According to the 3GPP 38.101 protocol, 5G NR mainly uses two frequency bands: FR1 band and FR2 band. The frequency range of FR1 band is 450MHz-6GHz, also known as sub 6GHz band; the frequency range of FR2 band is 24.25GHz-52.6GHz, which is usually called millimeter wave (mmWave). Strictly speaking, millimeter wave (mmWave) can only refer to the EHF band, that is, electromagnetic waves with a frequency range of 30GHz-300GHz. Because the wavelength of 30GHz electromagnetic waves is 10mm, and the wavelength of 300GHz electromagnetic waves is 1mm. The wavelength of 24.25GHz electromagnetic waves is 12.37mm, so it can be called millimeter wave or centimeter wave. But in fact, millimeter wave is just a conventional name, and no organization has ever clearly defined it. Therefore, some people believe that electromagnetic waves with a frequency range of 20GHz (wavelength 15mm) to 300GHz can be considered millimeter waves. For a long period of history, the millimeter wave band belonged to the wilderness. Why? The reason is simple, because there are almost no electronic components or devices that can send or receive millimeter waves. Why are there no electronic devices that send or receive millimeter waves? There are two reasons. The first reason is that millimeter waves are not practical. Although millimeter waves can provide larger bandwidth and higher data rates, previous mobile applications did not require such a large bandwidth and such a high data rate, and there was no market demand for millimeter waves. Millimeter waves also have some obvious limitations, such as too much propagation loss and too small coverage. The second reason is that millimeter waves are too expensive. It has always been a challenge to produce sub-micron integrated circuit components that can work in the millimeter wave band. Overcoming propagation loss and improving coverage also means a lot of money. However, everything has changed in the past decade or so. With the rapid development of mobile communications, frequency resources within 30GHz are almost exhausted. Governments and international standardization organizations have allocated all the "good" frequencies, but there are still frequency shortages and frequency conflicts. The development of 4G cellular systems and the upcoming 5G all rely on the proper allocation of frequencies. The problem is, there are almost no frequencies left. Now, frequencies are like houses, which can be described in one word: "expensive"! For houses, the first is location, the second is location, and the third is location. The same description applies to wireless frequencies. Millimeter waves are like the New World of America, providing mobile users and mobile operators with "endless" frequency resources. You can fill all the sub 30GHz bands we use now into the lower end of the millimeter wave band, and there are still at least 240GHz of vacant frequency. Millimeter waves bring large bandwidth and high speed. The maximum bandwidth that 4G LTE cellular systems based on the sub 6GHz band can use is 100MHz, and the data rate does not exceed 1Gbps. In the millimeter wave band, the maximum bandwidth that mobile applications can use is 400MHz, and the data rate is as high as 10Gbps or even more. Demand is always the biggest driving force for innovation. The technical difficulties of producing high-quality and low-cost millimeter-wave frequency integrated circuit components have been quickly overcome. By using new materials such as SiGe, GaAs, InP, GaN, and new production processes, transistors as small as tens or even a few nanometers have been integrated on chips working in the millimeter-wave band, greatly reducing costs. Can we now use any millimeter wave between 20GHz and 300GHz at will? Not yet. Some people divide the commonly used millimeter wave bands into four segments: Ka band 26.5GHz~40GHz; Q band: 33GHz~50GHz; V band: 50GHz~70GHz; W band: 75GHz~110GHz. 3GPP protocol 38.101-2 Table 5.2-1 defines three frequency bands for 5G NR FR2 bands, namely: n257 (26.5GHz~29.5GHz), n258 (24.25GHz~27.5GHz) and n260 (37GHz~40GHz), all of which use TDD standard. The US FCC recommends that 5G NR use the following frequency bands: 24-25 GHz (24.25-24.45/24.75-25.25 GHz), 32GHz (31.8-33.4 GHz), 42 GHz (42-42.5 GHz), 48 GHz (47.2-50.2 GHz), 51 GHz (50.4-52.6GHz), 70 GHz (71-76 GHz) and 80 GHz (81-86 GHz), and also recommends studying the use of frequencies higher than 95 GHz to carry 5G. Why can't millimeter wave frequencies be used at will? In addition to the consideration of large-scale economic benefits, some frequencies in millimeter waves have particularly bad "locations". Here, the factor affecting the "location" is the air, so to be precise, the "sky section" of these frequencies is particularly bad. When radio waves propagate, the atmosphere selectively absorbs electromagnetic waves of certain frequencies (wavelengths), causing particularly serious propagation losses of these electromagnetic waves. The main components of the atmosphere that absorb electromagnetic waves are oxygen and water vapor. The resonance caused by water vapor absorbs electromagnetic waves near 22GHz and 183GHz, while the resonance absorption of oxygen affects electromagnetic waves near 60GHz and 120GHz. So we can see that no matter which organization allocates millimeter wave resources, they will avoid the frequency bands near these four frequencies. Due to the technical difficulty, millimeter waves above 95GHz are not considered for the time being. In addition to this "sky segment" factor that can only be avoided, we can only face other limitations of millimeter waves and find ways to overcome them. Otherwise, millimeter waves cannot be used. One of the most critical limitations is that the propagation distance of millimeter waves is really limited. The laws of physics tell us that the shorter the wavelength, the shorter the propagation distance when the transmission power remains unchanged. In many scenarios, this limitation will result in the propagation distance of millimeter waves not exceeding 10 meters. According to the idealized free space propagation loss formula, the propagation loss L=92.4+20log(f)+20log(R), where f is the frequency in GHz, R is the distance in kilometers, and L is in dB. After a 70GHz millimeter wave propagates 10 meters, the loss reaches 89.3dB. Under non-ideal propagation conditions, the propagation loss is much greater. Developers of millimeter wave systems must compensate for such large propagation losses by increasing transmit power, antenna gain, and receiving sensitivity. Everything has two sides. The short propagation distance can sometimes be an advantage of millimeter wave systems. For example, it can reduce interference between millimeter wave signals. The high-gain antennas used in millimeter wave systems also have good directivity, which further eliminates interference. Such narrow beam antennas not only increase power and expand coverage, but also enhance security and reduce the probability of signal interception. [font=-apple-system,BlinkMacSystemFont,Helvetica Neue,PingFang SC,Microsoft YaHei,Source Han Sans SC,Noto Sans CJK SC,WenQuanYi Micro Hei,[sans-serif] In addition, the limiting factor of "high frequency" will reduce the size of the antenna, which is another unexpected surprise. Assuming that the size of the antenna we use is fixed relative to the wireless wavelength, such as 1/2 wavelength or 1/4 wavelength, then the increase in carrier frequency means that the antenna becomes smaller and smaller. For example, the length of a 900M GSM antenna is about tens of centimeters, while the millimeter wave antenna may be only a few millimeters. This means that in the same space, we can cram more and more high-frequency band antennas. Based on this fact, we can compensate for the high-frequency path loss by increasing the number of antennas without increasing the size of the antenna array. This makes it possible to use massive MIMO technology in 5G millimeter wave systems. After overcoming these limitations, the 5G system working on millimeter waves can provide many services that 4G cannot provide, such as high-definition video, virtual reality, augmented reality, wireless base station backhaul, short-range radar detection, information services in dense urban areas, wireless communication services in stadiums/concerts/shopping malls, factory automation control, telemedicine, security monitoring, intelligent transportation systems, airport security checks, etc. The development and utilization of millimeter wave bands provide a broad space and unlimited imagination for 5G applications. Since 3GPP decided that 5G NR will continue to use OFDM technology, compared with 4G, 5G does not have a disruptive technological innovation, and millimeter wave has almost become the biggest "novelty" of 5G. The introduction of other new technologies in 5G, such as massive MIMO, new numerology (subcarrier spacing, etc.), LDPC/Polar codes, etc., are closely related to millimeter waves, all of which are to allow OFDM technology to be better extended to the millimeter wave band. In order to adapt to the large bandwidth characteristics of millimeter waves, 5G defines multiple subcarrier spacings, among which the larger subcarrier spacings (60KHz and 120KHz) are specially designed for millimeter waves. The massive MIMO technology mentioned above is also tailored for millimeter