Research Article

Power Profiling of Context Aware Systems: A Contemporary Analysis and Framework for Power Conservation

Table 4

Comparative overview of contemporary power conservation techniques.

Conservation TechniquesAdvantagesDisadvantagesRequirements/SuitabilityTechnical Aspects

Threshold-Based Conservation(i) Simple and easy to implement(i) Lack of adaptability
(ii) Device level conservation
(i) Battery operated devices that need to shut down in a stable manner(i) Non-adaptive and hard threshold used for stable power-on and power-off procedures in battery operated devices

Demand-Based Conservation(i) Adaptive threshold(i) Device level conservation(i) Battery operated devices that need to predict consumption based on current usage(i) Rule based system
(ii) Predictive analysis

Cloud-Based Conservation(i) Use of cloud
(ii) Adaptive threshold
(iii) Reduction of power awareness overhead at device level
(iv) Power conservation at application level
(i) High communication overhead(i) Suitable for cloud assisted environment(i) Cloud-based system
(ii) Can allow classification techniques for adaptive thresholding

Context Change Detection-Based Conservation(i) Adaptive threshold
(ii) Context monitoring
(iii) Proactive approach
(iv) Power conservation at application level
(i) Context change detection is required a-priori(i) Context aware systems(i) Context awareness
(ii) Context differential

Statistics-Based Conservation(i) Adaptive threshold
(ii) Predictive conservation
(iii) Power conservation at application level
(i) History management overhead
(ii) Statistical calculation overhead
(i) Low power consumption history management module(i) Predictive analysis
(ii) Regression

Selective Sensor Access-Based Conservation(i) Selective sensor approach
(ii) Power conservation at sensor level
(i) Minimal list of sensors for each activity classification
(ii) Limited context classification scenarios
(i) Suitable for large number of sensors present in an environment(i) Power conservation for energy hungry sensors

Source Code Optimization-Based Conservation(i) Source code optimization keeping in view power constraint(i) Cumbersome to implement
(ii) Compile time conservation
(iii) Non-adaptive thresholding
(i) Suitable for embedded systems
(ii) Non-generic devices and applications
(i) Source code optimization
(ii) Power consumption measurement for programming constructs and procedures