资源有限连续双向拍卖环境下的ZIP2策略

ZIP2 strategy of limited resources under environment of continuous double auction

利用经济模型和多Agent技术研究网络资源分配问题已成为当前网络研究新的趋势。针对Agent资源有限的网络环境,提出了一种基于连续双向拍卖市场机制的ZIP2策略。ZIP2策略是一种包含价格和数量的二维报价策略,Agent采用该策略具有机器学习能力。最后通过模拟实验验证了ZIP2策略可以实现较高的网络资源分配效率,平均分配效率超过97%。

Using economic model and multi-Agent technology to research network resource allocation problem has become a new trend in the field of network study. Concerning the limitation of the Agent＇s resource in the network, the ZIP2 strategy based on market mechanism of continuous double auction was proposed. The ZIP2 strategy is a two-dimensional bidding strategy which contains price and quantity. The Agents adopting the strategy possess the machine learning ability. Finally, the results of simulation experiments show that the ZIP2 strategy can achieve high network resource allocation efficiency, the average allocation efficiency over 97%.