Hongyun Project -- China's low-earth orbit broadband Internet project

It might be tempting to dismiss this effort as small and behind the broadband satellite projects of companies like SpaceX, OneWeb and Telesat, but that would be a mistake.

Long March 11 rocket and Hongyun-1
satellite (source).

Last December, State-owned China Aerospace Science and Industry Corporation (CASIC) launched the first experimental Hongyun (rainbow cloud) Project satellite and they began testing it in March.

The 247 kg test satellite is in orbit at an altitude of around 1,100 km and they plan to launch four more test satellites this year and begin operating with a 156-satellite constellation in 2022. I don't know anything more about their plans, but with only 156 satellites I suspect they will focus on unserved regions in rural China and perhaps Latin America at first.

It might be tempting to dismiss this effort as small and behind the broadband satellite projects of companies like SpaceX, OneWeb and Telesat, but that would be a mistake. China has an ambitious, global Internet infrastructure and application program called the Digital Silk Road and the "road" is terrestrial with highways, ports, pipelines, and railways, undersea with cables and in space with the Hongyun Project, their Beidou satellite navigation system, which will be global next year, and the Digital Belt and Road Earth observation program. Our withdrawal from the Trans-Pacific Partnership and the current trade war were gifts to the Chinese.

(Other early short articles on the LEO project here and here).

Update 6/4/2019

CASIC broke ground on April 24 for a satellite industry park in Wuhan, Central China's Hubei Province, where they will produce satellites for the Hongyun project.

In keeping with China's policy of funding competitors, another production line operated by a satellite start-up, Spacety, based in Changsha, Central China's Hunan Province, began construction in January. Each facility is expected to produce 100 satellites per year. (China has historically funded Internet service competition).

Update 6/18/2019

U.S. military tracking data shows the satellite is in a nearly circular orbit averaging 1,067 km altitude at an inclination of 99.9 degrees and CASIC confirmed that Hongyun would emphasize service in China's remote regions.

Update 11/26/2019

Speaking at a conference last week, CASIC general manager Zou Guangbao confirmed their planned schedule and said they would serve the broadband communication, navigation & remote sensing markets in China and elsewhere. They are also developing a separate constellation of 80 Internet of things satellites

Update 12/14/2019

GalaxySpace is a second Chinese company working on a LEO broadband constellation. Their first satellite is under construction. The 200 kg satellite will have 10 Gbps capacity, orbit at 1,200 km with a 300,000 square km footprint and use high-frequency Q/V band radio.


Update 12/18/2019

Hongyun has expanded its broadband satellite plan. They are now working toward 864 satellites orbiting at 1,175 km with an 8 Terabytes per second capacity. They hope to serve 2 million 5G users through direct connections to base stations, 200,000 broadband users and 10 million Internet of things users. The focus will be on China and Belt and Road nations.

Update 12/21/2019

Liu Shiquan, Deputy General Manager of CASIC Hongyun satellite reported that Hongyun performance and function tests have been completed. He did not give details but said the tests included Web browsing, video chat, and high-resolution streaming.

Liu also gave a few schedule hints. The post quotes him as saying they would launch four more test satellites "by 2020" but I assume that meant "during 2020." He also said that by the beginning of 2020, users across China will be able to access the demonstration system. There was no elaboration on this, but I assume he is referring to a few test users. He also said they plan to have 156 satellites in operation by the middle of the 14th Five-Year Plan (2021-25).

I wonder if more detailed information is available on the Web in China.

Update 1/6/2020

The Hongyan (Wild Goose) project plans a constellation of around 320 LEO satellites. They have launched one test satellite so far and had hoped to launch 8 more by 2020, but did not make that deadline. They expect to have 60 satellites in orbit and operating around 2023 will be able to provide global coverage with the full constellation by 2025.

As shown in this illustration, they plan to connect buildings, ships, trains, and planes and to provide mobile backhaul and, most interestingly direct service to mobile phones. He Mu, Hongyan Application Director, promised the development of a "chip [that] can be integrated into the mobile phone so that everyone holding an ordinary mobile phone will have access to seamless satellite telecommunication with global coverage." (If this happens, it will be interesting to see how they differentiate this from terrestrial mobile service).

CASIC's Five Clouds

Hongyan is a project of China Aerospace Science and Technology Corporation (CASC) while Hongyun is a project of the China Aerospace Science and Industry Corporation (CASIC). CASIC has four other "five clouds" projects underway in addition to Hongyun:

• Feiyun, using solar-powered drones
• Kuaiyun, using near-space airships (dirigibles?)
• Tengyun, a project to develop reusable space plane.
• Xingyun, an 80-LEO narrowband IOT constellation using cubsats, the first of which has been launched.

Characteristics of Hongyun and Hongyan satellites

Update 1/20/2020

GalaxySpace has launched a LEO "5G" satellite, Yinhe-1, which is expected to test Q/V and Ka-band communications at up to 10 Gbps. I'm not sure what a "5G" satellite is, but note that the above diagram shows a satellite communicating directly with a mobile phone, as opposed to a mobile tower. Check out this short video on the satellite and launch:

Update 1/30/2020

See this post for some discussion of Chinese space policy.

Update 2/25/2021

Space Engineering Development (SED), a subsidiary of China Aerospace Science and Industry Corporation (CASIC), has begun testing its satellite manufacturing infrastructure and is expected to begin production in March. The plant will manufacture the Hongyun satellites and will be capable of producing about 240 satellites annually. Hongyun plans a 156-satellites constellation, so they should be ready next year when they plan to begin operation.

