Cooperative Abnormality Detection via Diffusive Molecular Communications.

Area of Science:

• Molecular Communications
• Networked Sensing
• Signal Processing

Background:

• Abnormality detection is crucial for networked systems.
• Molecular communications (MCs) offer a novel approach for sensor networks.
• Existing optimal detection methods in MCs can be computationally intensive.

Purpose of the Study:

• To develop computationally efficient abnormality detection algorithms for diffusive molecular communication networks.
• To compare the performance of different molecular release schemes (different type of molecule vs. same type of molecule).
• To propose low-complexity detectors suitable for resource-constrained fusion centers, such as nano-machines.

Main Methods:

• Derivation of optimal decision rules for both different type of molecule (DTM) and same type of molecule (STM) transmission schemes.
• Transformation of the optimal decision rule for STM into a low-complexity equivalent.
• Development of sub-optimal, low-complexity detectors for DTM based on log-likelihood ratio (LLR) approximations.
• Performance analysis using false alarm and missed detection probabilities.

Main Results:

• An equivalent low-complexity decision rule was found for the STM scheme.
• Sub-optimal, low-complexity detectors were proposed for the DTM scheme.
• The proposed low-complexity detectors demonstrate practical viability for MC systems with limited computational power.
• Simulation results validated analytical findings and highlighted performance-complexity tradeoffs.

Conclusions:

• Low-complexity abnormality detection is feasible and beneficial in diffusive molecular communication networks.
• The choice between DTM and STM schemes, along with detector complexity, impacts overall system performance.
• The proposed detectors offer a practical solution for abnormality detection in resource-limited MC environments.