LED Backhaul Project Engineer Blog

What's Li-Fi? (4) Will Li-Fi be used in smartphones? (Part 1)

Last Update: Sep 28th, 2021

---

Introduction.

In my previous article, I wrote that Li-Fi is most likely to spread in applications for high-speed communication in narrow spaces such as HDMI. Then, a question came from within the company, "Won't Li-Fi spread if it is installed in smartphones? Then, someone in our company asked, "Wouldn't Li-Fi spread if it was installed in smartphones? It is true that Li-Fi was originally expected to be installed in smartphones and laptop PCs to play a complementary role to Wi-Fi. A few years ago, there were even rumors that it might be used in the iPhone. There was even a rumor a few years ago that it might be included in the iPhone. Recently, however, such rumors have become less and less common. In this article, as an engineer of optical wireless communication, I would like to consider why Li-Fi is not used in smartphones and laptops, and whether there is a possibility that it will be used in the future.

This time, more than ever, it will completely include the author's "personal predictions as an optical wireless communication engineer". I would like to emphasize once again that this is not the official view of the company(Sangikyo).

The environment surrounding Li-Fi

As explained in the first installment of the Li-Fi series, the name Li-Fi was proposed by a British company called pureLi-Fi, and they are the ones who are leading the Li-Fi industry, including standardization. It is they who are leading the Li-Fi industry, including standardization. OLEDCOMM of France also sells small Li-Fi products, as I have mentioned several times in this blog. In Japan, Outstanding Technology has been developing and selling Li-Fi products for quite some time. We are not aware of all Li-Fi products in the world, but as a person involved in optical wireless communication, I have researched the industry to a certain extent. As you can see from the fact that there are only three companies, the market for Li-Fi (and optical wireless communication in general) is unfortunately very small at the moment. And all Li-Fi devices sold today are "dongle" type, that is, they are external USB devices. This means that you can't attach a dongle to a smartphone^*1. In this day and age of 100% built-in Wi-Fi in laptops, there is absolutely no way that dongle-type products will ever become a big market. In addition to the above, we have also added a new feature that allows users to connect to the Internet from anywhere in the world.

Can the challenges of Li-Fi be solved?

However, as I have repeatedly mentioned in previous articles, Li-Fi still has various technical hurdles to overcome. The biggest one is that it cannot be used outdoors in sunlight, but even if we ignore this premise for indoor use, the following issues remain. Even if we ignore this issue on the assumption of indoor use, the following issues still remain:

1. virtually no out-of-sight communication (NLOS)
2. area is required for gain
3. Maximum speed is slow (compared to Wi-Fi)

I wrote about the issue of 1 in past article. You might say that if you can't use it with NLOS, you can't use it because you hold your phone in different ways. However, since the display side is usually up when the phone is in use, it is not impossible to implement Li-Fi if the transmitter and receiver are exposed on the display side, as shown in Figure 1 below. However, if this becomes possible, there is still the issue of 2. When it comes to light, there is a physical principle that "the size of the lens = the brightness of the reception," just like a camera. Antennas in radio waves are similar, but the higher the frequency, the smaller the antenna, and since they are built-in, meaning they can be mounted inside the case, they can be placed in various places on the board. Li-Fi, on the other hand, needs to be exposed and must be placed on the display or top surface. In today's smartphones, the smaller the bezel width, the better, and even the notch for the front camera is considered "bad design". However, if foldable phones like those sold by Samsung and others become popular, there will be more room for Li-Fi to be implemented.

When it comes to laptop PCs, there is much more room for mounting compared to smartphones. Almost all laptop PCs sold today (including 2-in-1 and tablet PCs) have the web camera mounted on the top of the display (Figure 2), resulting in a wider bezel on the top than on the left and right. It should not be difficult to implement Li-Fi in the upper bezel of the display from a technical and design standpoint.

In other words, the problem of NLOS and installation location is not so big for laptops, while it is not impossible for smartphones, but it is difficult (foldable type is more likely). However, even if the problem of installation location is solved, there is still the problem of slow communication. The Li-Fi currently available on the market is a dongle type, which has a better light receiving area than the built-in type, but it can barely achieve 100Mbps. In this case, there is no reason to bother with built-in Li-Fi, since Wi-Fi can provide better speed unless the conditions are very bad. Regardless of the theoretical maximum speed, if the average speed in normal use is not overwhelmingly faster than Wi-Fi, we cannot expect the emergence of devices with built-in Li-Fi.

