Rebirth: Starting from Lighting Up the Tech Tree

Chapter 71 Breakthrough

Three days later, Ma Hao submitted the feasibility analysis report.

Zuo Cheng opened the report and studied it carefully. Ma Hao's experience in Shenzhen was indeed extensive; the report not only contained a detailed algorithm architecture but also specific implementation paths and expected performance evaluations.

According to the report's estimates, introducing machine learning can improve the spectral efficiency of adaptive coding and modulation by 18% to 25%. This figure is enough to give the 402 solution a significant advantage over Huaxin in the competition.

Zuo Cheng convened a meeting with the technical team to discuss the contents of the report.

"I think Ma Hao's plan is feasible," Chen Hao was the first to speak, "but it requires a lot of computing resources, and our existing servers may not be enough."

How much do you need?

"We'll need at least two more high-performance servers," Chen Hao said. "Training neural networks is very computationally intensive."

Zuo Cheng nodded: "Buy it. The bidding is imminent, money is not a problem."

Zhang Lei raised his hand: "I have another concern. Training machine learning models takes time. If we start now, will we have enough time to produce results before the bidding process?"

"Time is indeed tight," Ma Hao said, "but if we use the transfer learning approach, we can greatly shorten the training time. We can pre-train using a publicly available dataset and then fine-tune it using Tianqiong's data. This way, we can get results in one to two weeks."

Zuo Cheng pondered this in his mind. Two weeks, plus the preparation and testing of the plan, was indeed a tight schedule. But if successful, the rewards would be enormous.

"Let's do it." He slammed his hand on the table. "Ma Hao, you're in charge of leading the algorithm development. Chen Hao, you're in charge of hardware and data preparation. Zhang Lei, coordinate resources; just let me know if you need anything."

"clear!"

After the meeting, Zuo Cheng returned to his office. He opened the system panel to check the current status.

Points: 118.

Zuo Cheng pondered to himself. Ma Hao's algorithm was valuable, but with further optimization, the effect would be even better. He remembered that the system had a technology fusion function, so he could try to integrate different technological blades.

He checked the currently active leaves. The communications engineering branch has fourteen leaves, covering multiple fields such as signal processing, encoding and modulation, and network protocols.

Zuo Cheng selected two blades: one for adaptive coding modulation and the other for a signal prediction algorithm. Combining these two might produce unexpected results.

[Will we integrate technology?]

[Cost: 5 points]

[Expected Result: Generation of fused blades, increasing efficiency by 1.2 times]

Zuo Cheng confirmed the fusion. Five points were deducted from the total, leaving 113 points.

On the system panel, the two blades began to glow, slowly approached each other, and finally merged together to form a new blade.

【Successful integration】

[Achieving Fusion Blades: Intelligent Adaptive Tuning System]

[Results: Adaptive coding and modulation efficiency increased by 1.2, channel prediction accuracy improved by 15%]

Zuo Cheng's eyes lit up. The effect was even better than he had expected. The efficiency increased by 20%, and with the 20% bonus from the fused leaves, the overall improvement was quite considerable.

He closed the system panel and began to think about how to apply the integrated technology to the solution.

In the following two weeks, 402 entered a sprint phase.

Ma Hao led the algorithm team in working day and night, Chen Hao was responsible for setting up the hardware environment, and Zhang Lei coordinated resources from all parties. Zuo Cheng oversaw the overall situation while maintaining close communication with Professor Li's team.

Yu Ying also joined in. Although she still had classes, she would go to 402 whenever she had free time to help organize documents and test data.

"You don't need to work so hard," Zuo Cheng said with concern. "You also have your own research to do."

"My research is related to the Sky Dome project," Yu Ying said. "Getting involved is also a learning experience. Besides, I want to help you."

Zuo Cheng didn't say anything more, he just squeezed her hand.

Two weeks later, the algorithm training was completed.

The test results are in: the spectral efficiency improved by 23%, exceeding Ma Hao's initial expectations. This result is enough to make the 402 solution technically superior to Huaxin's.

Zuo Cheng looked at the test report and felt full of confidence.

"I'm preparing to write the tender document," he said. "We'll clearly explain our technological advantages, especially our adaptive coding modulation. Huaxin doesn't have this; it's our trump card."

Han Lu nodded: "I'll start preparing now."

Writing a tender document is a massive undertaking. Besides the technical proposal, it includes multiple sections such as a business plan, implementation schedule, and team introduction. Han Lu organized a dedicated writing team, with each member responsible for a different section.

Zuo Cheng personally oversaw the technical aspects. He included the integrated intelligent adaptive modulation system in the plan, but did not mention the system's existence, instead describing it using conventional engineering terminology.

"The core of the adaptive modulation algorithm is multi-parameter joint optimization," he wrote in the proposal. "By introducing a channel prediction mechanism, the system can sense channel change trends in advance, thereby selecting the optimal modulation and coding scheme. The advancement of this technology has already been verified in the second phase of Sky Dome."

This passage may sound unremarkable, but an expert can immediately recognize the technical sophistication behind it.

After the tender was completed, Zuo Cheng organized an internal review. Professor Li Weimin personally participated and offered some suggestions for revision.

"Overall, it's very good," Professor Li said. "The technical approach is clear, and the innovations are prominent. But there's one area that needs attention."

"You said."

"Lin Jianhua is the CTO of Huaxin Group on the bid evaluation committee," Professor Li paused, "He personally joined the committee and even has veto power."

Zuo Cheng nodded. He had already considered this issue. Lin Jianhua was Huaxin's top technical leader and Zuo Cheng's nightmare from his previous life. Now that he was personally overseeing the bidding committee, he was clearly determined to win.

"Although Lin Jianhua has a veto power," Professor Li said, "he generally doesn't use it lightly. Unless your proposal has obvious flaws, or some unexpected situation arises."

Zuo Cheng understood. This meant that their plan had to be flawless, leaving Lin Jianhua no opportunity to exploit.

"We'll be careful," he said. "The plan will be checked three more times to make sure there are no mistakes."

Professor Li nodded in satisfaction: "That's good. I trust you."

After the review concluded, Zuo Cheng returned to his office. Looking out the window at the sky, he felt both anticipation and pressure.

The third phase of the Sky Dome project is the largest project in the history of 402 Engineering Division. Winning the bid would mean that 402 Engineering Division would officially join the ranks of the top tier in the industry. Losing would mean that all the previous efforts would be greatly diminished.

He opened the system panel to check his current points. 113 points, enough to handle any unexpected situations.

Zuo Cheng took a deep breath and turned off the control panel.

He must win this battle.