The Internet of Things and Ubiquitous Intelligence (3)

Editor's Desk:
The traditional Internet is oriented towards person-to-person connection, whereas the Internet of things (IoT) is oriented towards connections between inanimate objects. IoT covers a larger range of connections and involves more semantics than traditional Internet. Traditional Internet and telecom networks focus on information transfer, but IoT focuses on information services. By combining sensor networks, Internet, telecom networks, and cloud computing platform, IoT can sense, recognize, affect, and control the physical world. The physical world can be unified with the virtual world and human perception. In this part, we discuss ubiquitous network convergence and cooperation technologies in terms of their application scenarios, technical advantages and research directions. We also briefly introduce IoT services.

5 Application Scenarios and Technical Advantages of Terminal Cooperative Technologies
    Terminal cooperative technologies are mainly applied in the following scenarios:

    Scenario 1: Self-organized networking and cooperation among multiple terminals of one user
With terminal cooperative technologies, multiple terminal devices of one user can be networked via radio interfaces in a self-organized way. A self-organizing network can manage and maintain multiple computing devices in a distributed environment. These devices are frequently changed and loosely coupled, and they can dynamically discover, select, aggregate and match terminal capabilities according to user preferences, service characteristics, locations and environments.  Thus, terminals can cooperatively support the same service.

    Scenario 2: Self-organizing among terminals in network architecture comprising a cellular network and wireless multihop technologies

    The greatest challenge for future cellular mobile networks is coverage. Most cellular networks operate at relatively high frequency bands and are likely to have poor penetration and poor non-line-of-sight (NLOS) capability in future high-transmission 4G networks. Coverage and transmission will be unstable, and coverage holes may even arise. One solution to these problems is to increase the number and power output of base stations. But this increases network construction cost greatly. Research shows that even advanced signal processing technologies, such as MIMO and adaptive antennas, cannot solve the coverage problem. An effective approach is to develop a new network architecture that combines cellular and multihop wireless technologies.

    Self-organizing relay between mobile terminals (mobile relay stations) is often applied in group communication, emergency communication, security mechanisms, and military communication. Mobile terminals are often regarded as valuable supplements to fixed relay stations in public communications, that is, fixed relay stations deployed by the operator. These have fewer functions and cost less than base stations and access points (APs).

    Scenario 3: Self-organizing among terminals in a hybrid network comprising cellular networks and wireless sensor networks

    In traditional wireless sensor networks, a large number of nodes are required to complete a user-specified task by means of peer-to-peer, multihop and self-organizing wireless communication. Networks based on static data access nodes have inherent problems, including unbalanced energy consumption, low transmission efficiency, inflexible deployment, and single network architecture. They are also prone to routing voids, coverage holes, and bottlenecks and overall network performance is decreased.

    A hierarchical network architecture that combines wireless sensor networks and cellular networks can effectively solve coverage and transmission problems. This type of network architecture is more suitable than others for future ubiquitous, heterogeneous and cooperative networks. Mobile terminals (mobile data access points) such as mobile phones, PDAs, and notebooks have distinct advantages in energy efficiency, bandwidth, reliability, transmission range, and mobility. Mobile terminals with multiple interfaces can use their relatively high-rate wireless interfaces to send and receive network data. They can also use the positioning functions of cellular networks and sensor networks to provide users with diverse location-based services. Cooperation among mobile terminals can also significantly enhance the performance of wireless sensor networks. By handing complex work such as data processing, access processing, data forwarding, and routing maintenance over to mobile terminals, wireless sensor networks can minimize data errors (or packet loss) arising from multihop wireless transmission. Moreover, they can use the powerful computing capability of mobile terminals to process information within the networks, thus lightening their own loads. Mobile terminals in a heterogeneous network can use their own wireless interfaces to communicate with the fixed cellular base stations in the upper layer or access points of other wireless networks or even to form an ad hoc network with other mobile data access points. In this way, mobile terminals can effectively exchange data with each other by means of cooperation and can also greatly expand the applications of wireless sensor networks.

    Terminal cooperative technologies can bring many advantages.
    First, cooperative technologies can exploit available terminal capability and user network resources for ubiquitous, heterogeneous convergence. By organically integrating different wireless technologies, cooperative technologies can take advantage of the capabilities of several terminals of one user. They can also take advantage of self-organizing relays among terminals and the rich capabilities of sensor networks to converge ubiquitous and heterogeneous networks and services.

    Second, cooperative technologies can increase the flexibility of communication. They offer users and terminals diverse access modes and service provision methods. They can use the advantages of both infrastructure and non-infrastructure networks to attain good network performance, flexibility and scalability. Convergence of infrastructure and non-infrastructure networks is likely to be important in future mobile communication systems.

    Third, cooperative technologies solve the coverage hole problem and expand the coverage of wireless networks. When cellular networks are combined with multihop wireless communication technologies, single-hop, poor-performance radio links are replaced with high-performance multihop ones. This eliminates coverage holes and increases the coverage of each base station or AP.

    Fourth, cooperative technologies increase system capacity. The multihop forwarding mechanism can reduce the transmit power, which enhances the spatial multiplexing of channels. When local services are set up in a self-organizing way, the bottleneck in access points is mitigated. By transferring the traffic of hotspot cells, resources are dynamically adjusted, and resource use is improved. All these help increase system capacity.

    Fifth, cooperative technologies balance load. The traffic of wireless coverage areas is often unbalanced and changeable. Dynamic self-organizing relays can distribute the traffic of one cell to surrounding cells, reducing the probability of service congestion and handover failure that arises from insufficient capacity.

