Rebirth: Starting from Lighting Up the Tech Tree

Chapter 7 Selection Defense

On Monday morning, Zuo Cheng received an official notification from the college's academic affairs office—the selection and defense for the Lanwan Communications horizontal research project was scheduled for 2 PM on Wednesday in the academic lecture hall on the third floor of the School of Telecommunications. A total of six people were participating in the selection: four graduate students and two undergraduate students.

Zuo Cheng is one of the undergraduate students.

He also recognized the name of another undergraduate student—Zhao Kai.

"Zhao Kai is participating too?" Zhang Lei almost spat out his steamed bun when he saw the list. "What level is he at that? How dare he go?"

"They've already secured a spot in the graduate program, and their resume includes two lab experiences, so they're qualified," Chen Hao said objectively.

Zuo Cheng paid no attention to Zhao Kai. What he cared about was the third name on the list—Ma Hao.

For a second-year doctoral student to participate in the selection of industry-funded research projects is a significant advantage. The research direction offered by Lanwan Communications is "Optimization of Intelligent Channel Estimation Algorithms in 5G Scenarios," which highly overlaps with the current research direction of Professor Lin's research group. Ma Hao has been doing related work for more than two years, and in terms of qualifications and experience, he is indeed the strongest among the six.

But Zuo Cheng wasn't worried at all.

Because what's in his mind can't be measured by the word "seniority".

On Tuesday evening, Zuo Cheng prepared his defense materials in his dormitory. He didn't make a PowerPoint presentation—Blue Bay Communications' selection rules were that each person had fifteen minutes for a free presentation followed by ten minutes for questions, with no restrictions on the format. Most people would make a PowerPoint presentation, but Zuo Cheng chose a different approach.

He printed out an eight-page technical proposal with only the title and name on the cover, without any fancy decorations.

The content is divided into three parts: The first part is an analysis of the current status of the field of 5G intelligent channel estimation, which is concise to only two pages, but every sentence hits the key point; the second part is a new algorithm framework he developed based on the channel modeling scheme, which directly addresses the needs of the Lanwan Communication project; the third part is preliminary simulation verification data, using real results he obtained from Professor Lin's laboratory.

Chen Hao leaned over and glanced at the algorithm framework in the second part. After a long silence, he said, "Brother Cheng, this is enough for a paper. Isn't it a bit of a waste to use it for a defense?"

"No waste." Zuo Cheng put the proposal into a file bag. "Whether I get the research project or not will determine my resources and platform for the next six months. I'll have plenty of time to publish the paper later."

At 1:40 pm on Wednesday, Zuo Cheng arrived at the academic lecture hall.

The hall was already quite full—besides the six candidates, there were several teachers from the School of Telecommunications, and three people from Lanwan Communications. In the center was a man in his forties, with a square face, wearing frameless glasses, and a name tag that read "Han Zhe, Deputy Director of Lanwan Communications Technology R&D Center".

Ma Hao sat in the front row, impeccably dressed in a suit and shirt, his laptop open with a PowerPoint presentation already on standby. Seeing Zuo Cheng enter, he glanced at him, his lips twitching, but he said nothing.

Zhao Kai, sitting in the back, couldn't resist making a sarcastic remark upon seeing Zuo Cheng: "Hey, Brother Cheng's here too, good luck!"

Zuo Cheng found a seat on the side and sat down.

Yu Ying was also there. She sat in the audience area, and Chong Zuocheng nodded slightly.

At 2:00 PM sharp, the defense began.

The first two graduate students' presentations were fairly standard; their technical approaches were sound, but nothing particularly groundbreaking. Han Zhe listened with a blank expression, occasionally jotting down notes in his notebook.

The third is Ma Hao.

His PowerPoint presentation was 47 pages long, covering everything from an overview of the field to the technical roadmap and experimental planning. The structure was complete, the data was comprehensive, and the pacing was just right. No wonder he's been doing this for over two years; his fundamental skills are truly solid.

When Ma Hao started explaining the core of the algorithm, Han Zhe looked up for the first time and asked a question: "How robust is your algorithm in high-speed moving scenarios? Have you done simulations at speeds exceeding 120 kilometers per hour?"

Ma Hao paused for a moment: "This scenario hasn't been specifically simulated yet, but based on theoretical model deductions—"

"That means it wasn't done." Han Zhe interrupted him, his tone not harsh but direct. "Theoretical deduction is not the same as measured data; you know that better than I do."

Ma Hao's expression darkened for a moment, but he quickly regained his smile: "Yes, President Han, this part does indeed require further verification."

He returned to his seat, looking fairly composed, but Zuo Cheng noticed that his knuckles were white as he gripped the mouse.

The fourth is Zhao Kai.

To be honest, Zhao Kai's preparation wasn't bad—his PPT was well-made, and his speaking skills were good; after all, he was a seasoned professional in social situations. However, his technical solution was clearly a hodgepodge, and the logical connections between the core formulas didn't hold up to scrutiny. Han Zhe asked a question about the convergence proof, and Zhao Kai hesitated for more than ten seconds without answering, finally ending with a hasty "This part is still under improvement."

When he returned to his seat, his face was already flushed. He glanced at Zuo Cheng out of the corner of his eye and realized that Zuo Cheng wasn't looking at him at all.

After the fifth respondent finished, it was Zuo Cheng's turn.

He didn't bring his computer; all he had was the file folder. Before going up to the stage, he handed out copies of the printed technical proposal to the judges one by one.

