Identification of Relationships between Electric Meter Boxes and Smart Meters in Power Communication

This paper investigates the relationship between electric meter boxes and smart meters in low-voltage distribution areas. Identifying this relationship accurately is critical for analyzing power line loss management, as it provides essential data support and a solid decision-making basis for optimizing distribution networks, reducing losses, and improving efficiency. To attain this objective, we first collect the Received Signal Strength Indicator (RSSI) information within the communication network of low-voltage distribution areas. Then, a distanceprobability mapping model is proposed to transform the RSSI data into a probability distribution that characterizes the relative distances between smart meters. Subsequently, the Markov Clustering Algorithm (MCL) is employed to analyze the RSSI data. Finally, validated across 38 real-world scenarios, the algorithm achieves an average recognition success rate of $97 \%$ in 33 urban distribution networks (indicating a 25-55 percentage-point improvement compared with traditional methods) and $72 \%$ in rural areas (indicating an 11-16 percentage-point increase), confirming its excellent performance under real-world variability.