There are several reasons for the problem of slow Li-Fi. For example, the backhaul type device (LED backhaul) that we sell has a physical layer speed of over 750Mbps (600Mbps for L2). If this speed can be achieved stably even with Li-Fi, there is a possibility that you will be able to experience "maybe it's faster than Wi-Fi? However, if the maximum speed is over 100Mbps, it will not be faster than Wi-Fi. What is the cause of such a difference in speed for the same optical fiber? The reason is that it is not possible to gain signal strength with the Li-Fi type. In backhaul devices, or devices that communicate one-to-one, the light is "squeezed" by a lens to prevent it from spreading, and the receiver also uses a large lens to focus the light. Li-Fi, on the other hand, requires the light to be spread out in order to fly over a wide area, and the size of the device limits the amount of light that can be focused by the lens. Even so, the sending side is the large size of the parent device, so there are ways to follow up, such as using high power LEDs or lining up many LEDs, but the bottleneck is the receiving side of the child device. For example, our LED backhaul uses a lens with a diameter of 10cm, but if this lens were to be 5mm in diameter (about the size of a smartphone camera), the area would be reduced to 1/400, and the amount of light collected would also be reduced to 1/400. The size of the sensor that can be mounted is also different. In terms of sensors, the image quality of smartphone cameras is getting better and better, and the evolution of light receiving sensors is remarkable. However, no matter how advanced smartphone cameras become, they still cannot compare to SLR cameras in terms of image quality. This is because the lens and sensor of an SLR camera are overwhelmingly larger, and the difference in the amount of light that can be collected will result in a difference in image quality. The same is true for optical communication, where the difference in size results in a difference in communication speed.

Game Changer LD Exists

The story so far is: "There is an implementation problem but a possible solution. But first, what Li-Fi needs is to be faster than Wi-Fi, but current Li-Fi is slow because of the size of the lens and sensor". As mentioned earlier, Li-Fi will theoretically never be faster than the backhaul type. However, it is possible that it will become faster than Wi-Fi by increasing the overall speed of optical wireless communication.

As I wrote in past article, the current speed of optical wireless communication is due to the limitation of the response speed of LEDs. I have also written that the use of laser diodes (LDs, also called semiconductor lasers) as the light source will allow us to break through this limitation and communicate at much higher speeds than before. There are many advantages of LDs, but the biggest advantage is their high response performance: fast on/off blinking and very high linearity (low distortion) of emitted light versus voltage. We are currently working on making the most of this characteristic. Currently, optical fibers** are making the most of these characteristics. LEDs used to be the light source of optical fibers (i.e., the light source of SFPs), but they are now almost entirely replaced by LDs. When the performance required of optical fibers exceeded 1 Gbps, LEDs became insufficient and were replaced by LDs, which enable higher speed communications. In the same way, Li-Fi will follow the same path, and it is quite possible to expect that the replacement of LEDs with LDs will enable communication at speeds exceeding 10 Gbps, just like optical fiber. It's not possible to achieve the same speed as optical fiber in noisy wireless communication, but if we can achieve a stable speed of 1 to 2 Gbps, it will be much faster than Wi-Fi, and we can see the possibility of using Li-Fi in smartphones and laptops. LDs are a game changer for Li-Fi (and optical wireless communication in general).

So why are LDs still not used in Li-Fi? The main reason is the output power issue^*2, Li-Fi needs to illuminate a certain area like lighting. LEDs are already the "mainstay of light sources for lighting", and high-powered ones can be obtained easily and cheaply. LEDs are already the "mainstream" light source for lighting, and high power LEDs are readily and inexpensively available, while there are no LDs that are high enough power and readily and inexpensively available for lighting. For example, BMW makes a laser headlight that uses LDs, and it is available in some high-end cars (as an option). This type of light uses ultraviolet LDs and phosphors to produce white light, so if you want to use it for lighting, you can, but... The price of this laser headlight is almost one million yen, even though it is just a light. It's a "brand price" that can only be installed on high-end BMW cars, but it's still not a price that can be used for Li-Fi.

The current high power LDs are high-end products that are only used in a few applications, but LDs have many advantages not only for communication but also for lighting, such as being able to send light farther^*3, having high color rendering, and being smaller than LEDs. For this reason, LDs are also referred to as next generation light sources, and various companies around the world are working on research and development for the practical application of high power LDs for lighting. We believe that high power and stable LDs will be readily available in the near future. When this happens, Li-Fi will become a product with one or two levels of performance improvement (our backhaul type will also be improved).

Such a high power LD is a dream come true, but whether it will be "I'll beat Wi-Fi in a flash when high power LDs go into mass production! Whether or not this will be the case depends on whether or not the "issues different from performance" will be solved.

To be continued in Part 2.　(Part 2 will be posted on October 11)

---