A Cognitive-Inspired Event-Based Control for Power-Aware Human Mobility Analysis in IoT Devices.

Area of Science:

• Computer Science
• Mobile Computing
• Internet of Things (IoT)

Background:

• Mobile Edge Computing (MEC) shifts processing to network edges, crucial for IoT device autonomy.
• Mobility-Based Services (MBSs) require energy-efficient and accurate location tracking on mobile devices.
• Existing solutions often struggle with uncertainties and optimizing energy consumption for location services.

Purpose of the Study:

• To propose and evaluate a cognitive controller for on-device Mobile Edge Computing (MEC).
• To achieve energy-efficient and accurate location tracking for Mobility-Based Services (MBSs).
• To leverage Cognitive Dynamic Systems (CDS) for exploiting mobility information and managing uncertainties.

Main Methods:

• Developed a fully on-device Cognitive Dynamic Systems (CDS) platform.
• Implemented a cognitive controller utilizing mobility-aware sampling policies.
• Conducted experiments on smartphone devices to evaluate the proposed strategy.

Main Results:

• The proposed cognitive control strategy significantly improved energy savings compared to existing methods.
• Demonstrated enhanced spatio-temporal accuracy in location tracking for Location-Based Services (LBS) and MBSs.
• Validated the effectiveness of the mobility-aware sampling policies in managing uncertainties.

Conclusions:

• The on-device cognitive controller effectively balances energy efficiency and location accuracy in MEC environments.
• This approach offers a promising solution for autonomous and self-aware mobile devices in the IoT.
• The cognitive controller's ability to exploit learned mobility information enhances performance for LBS and MBS.