The "concept" of the Internet of Things is in full swing. As a university that provides the effective strength of the industry, it is naturally unwilling to be lonely. As a result, various schools are preparing to set up and set up the Internet of Things major. Judging from the information from various channels, it seems that they want to impart a variety of knowledge, focusing on sensing, communication and an application field, such as smart grid and intelligent transportation. There seems to be a relatively unified direction, unlike the previous professional settings. In this case, first of all, we should make it clear that the Internet of Things should not be a traditional undergraduate major, at least it is not suitable to exist as an undergraduate major! In reality, many fields have this feature, which has a wide range of needs and is pregnant with great opportunities. However, because the knowledge involved is interdisciplinary and needs all kinds of basic and professional knowledge as a foreshadowing, I am a little reluctant to be an undergraduate major. I heard a long time ago that Ocean University should not recruit high school students, but should select them from undergraduate graduates who have studied related majors, because ocean research needs all kinds of professional knowledge, belongs to two disciplines, and needs some basic professional knowledge background, such as biology, chemistry, fluid mechanics and so on. High school students don't have it. If they want to study these things in the four years of Ocean College, it is estimated that they have little time to teach the graduates about the ocean. Compared with today's Internet of Things, it seems that there is also the shadow of "Ocean College". If the Internet of Things is promoted to the height advertised at present, it seems that it should also exist as two disciplines, and cultivate a talent who can cope with such a "vast" field. First of all, these people should learn some basic professional knowledge, such as measurement, sensing, communication, large-scale calculation and statistics. And then enter the Internet of Things major. Combining with my own professional knowledge background, I focus on learning a branch of the Internet of Things, such as designing a measurement part based on networking requirements, designing communication links and protocols based on measurement characteristics, or studying how to store and analyze the massive information collected by the "cloud" to make it effective. These are not directly faced by a high school student, but as the basic knowledge of the Internet of Things, time is not enough to teach the required professional knowledge. Originally, these were all four-year undergraduate courses. After learning these, the time left for imparting knowledge about the Internet of Things is limited, and only a student with no characteristics can be cultivated. Isn't this contrary to the original intention of setting up this major? Therefore, in order to cultivate real talents in the Internet of Things, it is difficult to be competent by four years of undergraduate study! The so-called "Internet of Things talents" in the true sense, according to my understanding, should be: they can bring the current real needs into the Internet of Things, thus creating revolutionary changes different from traditional solutions. It sounds a bit mysterious, but the current publicity is just so much, or even worse. Teng "cloud" driving "thing" is the most classic description seen in the media! When will it land? It seems that it is not ready yet. How should schools respond to the current Internet of Things craze? In my humble opinion: I don't want to train those "real Internet of Things talents" for the time being. Those future leaders cannot be trained by undergraduate education. As a large number of technical backbones to the society, engineering colleges should focus on how to cultivate a technician who can meet the needs of the Internet of Things in future work. To do this, we must first analyze what new things have been derived from the Internet of Things. The word "internet of things" actually expresses its meaning: first, things are the object of concern; Second, contact is the means used; Third, the network is a tool to solve problems. In these three parts, things have existed for a long time, and people can say that they are "concerned about things" and are still doing so. Although the "net" is younger, due to the progress of technical means, people have given full play to its functions, and the "net" has brought great changes to people's lives. The only new thing brought by the Internet of Things is "connection", that is, connecting "things" with "network"! There are many devices connected to the Internet, so why is it new to connect things to the Internet at this time? Let's first analyze the characteristics of previous networking devices. At present, networking devices are mainly computers, mobile phones, televisions and so on. , as well as some behind-the-scenes devices, such as switches and routers, which are characterized by interacting with people or serving for interacting with people. Therefore, the internet in the past can also be called "human networking"! The "things" referred to by the current "Internet of Things" are mostly things facing nature, focusing on their biological, physical and chemical characteristics, and some things without intelligence. Even if it is related to people, it pays attention to people's biological characteristics rather than intelligence, so it is completely different from the content carried by the existing "Internet". At present, most of the information on