposed in this thesis are truthful, and can deal with the problems about channel resource allocation existed in heterogeneous wireless network. Therefore, the research in this thesis is worthy to conduct.