The military-industrial scientific research system of the academic master.

Although Chang Haonan wrote in the document that communications were relayed by relying on a "ground station," this was just a general term used to make it easier for superiors and comrades in brother units to understand.

In fact, the nodes responsible for communication relay do not need to be rigidly bound by form.

After all, given that a low-orbit relay communication constellation will be planned in the near future, spending a lot of money to build expensive fixed communication sites in border and coastal areas is a waste of resources no matter how you look at it.

Moreover, even if we do not take this into consideration, the fixed relay station itself is a relatively vulnerable target in wartime and can easily be the focus of attention of the imaginary enemy.

Therefore, after the project was officially approved, Chang Haonan actually invested in the research and development of two platforms.

One is the stratospheric airship, which was developed by a Zhejiang university and had performed brilliantly during the embankment construction operation.

This is already a fairly mature technology, and Zhejiang University has not stopped upgrading and improving the equipment in recent years.

In addition, its original design purpose was to serve as a communication relay platform, which can be said to fully meet Chang Haonan's needs.

There is no need to make major changes to the overall architecture. It is only necessary to add encoding and decoding functions on the basis of the original simple reception-storage-forwarding function.

Since its floating altitude is over 20,000 meters, which is much higher than the normal flight altitude of early warning aircraft, it can be used as a simulated satellite unit during the test to test the antenna performance in the upper hemisphere of the aircraft.

Of course, because the distance is closer, the signal strength and signal-to-noise ratio are definitely much stronger than real space-based communications.

Moreover, the relay station itself is relatively static and cannot completely restore the space-based communication scenario.

This is also why Chang Haonan insisted on asking Zhang Weiyong to build the satellite.

The other platform is the vehicle-mounted ground communication relay station.

Compared with stratospheric airships, their functions are weaker and they are more susceptible to interference from clutter, but they can achieve "communication in motion", that is, communicating while moving.

It is used to test the antenna performance in the lower hemisphere, while simulating the characteristics of the satellite platform being constantly moving and the signal being weak.

There is no need for additional research and development on the vehicle chassis; it can be done by simply modifying the huge number of SX2190 trucks in stock.

If necessary, you can also move directly to other models.

As for the research and development of the entire vehicle-mounted relay station, Chang Haonan neither chose the 2000th Institute of China Electronics Technology Group Corporation, which was responsible for the development of KJ-14, nor the Space Technology Research Institute, which was responsible for the development of satellites.

Instead, they went to the 54th Institute of Electronic Science and Technology in Shimen.

Or strictly speaking, it was the leaders of the China Electronics Technology Group who took the initiative to recommend them to Chang Haonan.

Because the 54th Institute is currently developing a series of satellite communication/scattered communication vehicles for military communication units and emergency rescue units.

Although it is not exactly the same as Chang Haonan’s needs, the similarity is at least 90%.

And the technology is interconnected.

Chang Haonan had no objection to this.

The technical difficulty of the hardware of this communication relay vehicle is not high, and the core algorithm of the software is still developed by him, so it does not require the partner to have strong capabilities.

As long as you have strong execution and learning ability, it will be fine.

Besides, the 54th Institute, as a network communication research institute, is no ordinary one.

So, about a month after the QC300 project was completed, one day in late May, Wu Wei, the head of the general department of the th Institute of China Electronics Technology Group Corporation, brought a core team of four people to Beijing to find Chang Haonan in person.

The latter did not beat around the bush and stated his demands directly.

……

In Chang Haonan's office, Wu Wei looked at the densely marked technical data in his hand with a somewhat embarrassed expression:

"Boss Chang, using ground vehicles to relay communications to aircraft is not technically difficult to achieve, but..."

He carefully considered his words and said:

"But considering the size and power limitations of the vehicle platform, the range of action may not be particularly optimistic..."

When it comes to communications, the higher you stand, the farther you can transmit, so most ideas are to use airplanes in the sky to relay messages to the ground.

This was the first time Wu Wei had heard of such a request in his life.

"It doesn't matter, this is the effect I want."

Chang Haonan's answer was even more shocking:

"If the transmission range is too wide, it will be inconvenient to test the theoretical throughput performance of the multi-hop network."

Looking at Wu Wei, whose eyes were almost popping out, he had no choice but to continue explaining:

"This kind of test, whether for confidentiality or to reduce the impact on normal production and life, must be conducted far away from the city, so the test area will most likely be in the northwest, with a range of at most 200 to 300 kilometers. I have to ensure that at least three jumps are made, so the working range of the onboard system basically meets the requirements... The airship also has to reduce its power."

