Chapter 822 Testing with AWACS
In the next half an hour or so, Tang Lintian just felt dizzy.
Gone.
Both feet are off the ground.
Given the level of the mathematics major at Beijing University of Aeronautics and Astronautics, being able to invite a scholar of Perelman's caliber to give a report is something worth celebrating.
As a result, Chang Haonan turned around and gave him such a big job.
It's basically equivalent to a person worrying about having enough food to eat one second and suddenly finding out that he is the son of the world's richest man the next second.
This description is no exaggeration.
In the current situation, if Chang Haonan is willing, Tang Lintian probably won't care about trivial matters such as seniority...
In short, when the latter came to his senses again, he found himself in Chang Haonan's office.
I have absolutely no memory of what happened in between.
"Professor Chang."
As Tang Lintian spoke, he still felt his heart beating very fast, so he paused halfway through his speech:
"This result... probably... when will it be..."
As a man in his fifties, he has deeply felt the mark left on him by the passage of time.
But at this moment, if life could have a fast-forward button, Tang Lintian would press it without hesitation.
I can't wait any longer...
Poincare conjecture...
Even if he is not a mathematics professor, he knows this famous name.
It is a difficult problem that is as famous as the Goldbach conjecture and Fermat's Last Theorem and has troubled mathematicians around the world for more than a century (in fact, the Poincare conjecture is more significant, but those two are really out of the circle).
If it was resolved in Beijing Airlines...
Don't say anything else.
The lecture hall where the lecture was held yesterday can be directly promoted to a sacred place for mathematics.
Don't ask, the inspiration for solving the Poincare conjecture came from this place.
As the current president of Beijing Aviation University, Tang Lintian is definitely honored.
"It's hard to say."
Chang Haonan shook his head and replied:
“When it comes to math, the most time-consuming part is often not necessarily the most difficult part.”
He understood Tang Lintian's feelings, but since the current theoretical level LV3 was not enough to start the project, the Poincare conjecture was not directly solved through the system, and it was difficult to give a clear timetable.
All I can say is that in terms of supercomputing performance at the turn of the century, his own expectation was 3-4 months.
"Yes, also..."
Although he didn't hear the answer he wanted, Tang Lintian nodded repeatedly:
"A problem that has not been solved for a century must not be simple..."
After this topic came to an end, the office fell silent again.
Tang Lintian, who was still excited, was actually a little restless, but he couldn't jump up and down to vent his anger in front of Chang Haonan.
So after sitting there uncomfortably for a few minutes, he finally couldn't stand it anymore and stood up:
"I'll go find Professor Meng and talk about your project."
Chang Haonan was just about to say that it would be enough to just make a phone call, why bother making a special trip.
As a result, when I looked up, I saw Tang Lintian almost jumping out of the door...
"Ugh……"
He sighed, put down his hand that was about to pick up the phone receiver, smiled and shook his head.
About 40 minutes later, Tang Lintian, who seemed to have recovered, finally returned to Chang Haonan's office with Meng Zhenyuan.
After hearing that Chang Haonan had found a solution to his problem, Meng Zhenyuan's reaction was...
It's basically the same as Tang Lintian just now.
This time, Chang Haonan, who had experience, did not disturb the two people opposite him, but just busy himself with the things at hand.
Sure enough, after a while, Meng Zhenyuan finally came to his senses.
"No, Mr. Chang, this... is this true?"
Chang Haonan was amused:
"Of course."
He turned his eyes away from the computer screen and looked at the two people sitting on the sofa:
"Can I still lie to you?"
"No, no, no... Of course I believe Professor Chang..."
Meng Zhenyuan rubbed his hands.
Apparently, although he had recovered from the initial shock, his language and thinking abilities were still somewhat affected:
"Oh my god..."
Meng Zhenyuan sighed again, then looked at Chang Haonan and said sincerely:
"Mr. Chang, since you solved this problem, do you think there is any equipment or materials that need to be purchased? I can use the funds from this project to help you solve it!"
This is indeed a big trouble -
The project Meng Zhenyuan is working on is a vertical project, and the management of funds is much stricter than a horizontal one.
