Auction has been widely used to spectrum allocation. Most of the previous works supposed that all
the spectrums are identical. However, in reality, spectrums are quite different in different locations and
frequencies. Recently, some works studied the double auction mechanism for heterogeneous spectrums.
But their schemes are based on the assumption of “single-channel request”. To be more realistic, each
seller and buyer will bid at least one channel. The previous schemes will not work under multi-channel
assumption.
In this thesis, I proposed a truthful multi-channel double auction mechanism for heterogeneous spectrums. Our scheme allows sellers and buyers to sell or buy multi-channels for heterogeneous spectrums. We introduce a novel virtual grouping method to split sellers and buyers. We proved that the proposed scheme satisfies the economic properties: truthfulness, individual rationality and budget balance. Simulation results confirmed that our method achieves high auction efficiency and auction revenue. Beyond the double auction for heterogeneous spectrums, recent spectrum auction results have shown that small network providers cannot benefit from the auction directly because of the high price asked by the spectrum holders. Therefore, in this thesis, we proposed a truthful group buying-based double auction mechanism for cognitive radio networks. There are two single-round auction in our method. The first one is between secondary users and secondary access point, in which the secondary access point is the seller and the secondary users are the buyers. We call it the outer auction. The outer auction is based on single-sided buyer-only auction. The other one is between the secondary access points and the spectrum holders, in which the secondary access points are the buyers and the spectrum holders are the sellers. We refer to it as the inner auction. In the inner auction, we apply the double auction mechanism. We proved that our scheme satisfies the economic properties.
At last, we proposed a truthful multi-channel double auction mechanism for spectrum group-buying.
Since both sellers and buyers would require to trade multiple channels at the same time. No existing designs can meet multi-channel and group-buying requirements simultaneously. To solve this problem, we introduce a novel group splitting and budget calculation algorithm in the outer auction. We apply a proper winner determination and pricing mechanism in the inner auction. This scheme satisfies the economic properties as well.
History
Advisor
Liu, Derong
Department
Electrical and Computer Engineering
Degree Grantor
University of Illinois at Chicago
Degree Level
Doctoral
Committee Member
Anasari, Rashid
Mazumder, Sudip K.
Seferoglu, Hulya
Darabi, Houshang