the Internet is in the service of intelligence and belongs to the "superstructure", while the Internet of Things should pay attention to the "economic foundation". Because these "things" are relatively "low-energy", it is the essence of the Internet of Things to change the past networking mode accordingly to adapt to its characteristics. There are thousands of "things", all of which have corresponding specialties to explore, and "Web" also has corresponding branches to improve. As for data analysis, cloud computing, distributed computing, parallel computing, etc., it has long been a subject of in-depth research. In recent years, the latter two have been continuously integrated and improved, and great progress has been made. It's just that the former and the latter look a bit like parallel lines, and there is no intersection for the time being. The internet of things is to build a bridge between these two parallel lines, so that the information of the former can be sublimated with the help of the latter! However, due to the "low energy" and "diversity" characteristics of "things", the original networking means are no longer suitable, and it is this absence that has spawned the demand for Internet of Things teaching! Therefore, exploring networking methods suitable for all kinds of "things" should be the focus of cultivating an engineer who can meet the needs of the Internet of Things. Although networking should be the category of communication, the focus of communication in the past was media, links and protocols. As the application requirement of the Internet of Things, we should pay attention to the characteristics of various communication modes, and choose the appropriate communication scheme focusing on "selection" in combination with the personality of the "things" needed for networking. To make a suitable choice, we must accurately grasp the characteristics of various communication modes, foster strengths and avoid weaknesses, and rationally combine them, so as to achieve the most reliable and economical scheme. Not to mention that the Internet of Things is ZigBee and Wifi. In fact, each communication method has its advantages and disadvantages and applicable environment. ZigBee is not necessarily more suitable for detection on smart grid than "carrier"; It may be more economical and reliable for many nodes of smart home to be centralized through a simple "485 bus" and then transmitted to the access network through WiFi or ZigBee. Therefore, it is the core quality of a qualified Internet of Things application technician to accurately grasp the object characteristics and application scenarios and be familiar with the advantages and disadvantages of various communication technologies. As for the high-level application of the Internet of Things, that is, how to deal with those "clouds" that collect massive data? How to generate benefits? That's a matter for a few elites, which is the basis for this pyramid to need a large number of technicians. Specifically, according to the application requirements of the Internet of Things, we should pay attention to: 1 and understand the most basic communication principles, especially serial communication, because parallel communication is basically unnecessary as long as the distance is slightly longer. However, all kinds of communication in reality, whether it is wired RS232, 485, USB, Ethernet, infrared, Bluetooth or Wifi, are essentially serial communication, and the difference is only in media and protocols. 2. Through the simplest UART communication, understand the common problems in communication such as handshake, conflict, contention and arbitration, as well as the reliability problems such as frame loss and error code, and understand the function and significance of communication protocols. 3. Understand the needs of various scenarios through simple practical applications, and when to speed up? When is it reliable? When should I respond quickly? When do you want low power consumption? 4. According to the demand, try all kinds of communication, as long as you can use it, because with the ubiquity of single chip microcomputer, the use of various communication modules is becoming more and more simple. The seemingly mysterious 3G communication can also work like an ancient modem. For most people, it can only be used. Leave the design and production of this module to a few people on earth to worry about. Just like PC, it is used all over the world, but only Intel and AMD worry about it, and it doesn't seem to delay the popularization and use of PC! 5. Choose a branch to run through all levels systematically. It doesn't matter whether it is practical or not, as long as it covers the elements of the Internet of Things. It is important to feel the problems encountered after things are networked through actual experience. In a word, I think it is very important to cultivate an applied technician by using the existing professional knowledge and practicing around the above-mentioned key points, because the future applications are diverse and unpredictable, so it is very important to cultivate the ability to grasp the characteristics of the object. Finally, I recommend a book about all kinds of communication: development and application of embedded network communication, published by Beihang University, ISBN 978-7-5124-0179-2. Faced with so many communication modes, it is not easy to fully understand them. Many books only describe some of them, but this book comprehensively collects all kinds of existing communication methods. In order to choose the appropriate communication method, please refer to the detailed catalogue in "Good Book Communication".