At least it is impossible to be like the agreement signed by Chang Haonan and China Petroleum Pipeline Bureau, where if the project is completed ahead of schedule, the funds will be paid in one lump sum in the name of personnel expense subsidy at the end of the project.
For vertical projects, if you do not use up all the funds you applied for, not only will you not receive any reward, you may even be rejected.
The logic is also very simple -
Why did you apply for so much if you couldn't spend it all?
But even so, when applying for a project, no one will really think about how to spend the money first, and then get the money according to the planned amount.
The key is to get as much funding as possible and then find places to spend it.
Therefore, at the end of each year, or when a project is about to be completed, there will be a rush to spend money -
Of course, this is definitely a happy trouble.
For most research groups, the problem is not enough money, not too much money...
However, the problems Meng Zhenyuan is facing now have obviously exceeded all his past experiences.
It wasn't that he had some money left that needed to be spent, but that he was going to finish the project without spending a penny.
"Hmm...how many?"
Chang Haonan was not going to be polite, after all, he was the one who solved this problem.
"Three years, 20 per year."
Meng Zhenyuan replied.
"That doesn't seem to be enough..."
Chang Haonan held his forehead with one hand.
In fact, a project with a funding of 60 yuan was definitely considered a big project in 2000.
only……
There is no harm without comparison.
Ultimately, he decided to skip the topic:
"Forget it, keep it for now. Anyway, the earliest I can finish the project is the end of this year. I'll contact you when I need money for something."
……
The issue of funding is just a side topic.
The focus is still on the project itself.
Next, Chang Haonan spent about two hours introducing his improved manifold learning algorithm to Meng Zhenyuan and Tang Lintian, as well as the idea of using this algorithm to extract and analyze multi-band spectral data.
Simply put, in an era where sensors are already highly developed, the problem people face is not too little data but too much data.
It's impossible to tell which ones are useful and which ones are not.
Manifold learning technology can accurately extract the specific data needed for analysis from a radar image that is filled with trillions or even gigahertz of information.
Of course, this is just one of many application scenarios.
If the computer is powerful enough, it can also realize advanced functions such as face recognition and image search.
"That sounds... perfect."
After Chang Haonan's introduction, Meng Zhenyuan felt that he was almost convinced by this ingenious method:
"But there is still a problem..."
"Our satellite will not be launched for more than a month... Although I also think this idea is feasible, it still needs to be tested..."
After all, the former acted a little too fast.
"Yeah... indeed."
Chang Haonan nodded:
"It may take three months for the satellite to transmit the data back, and if the geological information needs to be verified, we have to wait for the results of on-site exploration to come out. The efficiency is really low..."
He believes that the algorithm he developed will definitely not be wrong.
However, in order to optimize the results to be good enough, we still need enough samples for comparison and parameter adjustment.
Relying on just one satellite, Resources-01 , and working with dozens of people at the Exploration and Development Research Institute, I'm afraid it would be hard to meet the efficiency requirements.
"Anyway, as long as it's telemetry data, then why not let the two Learjet S aircraft from the Academy of Sciences join in?"
Tang Lintian next to him suggested.
In the mid-80s, China imported two Learjet S/II business jet electronic reconnaissance aircraft from the United States, which were later registered under the name of the Academy of Sciences.
In addition to military use, it also performs some scientific surveying and mapping tasks.
Chang Haonan waved his hand:
"The shortcoming here is that the efficiency of on-site survey is not good. Even if there are two more planes..."
At this point, his eyes suddenly lit up.
Ultimately, the function of manifold learning algorithms is feature extraction.
We don't necessarily have to limit our focus to geological information.
Ground reconnaissance, battlefield surveillance, target identification...
That makes sense.
And because targets such as buildings and vehicles are controllable, even the effort of data verification can be saved.
At this moment, an excellent experimental platform suddenly flashed through his mind——
The KJ-200 which has already completed its maiden flight.
Although the centimeter-wave radar has limited accuracy in ground detection and cannot be compared with a regular synthetic aperture radar, considering the huge antenna area and power, it should not be a big problem if it is only used for technical verification.
Moreover, it kills two birds with one stone.
In addition to verifying the hyperspectral remote sensing equipment, it is also making technical preparations for China's future battlefield command aircraft, integrated reconnaissance aircraft, or strategic reconnaissance satellites!
