| dc.description.abstract | With the rapid development of Internet of Things (IoT) technology, smart services such as Industry 4.0, smart health and smart home have been derived. In order to easily deploy IoT devices, these smart services can be effectively
applied in our lives. Currently, there are already many kinds of low-energy wireless protocols, such as ZigBee and Thread, which support mesh networks. Although Bluetooth 5.0 also supports mesh networks, Bluetooth 5.0 uses a flood-based architecture that uses simple broadcast properties to communicate between devices. If the topology of the mesh network is larger, it will easily cause network congestion, accelerate energy consumption and shorten the life cycle of the network.
This study in thesis intends to build a scatternet network and the device feature value is normalized to the minimum value (Min-Max Normalization), and then the Role Suitability Metric (RSM) is calculated. The RSM value is used as an indicator of the formation of a distributed network. The scatternet network is divided into three procedures. The first procedure is a node discovery
procedure, the second procedure is the formation of the piconet that selects the master node according to the RSM value, and so the node with the maximum value becomes the master node, and the third procedure is the formation of a
scatternet network that selects relay nodes and combines two piconets to form a scatternet network. Only the master node and the relay node can transmit messages in the scatternet network, this design which can mainly improve the characteristics of repeated transmission packets in the flooded architecture, extend the network life cycle and reduce the overall network stability by reducing the transmission of messages between devices and the probability of network congestion.
The experimental results show that the device feature value is first normalized to the minimum value (Min-Max Normalization), which can be used to select the appropriate master node and relay node more accurately. The value
is not affected by the unit and the level is high or low. Adjusting the weight ratio can be consistent with practical applications, such as: reducing the weight of power consumption to reduce power consumption, increase battery life, and extend the network life cycle. | en_US |