China will be a formidable satellite Internet service competitor

In a study of the Internet in China in the late 1990s, my colleagues and I observed that "China has been able to execute plans effectively by allocating resources to competing, government-owned enterprises," and Kai-Fu Lee shows that they have pursued a similar strategy with respect to AI. Now they are doing the same with low-Earth orbit (LEO) broadband satellite constellations.

Characteristics of the Hongyun and Hongyan test satellites

Last December, state-owned China Aerospace Science and Industry Corporation (CASIC) launched their first experimental Hongyun (rainbow cloud) Project satellite and a week later China Aerospace Science and Technology Corporation (CASC) launched their first experimental Hongyan (wild goose) Project satellite. (Both CASIC and CASC have Wikipeida pages and their Fortune Global 500 ranks are 322 and 323).

As shown here, Hongyun launched a test satellite in December 2018 and said they planned four more during 2019, but there is no record of those having been launched as of today. They have, however, completed tests of Web browsing, video chat, and high-resolution streaming and said users across China would be able to access the demonstration system. (I assume that refers to test users).

They initially planned to begin operating with 156 satellites by the middle of the 14th Five-Year Plan (2021-25), emphasizing service in China's remote regions. Late last year, they expanded the constellation plan from 156 to 864 satellites orbiting at 1,175 km with an 8 Terabytes per second capacity. They hope to serve 2 million 5G users through direct connections to base stations, 200,000 broadband users and 10 million Internet of things (IoT) users. The focus will be on China and Belt and Road nations.

CASIC has also established two satellite factories in Hubei and Hunan provinces. This may have been necessitated by the increase in the number of planned Hongyun satellites or it may be another application of the strategy of creating competing state-owned enterprises.

Hongyan applications (source)

CASC's Hongyan project plans a constellation of around 320 LEO satellites. They have launched one test satellite so far and had hoped to launch 8 more by 2020, but did not make that deadline. They expect to have 60 operating satellites "around" 2023 and to be able to provide global coverage with the full constellation by 2025.

As shown in this illustration, they plan to connect buildings, ships, trains, and planes and to provide mobile backhaul and, most interestingly, direct service to mobile phones. He Mu, Hongyan Application Director, promised the development of a "chip [that] can be integrated into the mobile phone so that everyone holding an ordinary mobile phone will have access to seamless satellite telecommunication with global coverage." That does not sound like a mobile connection to a base station with satellite backhaul, but neither does it sound possible.

Earlier this month a third competitor, GalaxySpace, launched Yinhe-1, which is expected to test Q/V and Ka-band communications at up to 10 Gbps. They refer to Yinhe-1 as a "5G satellite." I'm not sure what a "5G" satellite is, but note again that the above diagram shows a satellite communicating directly with a mobile phone, as opposed to a mobile tower. Check out this short video on the satellite and launch:

CASIC has four other "five clouds" projects underway in addition to Hongyun: Feiyun, using solar-powered drones, Kuaiyun, using near-space airships (dirigibles?), Tengyun, a project to develop a reusable space plane, and Xingyun, an 80-LEO narrowband IoT constellation using cubsats, the first of which is to be launched soon.

As noted above, Chinese state-owned enterprises often compete with each other, but they also cooperate. For example, CASIC's Hongun-1 was launched on a CASC rocket. (I wonder how they arrived at the launch price). Will Hongyun and Hongan exchange traffic at shared ground stations? Will their satellites one day intercommunicate in order to optimize a joint constellation with different orbits? Will they intercommunicate with China's geostationary satellites and other space assets?

It is often argued that government ownership and subsidy are unfair to competitors and lead to a suboptimal allocation of resources. I assume this sort of government-brokered "coordinated competition" is more common in China than in the US, but even here, the lines between government-sponsored research and development, government procurement and industrial subsidy are a bit vague as are the criteria for anti-trust enforcement. People and organizations will learn to game either system, so both must be dynamic.

Hongyun, Hongyan and GalaxySpace are late to the game. OneWeb, SpaceX, and Telesat are beginning to sign up customers and will launch a lot of satellites this year. Amazon is also a late-comer, but they have a lot of money and complementary infrastructure. Like Amazon, China has funds for the long run, domestic infrastructure which can be shared by the three LEO projects and they are working on reusability. Furthermore, they have a political advantage in the "Digital Silk Road" nations of our increasingly divided world and divided Internet. China will be a formidable satellite Internet service competitor.

Update 4/26/2020

The US military has tracked the Galaxy satellite launched in January in a 637 by 621-kilometer altitude orbit inclined at 86 degrees. On April 23, Galaxy engineers conducted a three-minute video call via a WiFi hotspot that used the satellite for backhaul.

As noted above, Galaxy has referred to this as a "5G" satellite constellation and we speculated that they may have been implying direct connections from cell phones to orbiting 5G base stations, but these test results confirm and @Megaconstelai points out, they are using the satellite for backhaul from terrestrial 5G cells.

Update 5/26/2020

A few Hongyun test results are in. The brief note speaks of spectral thermometer tests and image data. There was no mention of broadband Internet tests.

Update 9/8/2020

A Galaxy Space test achieved 1 gigabit per second for each of its 16 Ka-band user beams and Blaine Curcio says 2020 is a turning point for Chinese commercial space. SpaceX and others are racing ahead and China has a strategic -- military, political, and commercial -- interest in keeping up with the West/US in space. The Digital Silk Road runs through space.

