【摘要】 随着通信平台种类的不断丰富，选择合适的终端对任务进行分配非常有必要，于是提出基于群智感知计算的多终端引擎任务分配方法。为了解决因环境等存在多方面差异而导致适配技术不同的问题，设计了多终端协同内容适配服务适应更加复杂的环境。根据用户连接的网络和可用终端生成网络组合方式，通过AHP计算出网络权重对网络进行评估，综合考虑带宽、功率和费用三个指标将用户偏好分为两类构建判断矩阵，使用户的服务质量最高。最后采用任务需求特征提取方法对感知任务类型的关键词进行提取，选择出与感知任务匹配度更高的终端引擎参与任务，确保任务的效率，以及资源利用的最大化。实验结果表明，所提方法可以较好的体现出用户的偏好程度，任务类别提取准确率较高。更多还原

【Abstract】 With the increasing variety of data platforms, it is very necessary to select appropriate terminals to allocate tasks. Therefore, a multi-terminal engine task allocation method based on swarm intelligence perception computing is proposed. In order to solve the problem of different adaptation technologies due to various differences in environment, a multi-terminal collaborative content adaptation service was designed to adapt to a more complex environment. Then, according to the network connected by the user and the network combination mode generated by the available terminal, the network weights were calculated by AHP to evaluate the network. Considering the three indicators of bandwidth, power and cost, the user preferences were divided into two categories to build a judgment matrix, so that the user’s service quality was the highest. Finally, the task requirement feature extraction method was used to extract the keywords of the perceived task type, and the terminal engine with higher matching degree with the perceived task was selected to participate in the task, ensuring the efficiency of tasks and maximizing the utilization of resources. The experimental results show that the proposed method can better reflect the user’s preferences, and the accuracy of task category extraction is high.更多还原