This paper proposes a mobile complex event processing (CEP) architecture for collaborative personal monitoring. The proposed architecture is designed by combining machine learning, distributed intelligence, efficient communication, and hybrid storage. With the merging trend of mobility and real-time monitoring, the proposed approach will open many opportunities to explore the collaborative potential of patient/person monitoring in various domains. Researchers in machine learning, mobile context-awareness, mobile situation-awareness, distributed computing, and ubiquitous/pervasive computing areas will want to study this work.

Different from most current remote monitoring systems that analyze the measured data in isolation, the proposed mobile CEP architecture extends the current remote monitoring system with a high-performance backend server that achieves collaborative monitoring by tracking information from four different contexts: personal real-time context regarding what a person is doing; environmental context regarding the environmental temperature, humidity, and so on; community context regarding data from other relevant users; and historical context regarding the past that can be used for generating different analytics against which the personal real-time data will be compared.

As shown by the cardio use case, with the exception of the collection of personal data such as the heart rate and running speed of jogging users, the proposed mobile CEP system is able to search all community users to find jogging users’ relevant users, for example, those who are also jogging near the monitored users, and execute the collaborative analysis over all the collected data to give objective users precise and reasonable recommendations. As a result, the context-aware adaptive behavior of users is enabled. However, the paper would be more informative if the authors had provided the architecture’s implementation details and fine-grained experimental results.