Update 9/20/2020 

GalaxySpace is building a factory to mass-produce low-cost satellites in East China. They will manufacture "more than one" satellite per day, joining SpaceX (four per day) and OneWeb (three per day) as mass-production LEO broadband satellite manufacturers.

Their current test satellite has verified high-frequency Q/V/Ka and other frequency band communications and they have achieved 10 Gps speed within a 300,000 Km^2 footprint. They also succeeded in a 5G wireless test, which I assume means they communicated through a 5G base station.

In a related development, China's National Development and Reform Commission added satellite broadband, 5G, and the Internet of Things to it's “New Infrastructures” list. That means more money will be invested in these technologies.

Consultant Blaine Curcio speculates that the Chinese government assumes increased importance of the Internet in a post-COVID world and rather than limiting the investment effort to the state-owned enterprises, the government has opened the doors to the private sector as well.

Update 1/19/2021

Construction of China's first smart manufacturing plant for satellites has been completed. Production will begin in March and the first product will be Hongyun broadband Internet satellites. The factory will be capable of making 240 small satellites a year and it sounds like it can be repurposed to build satellites for other constellations. SpaceX and OneWeb both claim to be able to produce satellites faster, but their factories may not be as easily reconfigured. Regardless, the Chinese space industry is progressing rapidly.

Chinese LEO satellite Internet update

China has pursued a strategy of competition among government-owned organizations, and it initiated two government-owned constellation projects, Hongyun and Hongyan, in 2018. In April 2020, China’s National Development and Reform Commission included “satellite internet” on its “new infrastructures” list, and China applied to the ITU for a new constellation, called  GW. Hongyun and Hongyan were dropped, and GW, also called China Satnet or Guowang, emerged as China’s global Internet service provider and it was followed by two others, Qianfan and Honghu-3. These are all far behind Starlink, but they will have protected markets, and China is developing new launch vehicles and satellite manufacturing capacity.

Gwowang

China’s 14th Five-Year Plan (2021-2025) for National Economic and Social Development and Long-Range Objectives Through the Year 2035 called for building an integrated communications, Earth observation, and satellite navigation system with global coverage. Guowang is the constellation they called for.

Guowang consists of two sub-constellations, designated GW-A59 (6,080 satellites) and GW-2 (6,912 satellites). GW-2 will orbit at 1,145 km, and  GW-A59 will orbit around half that. The ITU filing was in September of 2020, and after a long delay, the first ten GW-2 satellites were launched at the end of 2024, and they now have 81 in orbit. The cadence has picked up recently -- China just launched another batch of Guowang satellites. This was the ninth Guowang launch this year and the sixth in the last 30 days. Even at this cadence, it is unclear that they can manufacture and launch enough satellites to meet the ITU launch deadlines. Perhaps the Chinese have decided that, given launch and manufacturing resources, they would not be able to meet ITU deadlines for all of their constellations, so they are focusing on Guowang, which can be seen as most critical for the government.

Little technical information is available, but considering the capacities of the various rockets used to launch Guowang satellites and the number of satellites in each launch, it seems there are two sizes of satellite: large satellites of around 16,600 kg and smaller satellites of around 889 kg. While these are imprecise estimates, they indicate two classes of satellite with different capabilities and functions. (Note that the relatively high altitude GW-2 sub-constellation has both large and small satellites).

Several Guowang test satellites have also been launched, suggesting strategic government and military applications like Signals intelligence, positioning, navigation, and imaging applications in addition to Internet service.

Qianfan

Shanghai Spacecom Satellite Technology (SSST), a private company backed by the Shanghai municipal government and the Chinese Academy of Sciences, is developing the Qianfan constellation. The planned satellites will orbit at 1,160 km, which is higher than the other announced LEO satellite competitors except Telesat. While this will increase latency, collision risk, satellite lifespan, handoff frequency, and coverage footprint should improve.

Their plan called for 648 satellites providing regional service by the end of 2025 and global service with a second 648 satellites by the end of 2027. By 2030, they planned to have 15,000 satellites in orbit and offer direct-to-mobile service, but it does not look like they will make these goals.

It's been a year since the first Qianfan launch, but five months since the last one. Is the slowdown due to satellite or launch availability, or are they pausing for some redesign, or both? Curcio reports that they are “having a very hard time" finding rockets to send full batches of 18 satellites to orbit, but they have also had operational problems. The upper stage of the first launch fragmented, creating over 300 pieces of trackable debris, and ninety satellites are in orbit, but fourteen have not reached their operational altitude. Furthermore, the satellites are interfering with astronomy, and some are tumbling. Regardless of the cause for delays, Qianfan is unlikely to meet its ITU launch deadlines.

Qianfan is a more direct competitor to Starlink than Guowang, which is primarily focused on domestic telecommunications and national security. SSST has been actively marketing wholesale service through foreign telecom companies under the Sailspace brand name. They had MOUs with several nations in six initial target markets, as shown below, and they have subsequently been actively marketing in Asia, Africa, and Latin America. 

Honghu-3

Landspace Technology Corporation was founded in 2015, following a 2014 central government policy shift that opened the launch and small satellite sectors to private capital. Landscape owns 48% of Hongqing Technology, which is developing the 10,000-satellite Honghu-3 constellation. Honghu-3 satellites will be in six planes, ranging from 340-550 km. 

