Chapter 825 Testing, Target Feature Extraction

Of course, using ballistic missiles to counter aircraft carriers is definitely not that simple. It can be solved by developing a target classification algorithm.

First of all, the prerequisite for "target classification" is to detect the target.

But how to enable ballistic missiles to overcome black obstacles and use radar normally during re-entry into the atmosphere is a huge problem...

Reducing the reentry speed to less than 5 times the speed of sound can avoid the occurrence of black barriers from the source. However, low-speed ballistic targets are the easiest targets for the air defense system, which defeats the original purpose of using ballistic missiles to fight ships.

If you wait until you have passed the black barrier area before turning on the radar, the reaction time will be too short and it will not be enough for the missile to adjust its attitude.

Secondly, to find the location of the aircraft carrier battle group in the vast ocean before striking, a complete global ocean monitoring system is needed.

At this moment, China has just launched its first land-based remote sensing satellite.

The second and third ones are Resource No. 2 01 and 02 that will be launched soon——

This was also the original reason why Chang Haonan rushed out to conduct technical tests before the Spring Festival.

As for ocean remote sensing satellites...

Sorry, not even one.

In short, at least in 2000, we could only think about developing anti-ship ballistic missiles——

Most people wouldn't even think about it.

On the contrary, Wang Xiaomo's idea is more realistic.

In particular, Chang Haonan has previously considered designing an anti-submarine patrol aircraft based on the C808 platform with comfort and range that exceeds or is as good as the P3C.

Under the background that the main tone of the current confrontation is to fight without defeat, the importance of this equipment is not less, if not more important than conventional main battle equipment such as tactical aircraft.

Thinking of this, Chang Haonan immediately turned to look at Liang Shaoxiu, who was a bit out of touch with radar-related topics:

"Gong Liang, our Air Police 200 doesn't have hardware to shield low-altitude targets, right?"

Since the main function of early warning aircraft when developing is air detection, there are many models that directly limit the longitudinal depression angle of electromagnetic waves on the hardware, in order to solve the interference caused by ground echoes in the simplest and crudest way.

What's more, a shielding layer will be designed under the antenna to directly block the downward electromagnetic waves at large depression angles.

Chang Haonan definitely didn't have this thing when he made the plan, but there is no guarantee that it will be added in the past two years.

"No, I just limited the downward angle in the radar background program."

Liang Shaoxiu replied:

"And because it is still a prototype, this restriction can be lifted through a special command. If you have a new target recognition program, Mr. Chang, just modify the software part here."

Since pilots and ground crew are not professional aircraft designers, when the mass-produced aircraft is delivered to them, in order to prevent anyone from messing with it, the software programs are locked, leaving only some preset input and output ports. .

However, prototypes or technical verification opportunities have a status similar to "developer mode", allowing engineers to make modifications more conveniently.

"Don't worry, there is no need to make major changes to the current procedures."

Chang Haonan looked back at the staff who followed him.

The latter immediately understood and took a password box that he carried with him.

After opening it, there are several hard drives inside.

"I have encapsulated the target feature extraction and analysis functions into a module, which can be plugged into the existing data processing code."

Having said that, the operating computer on the early warning aircraft is not a home computer after all. Just make an installation file in .msi format and wait for it to be completed automatically.

Windows XP, which was widely used in military terminals in later generations, was still a trendy gadget that could only be installed on the latest PCs.

The hardware requirements are extremely high and it is also unstable.

It’s almost like win11 in 2023.

And it’s not officially released yet.

Therefore, it took Chang Haonan nearly a week to completely complete the program update and necessary ground testing.

It even took longer than it took him to write the program before.

Of course, Liang Shaoxiu and the others are not idle these days either.

Instead, he helped coordinate with the navy leaders and borrowed a batch of false targets for training.

It's not something fancy like a self-propelled decoy.

Instead, it is a simple method to install different styles of models or corner reflectors on the buoys.

Used to low-cost simulate targets of various sizes from sampans to destroyers in non-live-fire training.

It is a method that the army had no choice in the early years when it had to endure.

The advantage is that it can be deployed with a transport boat or tugboat, and it can be recycled without damage after use.

But later I found out that the effect was actually quite good.

However, because it is a buoy, it cannot be deployed in places such as the high seas and oceans. It can only play a role in relatively shallow locations.

…

The last day before New Year's Eve.

"Mr. Chang, comrades from the North Sea Fleet sent back a message saying that they have set up all the 35 preset false targets. In addition, the three warships 107, 110, and 519 that are training in nearby waters will also cooperate with us. Real ship target, compared with the signal characteristics of the corner reflector.”

Liang Shaoxiu hung up the phone and said to Chang Haonan who was not far away.

The latter was currently looking out of the floor-to-ceiling window of the tower, watching Air Police 200 drive from the apron onto the taxiway:

"Notify Unit 20001 to prepare for takeoff."

If it makes sense, then Chang Haonan, as the deputy chief designer, Liang Shaoxiu should be regarded as his leader in the Air Police 200 project.

However, the content of this test actually has nothing to do with the early warning aircraft itself, and the two parties are closer to cooperation.

The reason why it is carried out within the framework of Air Police 200 is simply for convenience.

Moreover, Liang Shaoxiu has actually been learning from Chang Haonan since the time he solved the problem of ice on the wings of the Y-8 platform at the end of 1996.

Like Lin Shikuan from Institute 601 back then, he was considered a half-disciple brought out by Chang Haonan.

Therefore, there was a strange scene in front of us where the chief engineer reported the situation to the deputy chief engineer, and then the deputy chief engineer issued instructions to the chief commander.

A few minutes later, Air Police 200, dragging four lines of faint black smoke, rose into the sky from the end of the runway and gradually disappeared from everyone's sight.

