AFFINITY PROPAGATION BASED SMART DATA ANALYTICS IN 5G NETWORK
Abstract : Network embedding assigns nodes in a network to low dimensional representations and effectively preserves the network structure. Recently, a significant amount of progresses have been made toward this emerging network analysis paradigm. In this survey, we focus on categorizing and then reviewing the current development on network embedding methods, and point out its future research directions. We first summarize the motivation of network embedding. We discuss the classical graph embedding algorithms on cognitive radio environment and their relationship with network embedding. Afterwards and primarily, we provide a comprehensive overview of a large number of network embedding methods in a systematic manner, covering the structure- and property-preserving network embedding methods, the network embedding methods with side information and the advanced information preserving network embedding methods. Moreover, several evaluation approaches for network embedding and some useful online resources, including the network data sets and software, are reviewed, too. Finally, we discuss the framework of exploiting these network embedding methods to build an effective system and point out some potential future directions.
? The existent adaptation mechanisms system are usually reactive, they solely react when a tangle happens.
? This for the most part limits the network ability to produce intelligent and efficient solutions, also with regard to inexperienced networking and advantageous business models.
? Cognitive Radio Networks (CRNs) give the rise of spectrum utilization by using unused or less used spectrum.
? Unauthorized users have access to licensed spectrum, below the condition that the interference perceived by the authorized users is lowest.
? High end-to-End delay and reduce network life time .
? The network situation is outlined as wireless detector network with 30 nodes at random deployed within the space of 500 m X 500 m. 250 m is that the transmission vary of every of the detector nodes within the network.
? Increase communication time go on between the supply node causing the packet and therefore the destination node receiving the packet.
? Node Pairing and Embedding is not possible and run time.
? If we regard network embedding as a way of network representation learning, the formation of the representation space can be further optimized and confined towards different target problems.
? Taking the network node classification problem as an example, if we have the labels of some network nodes, we can design a solution with network structure as input, node labels as supervised information, and embedding representation as latent middle layer, and the resulted network embedding is specific for node classification.
• The most effective way of spectrum sensing is to directly detect the primary Rx, because it is the Rx of a PU system that should be protected. In general, the PU systems can be divided into the following two categories:
1) One-way communication systems and
2) Two-way communication systems.
• One-way communication systems have only one direction communication from the primary Tx to the primary Rx, such as TV and radio broadcasts. The only way of detecting this kind of Rx’s is to sense the leakage signals from active Rx’s.
• Two-way communication systems have bidirectional communications, and there are interactions between the Tx and the Rx, which can be used for spectrum sensing. Next, we will introduce the sensing methods for the two kinds of systems, respectively.
? High tolerance of delays
? Large volume of CR device cost economies
? Low power is required
? very low mobility required
? No network of CR base stations required
|