An Optimization Approach to Multi-Sensor Operation for Multi-Context Recognition

Mobile devices and sensors have limited battery lifespans, limiting their feasibility for context recognition applications. As a result, there is a need to provide mechanisms for energy-efficient operation of sensors in settings where multiple contexts are monitored simultaneously. Past methods for efficient sensing operation have been hierarchical by first selecting the sensors with the least energy consumption, and then devising individual sensing schedules that trade-off energy and delays. The main limitation of the hierarchical approach is that it does not consider the combined impact of sensor scheduling and sensor selection. We aimed at addressing this limitation by considering the problem holistically and devising an optimization formulation that can simultaneously select the group of sensors while also considering the impact of their triggering schedule. The optimization solution is framed as a Viterbi algorithm that includes mathematical representations for multi-sensor reward functions and modeling of user behavior. Experiment results showed an average improvement of 31% compared to a hierarchical approach.

A General Framework for Making Context-Recognition Systems More Energy Efficient

Context recognition using wearable devices is a mature research area, but one of the biggest issues it faces is the high energy consumption of the device that is sensing and processing the data. In this work we propose three different methods for optimizing its energy use. We also show how to combine all three methods to further increase the energy savings. The methods work by adapting system settings (sensors used, sampling frequency, duty cycling, etc.) to both the detected context and directly to the sensor data. This is done by mathematically modeling the influence of different system settings and using multiobjective optimization to find the best ones. The proposed methodology is tested on four different context-recognition tasks where we show that it can generate accurate energy-efficient solutions—in one case reducing energy consumption by 95% in exchange for only four percentage points of accuracy. We also show that the method is general, requires next to no expert knowledge about the domain being optimized, and that it outperforms two approaches from the related work.