It took Wu Wei several minutes to digest the amount of information in this passage before he could fully understand Chang Haonan's ultimate goal:

"So... the network coding technology you just mentioned is for multi-hop communication?"

"Yes."

Chang Haonan nodded, then raised his hand and pointed at the document in Wu Wei's hand:

"If you turn two pages further, there will be relevant technical instructions."

The latter quickly lowered his head and turned to the position he mentioned.

"Of course, there's nothing much to say about network coding technology itself. You're an expert in this area, so you must have studied it... or at least learned about it."

At this time, Chang Haonan continued:

"The real core of our test this time is to use the maximum rank distance code to provide sufficient error correction capability for the network coding transmission process, ensuring that information can be accurately transmitted even in the face of strong interference caused by the opponent... or the natural environment."

"MRD code..."

Wu Wei put the documents in his hand on the table and thought:

"When the scatter communication vehicle project was just launched two years ago, I did research this technology. It can detect and correct errors in matrix data structures. The error correction efficiency can be improved by five orders of magnitude compared to the traditional low-density parity check code (LDPC)..."

Compared with Zhang Weiyong, who specializes in satellite systems, he obviously has a deeper understanding of this area.

“It’s good that you’ve done research.”

Chang Haonan put down the pen in his hand:

"In addition to its error correction capability, from a technical point of view, the MRD code is also an encryption technology that is almost impossible to crack, which can greatly improve the stability and security of communications."

The reason why he emphasized technically is that no encryption method can prevent physical theft.

In a system, the most reliable and weakest link is always people.

"But Mr. Chang, our research at that time was all done on supercomputers..."

Wu Wei's frown still didn't relax:

"The encoding and decoding process of MRD code depends on the check matrix and the generator matrix, which is very complex and its efficiency is heavily dependent on computing power. Even for supercomputers, the cycle of completing an encoding-decoding cycle sometimes takes hours. If it is to be applied to real-time communication, the computing power requirements for relay equipment will be too high..."

Chang Haonan was not surprised by this result.

In fact, the reason why the maximum rank distance code, a magical artifact with amazing effects, was proposed as early as 1985 but has never been put into use is mainly because the encoding and decoding process is too complicated.

And it was in this part that he achieved the technological breakthrough.

"I think... you were testing the Gabidulin code?"

"Yes."

Wu Wei nodded:

"After all, that is the most classic maximum rank distance code."

"But we have also tested several variants of the Gabidulin code. In general, the efficiency is better than the original one, but not very obvious..."

"But the maximum rank distance code I want to test does not rely entirely on the concept proposed by Gabidulin..."

Chang Haonan picked up the pen he had just put down:

"The technical details of this part are too long to explain in one or two hours. I will give you a special technical document later..."

"But to put it simply, it introduces a new concept called the self-duality of maximum rank distance codes... er... similar to the concept of matrix orthogonality, that is, when an MRD code C is equal to its dual code C⊥, we believe that it has self-duality. Under the premise of meeting certain conditions, an MRD code must be equivalent to a self-dual MRD code..."

"Combining this concept, we can design a fast decoding algorithm for Gabidulin codes... and of course other MRD codes, that is, for the syndrome S and i=t, t-1, t-2... compute the matrix Mi and the determinant det(Mi) until the first i=m occurs such that det(Mi)≠0, where m is the rank of the error vector e..."

"..."

"Furthermore, we can let the legitimate user A choose the matrices S, X and G as the private key, and then calculate the public key through Gcr=SG+X. Since it is almost impossible to reverse the secret matrix on the right side of the equal sign when Gcr is known, the security can be greatly improved..."

Although it was "simply put", Chang Haonan still spoke for almost half an hour.

This is based on the premise that Wu Wei has sufficient knowledge of encryption algorithms.

"In other words, we use traditional encryption methods to transmit the asymmetrically encrypted message itself, and then use asymmetric encryption to transmit the key required for decryption. In this way, we can reduce the amount of data that needs to be corrected by the relay station and ensure transmission efficiency?"

Wu Wei tentatively confirmed.

Chang Haonan nodded with satisfaction and leaned back in his chair.

Now, the latter at least understands the part about the technical route.

As for the specific algorithm level...

This can't be rushed.

Even he himself, with the help of the system, spent almost a month on this.

"I have tested this MRD code combined with network coding in a small-scale wired network."

Chang Haonan said lightly:

"The result is that in a four-router multi-hop network, the actual transmission capacity can be increased by about 15-20 times, but the processing delay is slightly higher, which has to be calculated in seconds."

"However, for satellite communications or space-to-earth communications, because the bandwidth is too low, transmission delay will replace processing delay and occupy the main position. The actual delay... will actually decrease!"