5 core technologies promote the development of the Internet of Things

Currently, there is no any wireless technology that can meet all needs of the Internet of Things in terms of effective range, cost, bandwidth, and power consumption. Most enterprises and organizations must therefore plan to use multiple technologies simultaneously.

From an innovation point of view, objects are divided into many categories. However, some objects cannot meet the traditional definition of the Internet of Things because they cannot link to the Internet or do not have the ability to sense and interact. For this study, we have expanded the scope of the object to the following four categories:

1. Identifyable objects: Passive objects that are identifiable but not intelligent, such as "Radio Frequency Identification System" (RFID) applications. Such objects cannot be networked due to their more primitive applications, but they still have interactive features.

2. Communicating/sensing objects: These objects contain sensors that transmit information about themselves or their environment (for example, wireless pressure sensors in car tires).

3. Controlled sensing objects: In addition to sensing and transmission, such objects can also receive messages and be manipulated. One example is a heating boiler that can be switched on and off remotely.

4. Smart Autonomous Object: For the complex object that can transmit messages, it can incorporate various sensors and precision functions to perform autonomous operations. An example is a car. Autonomous objects are often complex and may contain a variety of simple objects.

The various objects mentioned above will be applied to different fields of technology. In general, due to factors such as size, power consumption and cost, simpler objects such as communication/sensing objects and controllable sensing objects are more likely to utilize specialized technologies. Smarter objects, such as smart autonomous objects, are more likely to use traditional IT technologies, so class objects are usually larger and more resource-constrained. For emerging trend groups interested in reaching the Internet of Things, they must pay attention to the following five technologies:

First, low-power wireless networking technology

There are many wireless networking technologies developed or developed for the Internet of Things. The following three types are the most critical:

1. Personal area network: This type of low-power network can only be linked to sensors and human peripheral instruments within a range of a few meters, and can be used for other purposes such as medical care or personal electronics. Examples include Bluetooth LE, and the latest 802.15.6 local area network standard supporting three physical network transport technologies. There are also other special terms used to express this concept.

2. Long-Range Sensors and Mesh Networks: These technologies are designed for long-distance applications (from tens of meters to kilometers) but still retain low-power features. Some can also support multiple network topologies such as star, mesh, and point-to-point. At present, the most familiar technology in this field is ZigBee, and Dash-7 has gradually matured, and it has the potential to be a low-power application with an effective range of several kilometers. The market needs both honeycomb-wide coverage and low-power technologies, but there is no global standard that has become a climate. The two R&D technologies are more promising, including the Japanese Wide Area Universal Network (WAUN) and the wireless network of Neul. The industry also introduced 802.11h technology with a Wi-Fi variant with less than 1GHz bandwidth, which provides low-power, long-range, and low data rate capabilities, but it is difficult to achieve standardization by 2015.

3. Special application networks: There are several wireless technologies that have risen in special applications today. One part is proprietary, so its attractiveness is not as good as other standards that have already been widely accepted in the market. The ANT+ technology used in medical care and sports sensing is just one example. There are also wireless HART in the field of industrial automation. Highway Addressable Remote Transducer Protocol) and ZWave for home automation.

Currently, there is no any wireless technology that can meet all needs of the Internet of Things in terms of effective range, cost, bandwidth, and power consumption. Most enterprises and organizations must therefore plan to use multiple technologies simultaneously.

Second, optimize sensor data management

For many years the academic community has been devoted to the research of data transfer and query between small smart objects. The initial results include TinyDB, and systems such as SENFIS, AnduIN and Antelope that have just started in recent years.

Antelope is a distributed sensor database with a static RAM footprint of only 3.4Kb. The low imprint databases scattered across the sensor nodes, most of which still remain at the academic research stage, suggest that companies and institutions that want to track these technologies can use academic resources such as the ACM Digital Library.

Third, low-power embedded operating system platform

For simple objects that require long standby time, or “mote” (mote), TinyOS, IRIS, LiteOS, MansOS, Contiki, and other low-key operating systems have been developed. In the early days, such systems and their use of smart duster hardware lacked functions such as threading and memory protection. However, for most developers, these systems are indispensable for modern operating systems. However, recent versions have enhanced the above operating system features to make embedded systems closer to mainstream platforms.

Emerging trend groups should monitor such operating systems and the procedural and debugging techniques they use because they may be quite different from traditional IT trends.

IV. Internet of Things Power and Storage Technology

Power transmission and storage technologies that may play an important role in the development path of the Internet of Things include:

1. Innovative battery appearance: such as thin film printed batteries and flexible batteries, can be embedded in clothing or Thinergy, LG and other companies have recently displayed products with thin flexible cable batteries. Small-sized batteries are also important for the operation of millimeter-scale systems.

2. High power density batteries: With technologies such as silicon anodes that can improve the performance of lithium batteries, it is expected to improve battery performance and become smaller and lighter.

3. Electricity harvesting: Some systems can “harvest” the environment to recharge smart objects. Examples include electricity generated by mechanical motion such as vibrations, heat sources, solar cells, electrostatic charges, or electromagnetic radiation in the environment. Although electricity production usually generates only a small amount of electricity, it is sufficient for simple sensing and communication objects. MicroStrain has introduced a strain gauge sensor node that uses power generation technology.

4. Wireless charging: Not all objects with sensing and communication capabilities can supply the required power with a replacement battery or electricity. For example, smart clothing is very troublesome to change the battery. In addition, promising IoT applications such as implanting subcutaneous medical sensors cannot replace batteries. Under the sponsorship of the Wireless Power Consortium and its rival organization A4WP, several systems have already started initial production, while others are still in the R&D stage.

In addition, there is another power technology that has not yet developed the full potential for IoT applications. That is supercap, which is a capacitor with a high storage capacity. Although it is generally considered that it will become a substitute for batteries, the existing products have the problem of excessive leakage of electricity. However, emerging trend groups should still pay attention to whether the feasibility of future supercapacitors will increase.

V. Low-power/low-cost small processors

The design of low-power processors has made great gains in the field of simple computing. For example, the Phoenix chip developed by the University of Michigan in the United States consumes only 30 picowatts in sleep mode. Although the technology of the lowest-order 8-bit devices is mature and the processor cost will continue to decline, the cost has begun to reach its peak. Despite this, Gartner estimates that by 2016, 8-bit microprocessors will cost about $0.50.

Researchers have developed small systems that combine thin-film batteries, solar cells for power generation, processors, and simple wireless devices. The volume of such a system can be as small as 1 cubic millimeter, but it may still be several years away from industrial production. If feasible, this kind of millimeter-level computing system will implement concepts such as "smart dust", that is, small-sized sensing nodes that can be configured in large quantities.

Large-scale mass production of devices such as smartphones has reduced the cost of highly integrated and relatively high-end processors, some of which have been widely used in low-cost computing products, such as single-boards with Linux as the operating system for credit card size only ( The single-board device Raspberry Pi costs about $30. Such devices may help to reduce the threshold and produce inexpensive, yet autonomous, complex objects for applications where volume and power consumption are not constrained but where costs are important.