DCTP: Data collecting based on trajectory prediction in Smart Environment

Abstract

In this paper, we have proposed and designed a realtime distributed predicted data collection system-DCTP (Data Collection based on Trajectory Prediction according to Knowledge mined from trajectories) to solve the congestion and data loss caused by too many connections to sink node in indoor Smart Environment scenarios (like Smart Home, Smart Wireless Healthcare and so on). DCTP predicts and sends predicted data of the sensor nodes which people is going to pass at one time instead of sending the triggered data in several times. Firstly, our system applies data mining to get the knowledge of transition probability among sensor nodes from the historical binary motion data. Secondly, each sensor node stores the corresponding knowledge based on a special storage mechanism. Thirdly, each triggered sensor node predicts the next destinations people will arrive at according to the received message using HMM algorithm. At last, the sensor node sends its triggered data and the predicted data to the sink node. The significances of DCTP are as follows: (a) the procedure of DCTP is distributed; (b) it effectively reduces the connection between sensor nodes and sink node. The time complexities of the proposed algorithms are analyzed and the performance is evaluated by some designed experiments in a Smart Environment.

Modeling Users’ Data Traces in Multi-Resident Ambient Assisted Living Environments

“…Residents' behaviors, and in turn, their traces can be effectively modeled by Markov models (MMs). 2,5,[9][10][11][12] Although learning the parameters of such models is a straightforward task in single user environments, it is a challenging issue in multi-occupant environments. Because, in a multi-occupant environment, the traces are interwoven temporally; hence, a dataset of separated users' traces is not available beforehand for training or designing a model.…”