Abstract:
Identity theft in distributed networks, such as the internet, is becoming a significant
problem due to Sybil attacks. In these attacks, perpetrators create multiple fake
identities to manipulate network decisions. Traditional cryptographic methods for
detecting these attacks can be resource-intensive and sometimes ineffective. This
study suggests using Received Signal Strength R_S_S based techniques in massive
MIMO systems, along with diversity combining methods like Selection Combining
(SLC), Equal Gain Combining (EGC), and Maximal Ratio Combining (MRC). The
research also involves using machine learning algorithms such as K-means and
K-medoids for clustering to distinguish between legitimate nodes and attackers.
Machine learning techniques like KNN and Naive Bayes are used to conclude results.
The results indicate that configurations with 30-40 antennas and low noise levels can
achieve high detection accuracy. This method combines physical-layer security with
machine learning, making future wireless networks more resilient against Sybil
attacks.