Landspace has a pending IPO and is developing the Zhuque-3 rocket, which they plan to launch later this year. The Zhuque-3 will carry about 21,000 kg to LEO in an expendable configuration – less than an expendable Falcon 9, but more than a reusable Falcon 9. This connection to a rocket manufacturer is reminiscent of SpaceX's relationship with Starlink and Project Kuiper's with Blue Origin. (Several other Chinese companies are also working on reusable rockets).

Honghu-3 was announced after Guowang and Qianfan, and relatively little is known of their plans and technology, but Landspace has valuable experience as a private company. As you see in this conversation with ChatGPT, Landscape has a complex mix of private, state, and local government investors dating back to its founding, and it estimates the ownership breakdown as roughly 60% private, national government 15-20%, and provincial/municipal around 20%.

Update 9/14/2025

The Hong Kong Office of the Communications Authority has released a report on a Qianfan test using both the standard and high-performance terminals conducted on a cruise ship in Victoria Harbour, Hong Kong. The ship had an unobstructed view of eight satellites orbiting in a plane over Hong Kong.  They tested Web page loading, HD video playback, WeChat video calls, and large online games.

• The tester accessed the Baidu hot search page using a mobile browser, and images and text loaded quickly; video playback was also smooth.
• There was no lag or abnormal playback when streaming 4K high-definition video from the CCTV website.
• The WeChat video communication quality was comparable to that of terrestrial 4G/5G networks. The video remained stable during the test, with no noticeable lag.
• The user experience while playing League of Legends was good, and the network delay was kept at 60 to 70 milliseconds.

The report also lists the technical specifications of both the standard and high-performance terminals, and speed tests for both terminals are shown below.

Update 10/23/2025

Quinfan resumed launches with a batch of 18 satellites after a six-month delay, presumably to correct for the problems of an exploding first stage, tumbling satellites, and interference with astronomy. They have now launched 108 satellites, and 14 have failed and are decaying, 94 are working, and 67 are in their operational orbit. 

As mentioned above, Blaine Curcio noted that they were also “having a very hard time" finding rockets to send full batches of 18 satellites to orbit, but Andrew Jones reports that they have expanded launch procurement beyond state provider CASC, awarding $187 million in contracts to Landspace, Space Pioneer, and CAS Space. Still, there is no way they will meet the original goals of 648 satellites providing regional service by the end of 2025 and global service with a second 648 satellites by the end of 2027.

Update 1/9/2026

After the December 26 launch of nine satellites, there are now 136 Guowang spacecraft in space, functioning in and heading up to their operational orbit. They were launched by a Long March 8A rocket, so they were probably the smaller Guowang satellites. Guowang plans to launch 310 satellites in 2026, 900 in 2027, and 3,600 every year beginning in 2028. 

A Chinese fisherman was able to retrieve the fairing from the launch -- a cool souvenir.

Update 3/7/2026

Blaine C   

Zhuque-3’s next booster landing attempt is set for the second quarter of this year, and they hope to begin re-flying boosters during the fourth quarter. https://www.china-in-space.com/p/landspace-wants-to-reuse-zhuque-3 

They have also successfully tested their version of a “Pez dispenser” for satellite stacking and release, so they may be ready for full-scale satellite deployment later this year, serving the Chinese LEO satellite Internet service providers.

Update 5/13/2026

On May 12, another 18 Qianfan satellites were launched, bringing the total number in orbit to 144. They have evidently solved the technical problems they experienced in a five-month gap between launches at one point last year, but, as Blaine Curdio pointed out above,  they were “having a very hard time" finding rockets to send full batches of 18 satellites to orbit. 

Offshore launches offer a partial solution to that problem, and Qianfan's operator, SSST, has partnered with four other companies to finance Shanghai Commercial Aerospace Maritime Launch Technology Co Ltd., which will prepare launch missions near Shanghai and then perform them from nearby waters. SSST is one more company located in China's G60 Science and Technology Innovation Corridor.

Update 6/9/2026

Qianfan launched 18 satellites on June 4th and 18 more on June 5th. The pause for correcting problems is definitely over, and they now have 200 satellites in orbit (185 working). The goal of the "first phase" is to have launched 324 satellites (18 launches of 18 satellites) by the end of July, and the cost of each satellite is around 10 million yuan (about $1.5 million). Qianfan expects to provide service in China and Brazil by the end of the year.

Update 9/4/2026

Qianfan developer SSST, also known as Yuanxin Satellite, has received nearly 6.98 RMB in new investment from 18 B-round investors (including an Alibaba investment subsidiary), bringing these investors' share to 13.9384% ownership.

I asked ChatGPT to categorize the complicated investment vehicles into three sources: national-level funds, local state-owned asset platforms, and industrial capital and market-oriented venture capital institutions:

Shanghai government	~60%
National/central government	~25%
Private capital	~15%
Total	100%

We saw a similar ownership mix in the case of LandSpace. The Reagan-era world of Communists vs Capitalists is over.

Two Ways to Build the Internet in Space -- China, Inc. vs Starlink et al

SpaceX has over 10,000 working Starlink Internet satellites in orbit today and is far ahead of its Western and Chinese competitors, some of whom have not yet begun launching satellites. In my last post, I estimated the number of satellites each constellation would have in orbit at the end of the current Chinese five-year plan in 2030, and said one could think of the three Chinese constellations as divisions of a single enterprise, China, Inc. 

