In-network processing of an iceberg join query in wireless sensor networks based on 2-way fragment semijoins

Area of Science:

• Computer Science
• Wireless Sensor Networks

Background:

• Wireless Sensor Networks (WSNs) face challenges in processing complex database operations like join queries due to energy and computational constraints.
• Traditional iceberg join query processing prioritizes value matching over cardinality constraints, potentially leading to inefficiencies.

Purpose of the Study:

• To develop an energy-efficient in-network processing strategy for iceberg join queries in WSNs.
• To propose an alternative approach that addresses the cardinality constraint before value matching for improved performance.

Main Methods:

• Logical fragmentation of join operand relations based on aggregate counts of joining attribute values.
• Generation of an optimal sequence of 2-way fragment semijoins.
• Utilization of Bloom filters within fragment semijoins for efficient synopsis of attribute values.

Main Results:

• The proposed method filters non-joinable tuples more effectively and energy-efficiently compared to previous schemes.
• Experimental results demonstrate considerable performance improvements of the alternative approach.

Conclusions:

• The novel approach of meeting cardinality constraints first offers a more efficient strategy for iceberg join query processing in WSNs.
• This method enhances the feasibility of complex data processing within the energy-constrained environment of WSNs.