Population bursts management in digital city

In this paper, we study a novel problem, population burst detection, one of the important issues related population monitoring and management within a city. Based on the detected population bursts, we can trace and analyze these bursts, which can help security departments to infer the cause of bursts and be prepared for the future possible bursts. Though it is useful, it is not trivial to detect population bursts, especially under the condition that we can hardly get the population samples within a city from time to time. To address the difficulties of lacking real population data, we take the advantage of communication networks, specifically, mobile phone networks, which offer enormous communication data between peoples. Most importantly, we find the fact that we can use these communication data to infer the population samples. Therefore, we propose an effective and efficient algorithm to detect bursts over call volumes. We verify the performance of our proposed mechanism with an onsite case study and real calling data.