The Western approach of self-contained competitors has clear advantages. It allows rapid deployment, tight control over performance, and independence from other systems. The deployment of Starlink in Ukraine illustrates this point. On February 26, 2022, Ukraine requested Starlink service, and by the end of the day, service was activated, and terminals were en route. By March 19th, there were over 5,000 terminals in the country, and there were 15,000 by June 9th.

That’s the good news, but the bad news is that each provider must build and operate its own full-stack constellation, leading to duplication and a lack of global awareness. The Chinese approach to the Internet in space reflects their practice of making decisions within the context of five-year plans; it is slower but allows for a degree of optimization.

Rather than building multiple competing full-stack systems, Chinese constellations implement a division of labor: Guowang (backbone and sensitive government applications), Qianfan (general Internet service), and Honghu-3 (direct-to-device, Earth observation, IoT, etc.). For example, in providing global airline or maritime connectivity, Qianfan or Honghu-3 would both use Guowang for long-distance traffic. In this emerging architecture, the Internet stack is distributed across constellations. Instead of duplication, there is specialization and coordination.

The architectural difference becomes most visible in routing. In the Western stand-alone systems, routing is internal in each. In the Chinese layered systems, routing may span constellations, enabling system-level coordination and optimization. This does not imply fragmentation of the Internet, but it changes how traffic is managed and controlled.

It also alters our interpretation of my estimates of how many satellites each constellation will have in orbit in 2030. My estimate was that Starlink would have 16,083 satellites in orbit in 2030, far more than Western and Chinese runners-up Amazon LEO with 2,729 and Qianfan with 5,217. However, if we consider the Chinese constellations part of a unified, optimized collective with 11,692 satellites, the gap is smaller.

Note that, given today’s international political differences, the potential markets for Western and Chinese connectivity differ, as shown in this Venn diagram of the number and GDP of nations with Starlink connectivity, Chinese Belt and Road Initiative (BRI) projects, and both. The BRI projects tend to be in poorer nations, but there are more nations with BRI projects than with Starlink availability. The current success of Starlink suggests that the market is large enough to support it and the Chinese constellations. 

Finally, in a study of the Internet in India and China in the late 1990s, my colleagues and I observed that “China has been able to execute plans effectively by allocating resources to competing, government-owned enterprises.” That pattern is evident here. China’s earlier, now-discontinued, LEO broadband initiatives—Hongyun, sponsored by the state-owned China Aerospace Science and Industry Corporation, and Hongyan, sponsored by the state-owned China Aerospace Science and Technology Corporation—were initially conceived as competing systems. 

The creation of the China Satellite Network Group and the subsequent prioritization of the Guowang constellation following its initial announcement and ITU filing in 2020 marked a turning point. Experience gained from Hongyun and Hongyan must have informed this transition, with technology, personnel, and spectrum filings absorbed or redirected. Subsequently, Qianfan and Honghu-3 have emerged in more differentiated roles. The result is not simply the replacement of earlier projects, but a progression from competing proposals to a more coordinated, multi-layered architecture—one that reflects both accumulated experience and the broader organization of China’s space and telecommunications sectors.

Update 6/20/2026

It seems several collaborating organizations will also be working on orbiting data centers, while SpaceX goes it alone.

"Last week, Beijing quietly launched its first Space Computing Industry Innovation Center. Government-chartered, led by BUPT, a top state telecom university. The mandate: radiation-hardened space-native AI chips, compute satellite platforms, space-optimized LLMs, integrated space-ground networking, and "tokenized" orbital compute operations. 24 hours later, Beijing E-Town convened satellite makers, chip firms, and materials companies to plan a Space Intelligent Computing Research Institute."

Update 8/15/26

When I wrote this post, I characterized Qianfan as providing general Internet service and Honghu-3 as focusing on direct-to-device, Earth observation, IoT, etc., but I was wrong. Today, it appears that Honghu-3 will compete directly with Qianfan. The differences are not in target applications, but in maturity and the degree to which they are integrated and how they will be marketed.

Qianfan is a Starlink-like, vertically organized satellite Internet service provider, building and marketing a global broadband network. Hongqing Technology filed for the Honghu-3 constellation with the ITU and is developing the relevant satellite technology, but it is not yet clear whether they will eventually operate Honghu-3, establish a separate operating company, partner with Chinese telecom carriers, or primarily provide the constellation infrastructure while another company handles retail service and marketing.

Another important difference is Honghu-3's close relationship with LandSpace, a rocket and launch company that is very close to being able to launch satellites and reuse rockets.  LandSpace was a founding 48% investor in Hongqing Technology. That share has been reduced to around 15% by new investors, but a link remains between the two companies.

Starlink and the seven dwarfs

This post has been superseded, see: 

https://cis471.blogspot.com/2025/11/starlink-and-seven-dwarfs.html

Source: ChatGPT

In the 1960s, IBM dominated the computer market, which was often referred to as “IBM and the seven dwarfs.” IBM is prosperous today, but no longer dominant. The low-Earth orbit (LEO) satellite Internet service market today is reminiscent of that time, but it’s "SpaceX Starlink and the seven dwarfs." 

As of August 2025, Apple's market cap is approximately $3.363 trillion, while IBM's is around $223.03 billion. I do not expect Starlink to drop off that precipitously, but its lead will be significantly diminished. Five of IBM's dwarfs failed, and two merged to form Unisys, which has a market capitalization of $280.18 million; however, I expect all of Starlink's dwarfs to survive.

Let's look at today's dwarfs. 

OneWeb

Bill Gates and two partners founded Teledesic, a would-be LEO satellite Internet service provider, in 1990, but the technology was not yet ready, and Teledesic declared bankruptcy in 2002.