Han Zhe took the proposal, opened the first page, and raised an eyebrow—this was the first selection defense without a PowerPoint presentation today.

"Good morning, everyone. I'm Zuo Cheng, a senior student in the Department of Electrical Engineering." Standing on the podium, without notes or a teleprompter, he got straight to the point, "I won't be giving a demonstration today; I'll go straight to the solution. There's a diagram on the second page of your documents; it's a summary of my classification and bottlenecks for current 5G intelligent channel estimation algorithms."

Han Zhe turned to the second page, his gaze lingering there.

The image wasn't large, but it was extremely information-dense—it compared six mainstream algorithms based on three dimensions: accuracy, complexity, and scene adaptability. The core shortcomings of each algorithm were marked in red, making them immediately clear.

"Existing methods focus on two directions—either sacrificing computational efficiency for accuracy, or prioritizing efficiency but suffering accuracy collapse in complex scenarios." Zuo Cheng spoke slowly, but every word carried weight. "My solution attempts to find a third path—using adaptive tracking to allow the algorithm to dynamically adjust its structure based on the real-time state of the channel during operation. Instead of pursuing a one-step optimal solution, it approaches local optima at each time step, achieving global performance improvement through cumulative effects."

He turned to the fourth page of the solution and began to explain the core algorithm.

Not a single word wasted. The derivation process was as clear as a surgical scalpel, every step logical and well-founded, with key points supported by simulation data. The expressions of the graduate students in the audience gradually changed from indifference to focus, and then to disbelief.

Han Zhe's pen kept writing on the paper.

Zuo Cheng finished his presentation in twelve minutes, three minutes less than the allotted time.

The lecture hall was silent for a few seconds.

Han Zhe flipped to the last page of the proposal, where the simulation data was displayed. He looked up and asked his first question: "In this simulation, there's a high-speed movement scenario test, at 150 kilometers per hour. Was the data from a real run?"

"Yes. We're using Professor Lin Zhiyuan's laboratory simulation platform; the raw data and logs are available."

Han Zhe nodded, his gaze sharpening: "Then let me ask you a follow-up question. When your adaptive tracking mechanism experiences sudden changes in channel conditions, such as a sharp drop in signal strength when a vehicle passes through a tunnel, will it experience tracking divergence?"

This is a very tricky problem. Channel mutations are the nemesis of adaptive algorithms—if the tracking speed cannot keep up with the magnitude of the mutation, it will diverge and become uncontrollable.

Zuo Cheng had thought about this problem long ago.

"Yes," he admitted frankly. "Pure adaptive tracking does indeed have the risk of divergence when the channel changes abruptly. So I added an anomaly detection module to the algorithm—when the tracking error exceeds a dynamic threshold, the system automatically switches to conservative mode, replacing the real-time tracking value with the steady-state estimate from the previous moment, and then switches back after the channel stabilizes again. The specific switching logic and threshold calculation method are on page six of the proposal."

Han Zhe turned to the sixth page, looked at it for thirty seconds, and a very subtle smile appeared on the corner of his mouth.

"One last question." He closed the proposal, looked directly at Zuo Cheng, and asked, "Are you a senior undergraduate student?"

"Yes."

Han Zhe didn't say anything more and spent a long time writing on the score sheet.

After the defense, the six candidates waited for the results in the corridor. The atmosphere was subtle—Ma Hao leaned against the wall, eyes closed, his expression unreadable; Zhao Kai looked down at his phone, clearly lacking confidence; the other graduate students whispered among themselves.

As Yu Ying walked out of the lecture hall, she slowed her pace slightly as she passed Zuo Cheng.

"That anomaly detection module you just mentioned," she said in a low voice, "is very clever. I haven't seen this kind of processing in the relevant literature."

"Improvement on the spot."

Yu Ying couldn't help but chuckle: "You always say it's improvisation."

Fifteen minutes later, the results came out.

A teacher from the Academic Affairs Office came out and announced: Zuo Cheng was selected as the candidate for the horizontal research project cooperation of Lanwan Communications.

The corridor fell silent for a moment.

Several graduate students glanced over, their expressions varying. Zhao Kai nearly dropped his phone; he opened his mouth, then closed it again.

Ma Hao opened his eyes and glanced at Zuo Cheng. His glance held no anger, no resentment, only a cold, scrutinizing gaze—as if reassessing the threat level of an opponent.

Zuo Cheng didn't look at him.

Han Zhe walked out of the lecture hall, went straight to Zuo Cheng, and extended his hand.

"Zuo Cheng, our company will coordinate the details of the subsequent cooperation with the college. However, there's one thing I'd like to ask you privately first—would you be interested in interning at Blue Bay Communications' R&D center this winter break? The compensation is excellent."

Winter break internship. Blue Bay Communications R&D Center.

Zuo Cheng shook hands with him: "Thank you, Mr. Han. I'll think about it."

"Don't think about it for too long." Han Zhe smiled. "We're short-staffed."

He turned and left. Zuo Cheng stood in the corridor when his phone vibrated.

A system light screen popped up in my consciousness:

[Host detected to have obtained enterprise-level technology cooperation opportunities; main quest chain about to begin—]

[Note: The main quest chain differs from side quests; once started, it cannot be abandoned midway.]

[Confirm activation?]

Zuo Cheng's fingers tightened slightly.

Main quest chain.

This is the first time the system has displayed this concept.