What heterogeneous networks converge in the Internet of Things?

The integration and autonomy of heterogeneous networks is one of the most remarkable features of the Internet of Things. Due to the diversity of application requirements and network technologies, under the framework of the Internet of Things, a variety of networks will coexist, including personal local area networks, wired and wireless local area networks, metropolitan area networks and wide area networks. These networks with different performance characteristics complement and promote each other. How to realize seamless integration and autonomous management between them and meet users' needs more effectively and flexibly is one of the important technical challenges facing the Internet of Things.

The integration and autonomy of heterogeneous networks mainly include the management of massive addresses and data, the selection of access mechanisms and the autonomous management of heterogeneous resources. First of all, in the Internet of Things, the allocation and management of massive address space, the mapping of object addresses and labels, and the transmission and storage of massive data caused by a large number of objects have become the first problems to be solved in heterogeneous networks. Secondly, due to the different performance characteristics of various networks, it is difficult to find the truly optimal access selection scheme by using the traditional single-objective decision theory. Therefore, it is necessary to introduce multi-objective decision-making theory to find a balance between limited resources and multiple goals required by users, so as to achieve the goal of multi-objective optimization. Finally, due to the heterogeneity of Internet of Things resources and the dynamic nature of the network, the autonomous management of resources is the key content of the research. In the information perception layer with self-organization as the main form, the key is the core protocols of self-perception and self-configuration, including time synchronization protocol, distributed positioning protocol, topology control protocol, self-organizing routing protocol and energy management protocol. In the access/network layer, in order to support the mobility of users and nodes, not only the horizontal handover technology between different cells in the same network, but also the vertical handover technology from one network to another network is needed. Due to the differences in data rate, spectrum and QoS of heterogeneous networks, the precise location determination and fast handover mechanism required for vertical handover will be more complicated. At the same time, in the heterogeneous environment, self-aware dynamic allocation of distributed spectrum (bandwidth) based on context-aware technology is also one of the trends of resource management. Multi-radio cooperation (MRC) is the key technology to realize the above resource management, that is, a single node is equipped with multiple independent radio systems, and each radio system can use different access technologies and different channels. Because a node can establish connections with different access systems at the same time, it can also keep multiple connections with one access system at the same time, which is helpful to realize fast vertical handover and dynamic resource allocation.

(1) Data Fusion and Information Processing

The nodes in the Internet of Things have the characteristics of large number, small volume, limited energy and massive data. Therefore, from the perspective of improving information accuracy and reducing energy consumption, effective data fusion and information processing technologies are needed. These technologies permeate all levels of the Internet of Things. In the information perception layer, energy can be saved by moving relay, grouping nodes working in turn, selecting representative reporting nodes, compressing sensing and other mechanisms, while ensuring the integrity and accuracy of information; In the access/network layer, the data is reorganized and merged mainly through convergence processing and various routing control protocols to reduce the data transmission volume; In the application service layer, the distributed database technology is mainly used to further screen the received data to achieve the purpose of data fusion; At the same time, according to the dynamic characteristics of user and environment data information changing with time and space, context-aware processing based on multi-level fusion is carried out.

(2) service search and discovery

Compared with the traditional telecom network and Internet service mode, the service of the Internet of Things is different in that it emphasizes the active provision of services, so it needs more advanced and complex service search and discovery technology. At present, there are three Web service search and discovery technologies: direct search, centralized architecture search and distributed architecture search. Direct search means that the user directly asks the service provider for a copy of the service description; Centralized architecture search means that service providers register services in a central directory and publish service announcements and references for users to retrieve; Distributed architecture search means that there is a reference to the service description at the service provider's point on the Web site, and users can obtain available Web services by specifying to check the Web site. The search and discovery of Internet of Things services need to add an active link on the basis of the above technology, that is, automatically search, discover and assemble the appropriate services according to the needs of users, so as to realize the reliable transmission and active provision of services in the dynamically changing heterogeneous network environment.

(3) Guarantee of safety and reliability

Security and reliability in the Internet of Things are mainly embodied in two aspects: network security and information security. Network security includes system security, including hardware platform, operating system and application software, as well as continuous and reliable normal operation of the system and uninterrupted network services. Information security refers to protecting the accuracy, authenticity, confidentiality, integrity, availability and controllability of information. Compared with the traditional Internet, due to the miniaturization of nodes and the limitation of energy capacity, the Internet of Things needs to consider the balance between the computational strength and the security strength of the algorithm, that is, how to achieve as strong security as possible through simpler algorithms and lower energy consumption.