The next would-be LEO Internet service provider, OneWeb, was founded by Greg Wyler, who had extensive experience with networking in developing nations, with the mission of “bridging the digital divide by 2027”, but it entered bankruptcy in 2020. The company was reorganized and emerged from bankruptcy, and in 2023 merged with established geo-stationary satellite (GEO) operator Eutelsat, creating the "Eutelsat Group" company, with subsidiaries "Eutelsat" and "Eutelsat OneWeb.”

In spite of that rocky start, OneWeb is the only company other than Starlink that is offering LEO satellite Internet service today. OneWeb LEO revenue was 187 million euros ($216 million) for the 12 months ending June 30,2025, representing around 15% of total Eutelsat Group sales. Starlink revenue for 2024 was $2.7 billion. OneWeb’s market share, and more importantly, global capacity, are minuscule compared to those of Starlink.

That sounds grim, but given Elon Musk’s political activity, Trump’s MAGA/isolationist policy, and the military value of LEO Internet, Europe will not let OneWeb fail unless there are viable alternatives to Starlink. This is evidenced by European support of OneWeb in Ukraine, including German funding of OneWeb and a recent British investment.

Finally, note that the Eutelsat Group can offer multi-orbit service, switching seamlessly between Eutelsat GEO and OneWeb LEO satellites or offering OneWeb service to other GEO providers. They signed their first multi-orbit contract three years ago and have added others since.

Telesat

Telesat, an established Canadian GEO satellite operator, was the next LEO Internet company. Telesat recognized the trend to LEO, but decided not to offer consumer connectivity. 

Telesat has been beset by delays and has reduced its initially planned constellation size, but they have contracted (with SpaceX) to begin launching satellites next year. Trump’s immigration and tariff policies, along with talk of annexing Canada, assure us that the Canadian government, which, along with Quebec, has invested in Telesat, will not allow it to fail.

The initial “Lightspeed” constellation will consist of 198 satellites with a mass of 750 kg, roughly that of Starlink V2 mini satellites.  SpaceX is slated to deploy them over the course of a year, starting in mid-2026. Telesat has been booking customers, and their LEO backlog now exceeds their GEO backlog. They plan to provide global service with polar and inclined sub-constellations, are seeking a ground station partner, and have terrestrial deals with Vocus, Orange, and Space Norway.

While Telesat will not bundle its own LEO and GEO services, they have tested a hybrid deployment between LEO and GEO using the Telesat Lightspeed emulator, showing seamless integration without any issues. Software like the emulator is part of their strategic decision to use Aalyria Spacetime, a multi-layer, multi-orbit operating system for a temporospatial network, which they acquired from Google when the Loon project was abandoned.

Amazon Project Kuiper

Project Kuiper, which has only 101 operational satellites in orbit today, is far behind Starlink, which has over 8,000, but Amazon has many things going for it. From the time it was founded, Amazonwas an infrastructure company, and Project Kuiper is an orbiting infrastructure that willbe strategically paired with Amazon’s complementary terrestrial infrastructure, like fiber and datacenters. Amazon hasvast experience in manufacturing and logistics that will stand them in good stead with the manufacturing of terminals as well as satellites. 

Project Kuiper is a wholly-owned subsidiary and an initiative of Amazon, and Jeff Bezos is the founder of both Amazon and the Blue Origin launch company, which will launch some Project Kuiper satellites. Amazon itself will also be a significant Kuiper user, and Kuiper will use Amazon’s ground station service. 

That’s the good news, but Amazon faces an FCC deadline to launch half the constellation by July 30, 2026, and the remainder by July 30, 2029. They say they will be able to receive, test, and pack 100+ Kuiper satellites per month into the appropriate fairing and claim to have secured 80 launches, but how fast can they manufacture them? They will apply for a waiver from the FCC, if necessary, and. like Elon Musk, Jeff Bezos has a lot of money and attended Trump’s inauguration. Earlier, Musk might have stopped an Amazon waiver, but now Trump is looking into deporting him, and  Bezos has made editorial changes at the Washington Post, which he owns. A political contribution might solve the FCC deadline.

IRIS²

IRIS² (Infrastructure for Resilience, Interconnectivity and Security by Satellite) is a €10.6bn project with 61 per cent funded publicly and the balance coming from the SpaceRise industrial consortium, led by Eutelsat, Hispasat, and SES. SpaceRISE will design, deliver, and operate IRIS² for a period of 12 years.

They have contracted for 274 satellites in LEO and 18 in MEO, with first launches anticipated for 2029 and completion in 2030. Eutelsat will act as prime contractor leading the design of the LEO segment and co-leading the development of common system elements. SES will be responsible for procurement, integration, and operation of the MEO satellites, and Hispasat will lead the very low orbital layer (Low LEO) of the constellation and design, deliver, and operate the ground segment, manage operations, and interconnection with terrestrial networks. They also expect to eventually add a GEO sub-constellation.

This is a unique and complex organization that will have to manage suppliers like Airbus, Thales, OHB, Deutsche Telekom, and  Orange. Bureaucracy might be a problem, but Europe can not rely on Starlink as Ukraine has during the war with Russia.

