The pitfalls of low-power wide area networks! Take a look at the stories of Sigfox and LoRa manufacturers

In the past decade of development of the communications industry, it is very important to build a network with broad and deep coverage, because the services provided by operators to users mainly rely on this network. However, after the mobile phone users have reached saturation, facing the blue ocean of the Internet of Things, is it also urgent to build a network with broad and deep coverage first? From the current development situation, the approach of building a network first and then developing applications is full of difficulties. As an industrial economy driven by application orientation, the Internet of Things should be a model of building a network on demand.

[[242303]]

The supply and demand environment has changed fundamentally

"Build the network first, and people will naturally use it." This concept has become a consensus among many people working in the communications industry. The development history of 2G/3G/4G in China has indeed confirmed this view. Of course, this understanding is also due to its historical background. If we regard the communications network as a commodity, we can clearly draw this conclusion from the power comparison between supply and demand. Previously, cellular networks could be built quickly in a short period of time. I believe that this was mainly due to the strong power of the demand side:

[[242304]]

(1) Strong demand drives network construction

For most of the veteran telecommunications professionals who have been in the industry for more than ten years, the golden age they experienced was definitely not the recent years, but the 2G era. Some practitioners recalled that the two major operators previously launched a competition to build networks. After a base station was built and covered in a certain community, the marketing staff followed up with marketing in the local area, and a large number of users joined the network in a short period of time.

It can be seen that in the early days of 2G network popularization, residents' strong demand for mobile communications, coupled with dissatisfaction with the high cost of fixed-line phones, put the mobile communication network in a state of supply exceeding demand. At this time, the operator's job is to build a good network at full capacity and ensure the quality of the network, and users will naturally access it. Even in the 3G/4G era, although the traffic costs were high in the early days, user enthusiasm remained unabated.

(2) The pace of innovation is too fast, and network construction needs to keep up with the pace as soon as possible

The shift from voice networks to data networks has given the industry plenty of room for innovation, especially with the commercialization of 3G/4G networks. Before the commercialization of 3G networks, many people in the industry believed that video calls would become the killer application of the 3G era. However, the emergence of the APP Store model has spawned a variety of consumer applications, and the emergence of killer applications has also changed people's lifestyles to a great extent. Currently, our food, clothing, housing, transportation and other ways have formed new models due to the emergence of mobile Internet.

These mobile Internet innovations have continuously put pressure on the communication network. All innovative applications have generated explosive growth in traffic. Operators have been continuously piped, but they still have to carry out deep network coverage, capacity expansion, and optimization. I remember that in the early days of 3G commercial use, China Mobile could only divert traffic by deploying a large number of WLAN hotspots due to some deficiencies in the TD industry chain and technology, but it still could not meet the traffic demand brought by application innovation. At this time, it was not just "build the network first, and people will naturally use it", but the speed of network construction could not keep up with the pace of application innovation.

However, the supply and demand relationship in IoT applications has changed greatly. On the one hand, the IoT users in various industries of the national economy do not have the same urgent demand for the network as in the 2G/3G/4G era; on the other hand, the fragmentation of various industry applications has not yet formed a small number of large-scale scenarios. Even the small number of large-scale scenarios that have been formed are not comparable to the network traffic and network coverage brought by previous mobile Internet applications. The changes in the supply and demand environment have made the IoT communication network no longer a bottleneck that urgently needs to be broken. The situation of "build the network first, and people will naturally use it" has not appeared in the short term.

Lessons from Sigfox and other LoRa pioneers

[[242305]]

It is not my wishful thinking that the situation of “build the network first and people will naturally use it” will not occur. Many pioneers at home and abroad have given priority to the construction of IoT communication networks, but there has been no large-scale application access for a long time after the network was built.

A typical example is Sigfox. This company has long been a star enterprise in the field of the Internet of Things, and it also attracted people's attention because it received a huge amount of financing in early 2015, becoming an important event in the fermentation of low-power wide area networks (LPWAN) in the field of the Internet of Things.

As we all know, Sigfox adopts the most typical "network construction and road-breaking" model. It has been frantically expanding its territory around the world and has expanded very rapidly. It has currently carried out network deployment in more than 40 countries and regions, and is poised to become the "Vodafone of the Internet of Things."

Such a fast expansion speed has put pressure on other manufacturers that provide low-power wide area networks, such as NB-IoT and LoRa. However, the global race is also a money-burning game, which requires continuous financial support. By the end of 2017, Sigfox had completed six rounds of financing, as shown in the following table:

Sigfox financing situation (Source: crunchbase)

In addition to the French government fund Bpifrance, the Malaysian government fund Khazanah Nasional, and the World Bank fund IFC, the investors also include industrial investment funds such as Intel and Samsung, telecom operators such as SK Telecom, Telefonica, and Docomo, as well as LPWAN user companies such as Total, French Liquefied Gas, and French Gas. It can be said that the investment lineup is very luxurious.

Based on this calculation, the disclosed financing amount has reached 277 million euros. Previously, the president of Sigfox Asia Pacific revealed that Sigfox raised another 150 million euros in 2017. Based on this, it is estimated that it is likely to come from the investment of the Malaysian government fund Khazanah Nasional, so the completed financing is as high as 427 million euros.

However, the huge amount of 427 million euros in financing cannot meet Sigfox's global network deployment needs. According to foreign media Light Reading, Sigfox hopes to raise another 40 million euros from existing shareholders this year. Along with the global expansion and network construction, there are constant rumors that Sigfox is in trouble, a typical example of which is the large number of its executives leaving, including the recently resigned CFO.

Although networks have been deployed in more than 40 countries and regions, there has been no large-scale access to applications. At the beginning of 2018, the number of terminals connected to Sigfox's global network was only 2.5 million. Sigfox had previously planned to achieve 10 million connections in 2018, but quickly revised this goal to 6 million a few days later. In terms of revenue, Sigfox announced that its revenue in 2016 was 30 million euros and reached 50 million euros in 2017. However, foreign media commented that this figure was lower than its expected target of more than 60 million euros, so the original IPO planned for 2018 was postponed. Previously, Sigfox was hotly discussed for its business model of charging 1 euro/year for each terminal. At present, it is far more than that, including revenue from software platforms, licensing fees, etc. However, compared with its hundreds of millions of euros in financing and burning money in recent years, the growth rate of these revenues is still too slow.

[[242306]]

It is still hard to say how Sigfox will develop in the future, but it can be seen that the company has further strengthened its exploration of applications and cooperated with various industry partners to promote some application innovations, rather than just preempting network deployment. It can be said that Sigfox's development in recent years has verified that the concept of "build a good network first, and people will naturally use it" is difficult to apply in the field of the Internet of Things.

Coincidentally, many domestic manufacturers have also tried to build a network in a small area in the past few years, but it turned out that this decision was a mistake. In these attempts, many companies basically deployed a LoRa network in their cities. Although only dozens of LoRa base stations were built, they had to pay tower rental, power, manpower maintenance and other costs every month. There were not many applications connected for a long time, and the operating income was extremely mismatched with the cost. Therefore, many manufacturers eventually dismantled the LoRa base stations deployed on the towers and only retained the base stations with application access.

It can be seen that the network is an important factor in the IoT solution, but it is not the most urgent factor to be solved. At present, the deployment of various LPWAN networks is very flexible and the deployment cycle is very short. When there is a clear demand and large-scale application, and a healthy business model can be formed, it is not too late to build the network on demand. Of course, traditional telecom operators still first achieve network coverage when facing the IoT. This method has its rationality among the operator group. If other manufacturers directly imitate it, they will get themselves into trouble.