摘要 : Recently, a new architecture regarding a crowd of mobile devices as mobile cloud has been proposed, where the neighboring mobile devices are pooled together for resource sharing. Most existing works focus on mobile cloud framework... 展开 Recently, a new architecture regarding a crowd of mobile devices as mobile cloud has been proposed, where the neighboring mobile devices are pooled together for resource sharing. Most existing works focus on mobile cloud framework design and assume that the nearby users are willing to share their resources. Since each user is rational and self-interested, a proper market mechanism is needed to incentivize users to participate and allocate resources among multiple users. In this paper, we consider a broker-based mobile cloud, where multiple supplying users have idle resources to share, and multiple demanding users suffer from resource limitations. We propose and analyze a double-sided bidding mechanism where each demanding user submits a bid to choose a demand resource-price function, and each supplying user submits a bid to choose a supply resource-price function. We consider the cases with price-taking users and price-anticipating users who would anticipate the influence of their own bids on the price, respectively. For the case with price-taking users, we first show that the proposed mechanism admits a unique competitive equilibrium that maximizes the social welfare of the mobile cloud and then develop an optimal distributed algorithm to achieve the desired equilibrium point. For the case with price-anticipating users, we first formulate the interaction among multiple users as a strategic game and prove the existence and uniqueness of the Nash equilibrium and then develop a distributed algorithm to compute the Nash equilibrium. Numerical results validate the efficacy of our proposed algorithms and illustrate that the social welfare achieved at the Nash equilibrium is very close to the optimal social welfare. 收起