Bloom Filter Approach for Autonomous Data Acquisition in the Edge-Based MCS Scenario.

Area of Science:

• Computer Science
• Ubiquitous Computing
• Data Science

Background:

• Mobile crowdsensing (MCS) generates massive datasets from numerous mobile and wearable devices.
• User mobility causes data obsolescence, necessitating efficient data processing techniques.
• Bloom filters (BF) have shown promise in reducing redundant data in hierarchical edge-based MCS.

Purpose of the Study:

• To extend the Bloom filter (BF) algorithm for MCS to handle multiple data readings of the same type at a single location.
• To evaluate the communication overhead and overall performance of the enhanced BF algorithm in a real-world MCS dataset.
• To demonstrate significant data reduction and energy savings in mobile crowdsensing systems.

Main Methods:

• Implementation of an extended Bloom filter (BF) algorithm capable of managing multiple data readings.
• Evaluation using a real-world dataset to assess communication overhead between edge servers and end-user devices.
• Comparative analysis against a baseline approach to quantify performance improvements.

Main Results:

• The enhanced Bloom filter (BF) algorithm significantly reduces the volume of transmitted data in MCS.
• Energy savings of up to 62% were achieved compared to the baseline approach.
• The algorithm effectively manages redundant data and supports MCS tasks requiring multiple readings at specific locations.

Conclusions:

• The proposed extended Bloom filter (BF) algorithm is an effective method for optimizing data transmission and energy consumption in mobile crowdsensing (MCS).
• This approach enables more efficient data processing and informed decision-making for users in MCS environments.
• The findings highlight the potential of advanced filtering techniques for scalable and sustainable crowdsensing applications.