Beyond theory: a comprehensive survey of clustering algorithm applications in wireless and media communications

Abstract

The burgeoning advancement in communication technologies, coupled with the widespread adoption of mobile devices, has led to an exponential increase in communication data volumes. Navigating this deluge of data in the big data era poses significant challenges in terms of processing and analysis. Clustering, an unsupervised learning approach, stands out as a potent analytical tool in this context, capable of segmenting data into groups of similar entities to unearth underlying patterns and structures. Given its capacity to discern intrinsic data relationships, clustering has garnered considerable interest for its potential applications within the communication field. Despite the wealth of research, many existing surveys either focus solely on the theoretical aspects of clustering algorithms or restrict their scope to a single domain, often overlooking the intrinsic synergy between data transmission and content processing. Addressing this gap, this survey delivers an exhaustive review of clustering algorithm applications from a unique cross-disciplinary perspective, bridging the realms of wireless communications and media communications. We argue that as communication systems evolve towards semantic and content-aware networking, a unified view of these domains is essential. This comprehensive review not only aids readers in grasping the varied applications of clustering algorithms—ranging from physical layer security to semantic image analysis—but also empowers researchers to refine algorithm design by drawing on analogous use cases across these sectors. To enhance comprehension, we elucidate diverse clustering algorithm implementations within wireless and media communications, illustrating with detailed examples. Moreover, we highlight prevailing challenges and delineate future research trajectories for clustering algorithm applications in these fields.