waves. Therefore, 5G can also be called "enhanced 4G extended to millimeter waves" or "enhanced LTE extended to millimeter waves." If one day millimeter waves are also congested, how should mobile communication systems expand new territories? If the wavelength is less than 1 mm, it enters the wavelength range of light (the wavelength range of the infrared band is 0.76 microns to 1 mm). Transistors above 100 GHz have been developed in the laboratory. However, this type of transistor is basically useless at around 300 GHz. So what electronic components should be used? Infrared rays work at 150 THz to 430 THz, visible light works at 430 THz to 750 THz, and ultraviolet rays work above 740 GHz. Laser devices, LEDs, and diodes can generate and detect these lights. However, these devices cannot work in the frequency range of 300 GHz to 100 THz. This frequency range seems to have become a blind spot at present. However, this phenomenon is temporary. As long as there is demand, new technologies and new components will definitely eliminate this blind spot. [sans-serif] Since 3GPP decided that 5G NR will continue to use OFDM technology, compared with 4G, 5G does not have any disruptive technological innovation, and millimeter wave has almost become the biggest "novelty" of 5G. The introduction of other new technologies in 5G, such as massive MIMO, new numerology (subcarrier spacing, etc.), LDPC/Polar codes, etc., are closely related to millimeter waves, all of which are to allow OFDM technology to be better extended to the millimeter wave band. In order to adapt to the large bandwidth characteristics of millimeter waves, 5G defines multiple subcarrier spacings, among which the larger subcarrier spacings (60KHz and 120KHz) are specially designed for millimeter waves. The massive MIMO technology mentioned above is also tailored for millimeter waves. Therefore, 5G can also be called "enhanced 4G extended to millimeter waves" or "enhanced LTE extended to millimeter waves." If one day millimeter waves are also congested, how can mobile communication systems expand new territories? If the wavelength is less than 1 mm, it enters the wavelength range of light (the wavelength range of the infrared band is 0.76 microns to 1 mm). Transistors above 100 GHz have been developed in the laboratory. However, this type of transistor is basically useless at around 300 GHz. So what electronic components should be used? Infrared works at 150THz~430THz, visible light works at 430THz~750THz, and ultraviolet works above 740GHz. Laser devices, LEDs, and diodes can generate and detect these lights. However, these devices cannot work in the frequency range of 300GHz~100THz. This frequency range seems to have become a blind spot at present. However, this phenomenon is temporary. As long as there is demand, new technologies and new components will definitely eliminate this blind spot. [sans-serif] Since 3GPP decided that 5G NR will continue to use OFDM technology, compared with 4G, 5G does not have any disruptive technological innovation, and millimeter wave has almost become the biggest "novelty" of 5G. The introduction of other new technologies in 5G, such as massive MIMO, new numerology (subcarrier spacing, etc.), LDPC/Polar codes, etc., are closely related to millimeter waves, all of which are to allow OFDM technology to be better extended to the millimeter wave band. In order to adapt to the large bandwidth characteristics of millimeter waves, 5G defines multiple subcarrier spacings, among which the larger subcarrier spacings (60KHz and 120KHz) are specially designed for millimeter waves. The massive MIMO technology mentioned above is also tailored for millimeter waves. Therefore, 5G can also be called "enhanced 4G extended to millimeter waves" or "enhanced LTE extended to millimeter waves." If one day millimeter waves are also congested, how can mobile communication systems expand new territories? If the wavelength is less than 1 mm, it enters the wavelength range of light (the wavelength range of the infrared band is 0.76 microns to 1 mm). Transistors above 100 GHz have been developed in the laboratory. However, this type of transistor is basically useless at around 300 GHz. So what electronic components should be used? Infrared works at 150THz~430THz, visible light works at 430THz~750THz, and ultraviolet works above 740GHz. Laser devices, LEDs, and diodes can generate and detect these lights. However, these devices cannot work in the frequency range of 300GHz~100THz. This frequency range seems to have become a blind spot at present. However, this phenomenon is temporary. As long as there is demand, new technologies and new components will definitely eliminate this blind spot.