    Sixth, cooperative technologies save radio resources. Mobile nodes that are close to each other can directly communicate with each other in a self-organizing way. In this way, the system load is decreased and system capacity is increased.

6 Research Direction of Terminal Cooperative Technologies
    Terminal cooperative technologies support heterogeneous network convergence and are regarded as one of the core components in future mobile communications. At present, they are a focus of research. However, in an environment where infrastructure networks and non-infrastructure networks are converged on the basis of terminal cooperation, there are many problems that arise from heterogeneous wireless technologies and many problems that arise from dynamic, multihop self-organizing non-infrastructure networks and cooperation among terminals. These problems involve routing protocol, network control, mobility management, service provision, and evaluation of system performance.

    Among other issues, the following are in urgent need of attention:

    Issue 1: Cooperation-based self-organizing adaptive network control mechanism
The infrastructure network and non-infrastructure network have different network control mechanisms. Most infrastructure networks have a centralized network control mechanism, whereas decentralized non-infrastructure networks have a distributed control mechanism. The differences between the two control mechanisms is reflected in their control and management technologies, that is, routing protocol, QoS guarantee, and resource management. It is necessary to develop a self-organizing and adaptive network control mechanism that combines the advantages of centralized and distributed mechanisms. Such a mechanism should effectively control dynamic, multihop, multipath wireless communication that comes with cooperative technologies.

    Issue 2: Mobility management technology
    Most traditional mobility management technologies focus on location management and horizontal handover control within a specific network. Mobility management technologies for heterogeneous networks focus on hierarchical location management and vertical handover control over different access technologies. In infrastructure networks, mobility management technologies are relatively mature; but in non-infrastructure networks, there is still no systematic mobility management technology. The introduction of cooperative technologies enables infrastructure and non-infrastructure networks to be organically integrated. A mobile terminal may have several network interfaces and support several communication modes, including cellular, self-organizing and hybrid. It needs information about the relative and absolute location of other mobiles for effective, self-organizing cooperation. Therefore, cooperation-based mobility management technologies need to integrate location management, positioning technologies, and location-based services in order to implement handover control in self-organizing adaptive communication.

    Issue 3: Service cooperation technology
    In ubiquitous, heterogeneous networks, there are a large number of intelligent devices that provide a range of information and services. The peer-to-peer feature of nodes in non-infrastructure networks enables each mobile terminal to act as a service provider. Hence, self-organizing among terminals requires efficient service cooperation technologies, including cooperative service provision modes as well as broadcast, discovery and request mechanisms for dynamic, self-organizing, and lightweight services. Efficient use of network resources and services is key for automatic configuration of application services.

    Issue 4: Evaluating the impact of terminal cooperation on system performance
    Existing cooperation schemes are mainly designed to address a specific problem in a specific scenario with the goal of optimizing local performance. In these schemes, only intuition and qualitative analysis are used to evaluate the impact of self-organizing multihop communication and multihop access on system performance. There is still no systematic quantitative analysis, and the advantages of cooperative technologies to performance are still in dispute. It is necessary to quantitatively analyze the impact of cooperative technologies on network coverage, system capacity, service availability, and reliability. This will provide a useful reference for network planning, cost accounting, and determining network performance and optimization.

7 Services of the Internet of Things
    IoT has applications in daily life as well as in industry automation. Typical applications include intelligent power grids, intelligent banking, intelligent transportation, intelligent home, and intelligent medical systems. The services of IoT can be classified according to different criteria. They can be roughly divided into four types according to technical features and development trend: identity-related, information aggregation, cooperative sensing, and ubiquitous intelligence [1].

    Identity-related services are delivered using identification technologies such as RFID and two-dimensional code. In information aggregation services, the IoT platform manages the terminals, data, applications, and services in a unified way. Terminals collect and report data and do not need to communicate with each other. Because terminal technologies tend to be diverse, intelligent, and multimode in IoT, communication between the terminals and between terminals and individuals is more frequent and complicated than ever. As a result, one trend in the development of IoT is that terminals cooperate to complete a service. The vision of IoT is to provide ubiquitous intelligent services anytime, anywhere and in any way.

    Ambient Intelligence (AmI) was proposed by the Information Society and Technology Advisory Group (ISTAG) of the European Commission. It was used for the launch of the sixth framework programme (FP6) in Information, Society and Technology (IST). AmI builds on pervasive computing, ubiquitous computing, and human-centered computer interaction design. It is an adaptive digital environment in which many sensing and computing devices are embedded. It can judge a person’s intention and reaction by identifying the person’s posture, physiological condition, gestures, and voice in different situations. AmI has the following features:

• context-awareness. AmI can recognize the situational context of a user. Sensors embedded in the context change the physical information into valid statuses or operations, such as identity authentication and acquisition of the user’s current location.
• personalization and adaptability. AmI can be tailored to a user’s specific needs and can meet changes in user needs.
• anticipation. AmI can anticipate the desires of users without conscious mediation. This enables users to unconsciously interact with the environment, process information in a simple and natural way.

  In brief, ubiquitous and intelligent services are new implicit services that are designed to be user-centered in human living and work environments. Delivering these services requires IoT in the ubiquitous sensing layer as well as the powerful information processing of cloud computing. (To be continued)

Reference
[1] Xiaojiang Xing, Jianli Wang, Mingdong Li, “The Internet of Things and its Key Technologies,” ZTE Communications, vol.2, no.16,pp.27-30.