Abstract:

With the continuous advancements in computer networks and communication technology, covert connections have become more accessible, faster, and increasingly secure, making them challenging to detect. These covert operations involve transmitting secret messages through channels that breach system security policies, posing significant risks to data security. Traditional security measures are inadequate in identifying these hidden dangers since covert channels exploit unconventional means of communication. This comprehensive review focuses on covert operations, covering their definitions, types, and advancements, with a special emphasis on leveraging machine learning (ML) for detection. ML methods have proven effective in analyzing vast amounts of data and identifying patterns associated with covert communication. Various ML strategies, including supervised learning, unsupervised learning, and deep learning, are examined in the context of combating covert channels. The review delves into the accomplishments and limitations of ML approaches, highlighting the progress made in detecting hidden routes. To evaluate the performance of ML classifiers, a comparative experimental investigation is conducted, considering detection accuracy, false positive rates, and computational efficiency. The findings offer valuable insights into the strengths and weaknesses of ML techniques when countering covert communication. Ultimately, this review emphasizes the ongoing need for vigilance and continued research in detecting and countering covert channels, as data security remains at risk in today's interconnected world

Description:

.