Three Chinese Dwarfs

China has pursued a strategy of competition among government-owned organizations, and it initiated two government-owned constellation projects, Hongyun and Hongyan, in 2016. In April 2020, China’s National Development and Reform Commission included “satellite internet” on its “new infrastructures” list, and China applied to the ITU for a new constellation, called  GW. Hongyun and Hongyan were dropped, and GW, also called China Satnet or Guowang, emerged as China’s global Internet service provider and it was followed by two others, Qianfan and Honghu. These are all far behind Starlink, but they will have protected markets

Gwowang

China’s 14th Five-Year Plan (2021-2025) for National Economic and Social Development and Long-Range Objectives Through the Year 2035 called for building an integrated communications, Earth observation, and satellite navigation system with global coverage. Guowang is the constellation they called for.

Guowang consists of two sub-constellations, designated GW-A59 (6,080 satellites) and GW-2 (6,912 satellites). GW-2 will orbit at 1,145 km, and  GW-A59 will orbit around half that. The ITU filing was in September of 2020, and after a long delay, the first ten GW-2 satellites were launched at the end of 2024, and they now have 81 in orbit. The cadence has picked up recently -- China has launched another batch of Guowang satellites. This was the ninth Guowang launch this year and the sixth in the last 30 days. Even at this cadence, it is unclear that they can manufacture and launch enough satellites to meet the ITU launch deadlines. Perhaps the Chinese have decided that, given launch and manufacturing resources, they would not be able to meet ITU deadlines for all of their constellations, so they are focusing on Guowang, which can be seen as most critical for the government

Little technical information is available, but considering the capacities of the various rockets used to launch Guowang satellites and the number of satellites in each launch, it seems there are two sizes of satellite: large satellites of around 16,600 kg and smaller satellites of around 889 kg. While these are imprecise estimates, they indicate two classes of satellite with different capabilities and functions. (Note that the relatively high altitude GW-2 sub-constellation has both large and small satellites).

Several Guowang test satellites have also been launched, suggesting strategic government and military applications like Signals intelligence, positioning, navigation, and imaging applications in addition to Internet service.

Qianfan

Shanghai Spacecom Satellite Technology (SSST), a private company backed by the Shanghai municipal government and the Chinese Academy of Sciences, is developing the Qianfan constellation. The planned satellites will orbit at 1,160 km, which is higher than the other announced LEO satellite competitors except Telesat. While this will increase latency, collision risk, satellite lifespan, handoff frequency, and coverage footprint should improve.

Their plan called for 648 satellites providing regional service by the end of 2025 and global service with a second 648 satellites by the end of 2027. By 2030, they planned to have 15,000 satellites in orbit and offer direct-to-mobile service, but it does not look like they will make these goals.

It's been a year since the first Qianfan launch, but five months since the last one. Is the slowdown due to satellite or launch availability, or are they pausing for some redesign, or both? Blaine Curcio reports that they are “having a very hard time finding rockets to send full batches of 18 satellites to orbit, but they have also had operational problems. The upper stage of the first launch fragmented, creating over 300 pieces of trackable debris, and ninety satellites are in orbit, but fourteen have not reached their operational altitude. Furthermore, the satellites are interfering with astronomy, and some are tumbling. Regardless of the cause for delays, Qianfan is unlikely to meet its ITU launch deadlines.

Qianfan is a more direct competitor to Starlink than Guowang, which is primarily focused on domestic telecommunications and national security. SSST has been actively marketing wholesale service through foreign telecom companies under the Sailspace brand name. They had MOUs with several nations in six initial target markets, as shown below, and they have subsequently been actively marketing in Africa.

Honghu-3

Landspace Technology Corporation was founded in 2015, following a 2014 central government policy shift that opened the launch and small satellite sectors to private capital. Landscape owns 48% of Hongqing Technology, which is developing the 10,000-satellite Honghu-3 constellation. Honghu-3 satellites will be in six planes, ranging from 340-550 km. 

Landspace has a pending IPO and is developing the Zhuque-3rocket, which they plan to launch later this year. The Zhuque-3 will carry about 21,000 kg to LEO in an expendable configuration – less than an expendable Falcon 9, but more than a reusable Falcon 9. This connection to a rocket manufacturer is reminiscent of SpaceX's relationship with Starlink and Project Kuiper's with Blue Origin.

Honghu-3 was announced after Guowang and Qianfan, and relatively little is known of their plans and technology, but Landspace has experience as a private company. (I asked the ChatGPT, Gemini, and Copilot chatbots how much state and private capital Landspace had received since it was founded, and the answers and explanations varied so much as to make them worthless, but they all agreed that the private investment was greater than the public.)

Inter-satellite laser link update

SpaceX satellite mesh with four
laser terminals on each satellite

Inter-satellite laser links (ISLLs) and electronically steerable flat panel antennas are critical technologies for constellations of low-Earth orbit (LEO) Internet-service satellites. Low-cost antennas are critical for the mass consumer market and ISLLs are required for an effective Internet backbone in space. In an earlier post, we saw that progress is being made on antennas, this one looks at ISLLs.

The figure to the right is taken from a simulation of the first phase of SpaceX's planned broadband Internet service, Starlink. It shows 66 satellites in each of 24 53-degree orbital planes -- a total of 1,584 satellites at an altitude of 550 km. Each satellite has four laser-communication terminals. Two on the front and back and two on the sides. Since the front and back lasers link to satellites in the same orbital plane, they remain at the same place in the sky relative to each other while the side lasers must move to track one another. (To visualize the dynamic nature of the links between the constantly moving satellites, check this clip from the animated simulation).