"Mr. Chang, Comrade Shaoxiu has complained to me many times before, saying that the performance of the turboprop 6 is still lagging behind..."

Wang Xiaomo looked at the black trail that had not completely dispersed in mid-air and said to Chang Haonan next to him:

"I heard that when your aerospace power group was established last year, you mentioned a development plan for a turboprop engine?"

Obviously, this is a hint...or a hint that he wants a turboprop engine with better performance.

In other words, a flying platform with better performance.

If Liang Shaoxiu had said this, it would have been difficult to determine the balance between suggestions and dissatisfaction.

But Wang Xiaomo has nothing to do with the aviation system, so he has no worries in this regard.

"Indeed."

Chang Haonan nodded:

"But the design power of that machine is only 1,600kw. It is planned to be used for 10-ton helicopters or 20-ton turboprops like the Y-7, which is not the same level."

"As for 4,000kW-level turboprops... I'm not talking about those imaginary ones. They really don't have a high priority at the moment."

"We actually have R&D capabilities here, but the number of technical personnel is still too small, and we can't take care of too many models at the same time. We can only work on the most urgently needed ones first..."

"..."

When the Air Police 200 landed back, it was almost three hours later.

According to Chang Haonan's test requirements, the aircraft must conduct four rounds of searches for targets from different angles, that is, two zigzag routes that are generally perpendicular to each other.

For the Air Police 200, which has a somewhat limited range, this is even the longest flight since the first flight of the prototype...

While the ground crew rushed over to perform maintenance on the aircraft, the engineer on board had also taken down the recorded radar data and handed it to Chang Haonan.

Naturally, it is impossible for supercomputing to be used here in Xingcheng Base.

Fortunately, if you just want to check the results or simply adjust parameters, you can use an ordinary computer.

But if the algorithm needs to be improved in the future, then we have to go back to the capital and do it again.

"We marked targets 1-39 in the order we completed the search for the first time."

The radar operator responsible for following up on the test flight was a naval lieutenant colonel named Su Changyun. He was sitting behind Chang Haonan and Wang Xiaomo, introducing the details of the flight just now:

"Originally, there were 35 false targets plus 3 ships, a total of 38, but we found an illegal fishing boat in the target waters, so one more was added."

"Then this fishing boat ended up..."

Although this matter had nothing to do with the test, Chang Haonan still couldn't restrain his curiosity.

"We contacted the 110 Dalian ship through the North Sea Fleet Command, and finally caught them and found out that they were a group of people stowaway."

"Okay...that's pretty good."

Wang Xiaomo next to him showed a helpless smile:

"To be able to become the first combat result under the command of an early warning aircraft in the history of our country is to a certain extent recorded in history..."

"And it also provides us with a typical signal characteristic of a fishing boat."

Chang Haonan added next to him:

"In the future, you can add it to the white list of anti-ship weapons, or at least arrange the aiming coefficient to the end to avoid accidental hits..."

Naturally, other people here didn't know about Xiongfeng 3's fishing boats, so they just thought he was preparing for a rainy day.

After a brief interlude, everyone returned their attention to the radar data.

There are two sets of data in total, one is original and the other is processed by the recognition algorithm.

In addition to being used for comparison, the difference between the two sets of data itself can also be used as a basis for judging the nature of the target——

Since the radar display is not specially set, the result output to it is a dot.

But in the background running log, the difference can be clearly seen.

The most intuitive thing is that the latter has richer, or clearer, data levels than the former.

"Two targets 1 and 2, four directions, each with two sets of data. The first set was directly hidden in the sea clutter and was not seen. The second set was seen but very faint. It should be a simulated periscope... "

"The RCS of numbers 3 to 16 are quite large, but it was revealed in the second set of data that they are completely different from the signal characteristics of the three targets 36, 37, and 38... considering the difference in quantity , the dozen in front should be different large corner reflectors, and the last three should be the three real warships..."

"17 and 18, the signal characteristics detected in four directions are almost the same. They should be small corner reflectors used to simulate speedboats or patrol boats..."

"..."

After Chang Haonan finished speaking, Su Changyun compared it with the "standard answer" sent by the fleet.

In the end, except for the wrong judgment of "reef" and "small boat", the rest were all correct.

As far as the first test is concerned, the result is quite good -

After all, radar and computers do not know what humans define as targets, they are only responsible for finding the difference.

As for summarizing and analyzing signals with different characteristics, it needs to be calibrated slowly later.

For example, targets with feature A are large and medium-sized surface ships, targets with feature B are ordinary bulk carriers, and targets with feature C are container ships...

In fact, this is exactly the fundamental purpose of Chang Haonan coming here to conduct the test.

Although the situation of the early warning aircraft cannot be directly applied to the remote sensing satellite, the method of feature extraction and selection of calibration anchor points is the same.

"Although it is not perfect yet, it has great potential."

After reading all the results, Wang Xiaomo came to a conclusion almost immediately:

"Especially in the aspect of bait identification, it is absolutely accurate at a glance, but it is a pity that the storage and computing capacity required are too large, and it is not easy to put it on the radar of a fighter."

"That may not be the case."

Chang Haonan was already adjusting the parameters and calibration methods, but he stopped what he was doing after hearing what the other party said:

"The main function of fighter radar is still air search, the background is relatively clean and the target characteristics are more monotonous. I can write a simplified version for the 1473 and 1493 series radars..."

"!"

Wang Xiaomo just said it casually, but he didn't expect Chang Haonan to come up with a feasible plan.

He stood up:

"Then I will contact the 14th Institute. It happens that these radars, oh, and 346, have not been finalized yet. We can take this opportunity to do them together..."