When Elon Musk introduced his Starlink plan to prospective employees in 2015, he said his goal was to transport "a majority of long-distance Internet traffic" and "about 10 percent of local consumer and business traffic." He pointed out that satellites would have an advantage over terrestrial links since the speed of light is faster in space than through optical fiber and fewer router hops would be needed to reach a distant location.

In addition to mitigating the digital divide by serving rural areas and small organizations, Musk and his competitors at OneWeb, Telesat, Amazon, and Leosat hope to service high-end, high-margin customers like enterprises, governments and maritime, airline and mobile phone companies. ISLLs are necessary for serving those lucrative high-end markets.

Initially, SpaceX proposed five ISLLs for each satellite -- the fifth would have been a link to a satellite in the crossing plane, but last November they cut back to four. The fifth terminal would have been difficult to engineer because while the front, back and side-mounted terminals move slowly relative to each other, this simulation shows that satellites in crossing planes would have been traveling at 7.3 km/second relative to each other. Furthermore, links between satellites in crossing planes would be of short duration. Designing and manufacturing them would have taken time and money.

Furthermore, because of the 53-degree orbit inclination, about half the satellites are moving northeast and half are moving southeast at any time and place. That favors east-west links over north-south links and since most of the lucrative low-latency, long-link traffic is in the northern hemisphere, they could not justify the cost or possible deployment delay. That is not to say they will not deploy them in the future. (Note that the initial five-link constellation was to orbit at an altitude of 1,100, not 550 km. Future plans call for constellations at 1,100 and 335-345 km and there may be ISLLs between all of them).

Tesat laser communication roadmap

But even with 4-links, the terminals are still under development and will be expensive. At least two companies are working on ISLLs, Mynaric and Tesat.

Tesat already markets a laser communication terminal for LEO to ground transmission from CubeSats. Their CubeLCT is 9 x 9.5 x 3.5 cm, has a mass of 360 grams, consumes 8 Watts of power and communicates through the atmosphere to the Earth at 100 Mbps, with a 1 Mbps channel from the ground to LEO. They are developing an ISLL terminal based on that experience and, judging from the diagram shown here, they are pursuing laser communication between the ground, LEO and geostationary satellites.

Mynarc has announced that their ISLL terminal, the MLT-80, will be available in high-volume production this year and both companies are working on faster terminals. A while ago, I suggested that SpaceX would probably develop their own ISLL, but last March, Bulent Altan, a former SpaceX Vice President, joined Mynarc as co-CEO and a few days later Mynaric announced that they had raised $12.5 million from mystery constellation customer. Might the mystery company be SpaceX? Might it be Amazon, which entered the race late and has enough money to pay for terminals or even buy a stake in Mynaric or Tesat? We will know soon because test satellites equipped with Mynaric’s terminals should be launched in late-2019.

The following are selected characteristics of their forthcoming ISSLs:

	Mynaric	Tesat
Link distance	4,500 km	6,000 km
Data rate (full duplex)	10 Gbps	10 Gbps
Target mass	<20 kg	<15 kg
Power consumption	<60 W	80 W

Sources Tesat, Myarnic

The SpaceX simulation shown above was for satellites with 4 ISLLs, but SpaceX launched their first 60 satellites without the ISLLs and, as far as I know, has not said if forthcoming satellites will have them or not. Arthur Sauzay, a French environment and space lawyer has pointed out that SpaceX argued for the allocation of radio frequencies for ISLs in a comment to a recent Whitehouse report on the impact of emerging technologies and their impact on non-federal spectrum demand, but they seem too large, heavy and slow to support a LEO network with long-distance, low-latency links.

OneWeb has decided not to use ISLs in their first constellation and will route traffic through terrestrial gateways. This decision seems to have been at least partially motivated by Russian insistence that satellite traffic passes through gateways within their borders. I imagine China and other nations will impose the same restriction.

Telesat remains committed to ISSLs, but say they will have the flexibility in their network-control system to route traffic coming to a country over satellite or terrestrial links. Erwin Hudson, vice president of Telesat LEO is confident that ISLs will be cheap enough to allow them to compete successfully with terrestrial fiber and 5G, offering fast, 30 ms latency broadband. They also have a $2.8 million contract to study inter-satellite laser links between their constellation and Blackjack, DARPA's 20 LEO satellite constellation and they are collaborating with Google on software, so we might see laser links between Telesat satellites and Google's balloons.

LeoSat is unique in that they are not pursuing the consumer and small organization markets, but are focused exclusively on large, high-end customers. They will provide fast, low latency, encrypted, reliable point-to-point connections to governments at up to 1.2 Gbps with latency under 50 ms and they have over $1 billion in pre-launch customer agreements. ISLLs are mandatory for the markets they are pursuing and since two geostationary satellite operators, Jsat and Hispasat, are investors in LeoSat, they may very well link to them in the future to offer a service similar to the SpaceDataHighway of Airbus and the European Space Agency.

China's Hongyun LEO broadband project is an ISLL unknown. China is doubtless working on laser communication in space, but I have no idea whether or not they will use it in their broadband constellation. Since they say the goal of the project is to serve rural China and they regulate Internet links to the outside world, Hongyun satellites may serve exclusively as "bent pipe" relays between rural locations and China's terrestrial network.

ISLLs will be needed if the Internet backbone in space is to compete with the terrestrial backbone and serve high-value applications. It seems that making cost-effective ISLLs for LEO constellations was harder than Elon Musk and others anticipated, but first production models are now on the horizon and they will improve over time.

For a copy of the PowerPoint presentation I use for teaching this topic click here.