radiothebook Chapter 9 Hidden Dagger
Chapter 9 Hidden Dagger
The project progressed more smoothly than Zuo Cheng had anticipated during the first week after it officially started.
Chen Hao's coding skills are indeed strong—Zuo Cheng gave him the algorithm framework and core formulas, and he was able to transform the theory into a runnable program within half a day. Moreover, the code structure was clean, the comments were clear, and it didn't look like something written by an undergraduate.
Yu Ying also performed admirably. After taking over the real-time decomposition module for high-dimensional matrices, she got the first version running in just three days, with performance exceeding Zuo Cheng's own estimates by 15%. Her expertise in numerical analysis was truly remarkable—she could not only perform the calculations but also analyze the results to uncover potential flaws in the algorithm design.
The three of them had a clear division of labor and met in the lab for two hours every afternoon to exchange progress and discuss problems. Their efficiency was frighteningly high.
The real trouble began on the eighth day.
That afternoon, Zuo Cheng planned to run a set of key multi-scenario comparative tests on the laboratory's simulation platform. This set of tests required access to the BlueStar 5G base station channel measurement database on the platform—a core asset collected by Professor Lin's research group over two years. It contains a large amount of data and covers a wide range of scenarios, making it an indispensable basic resource for channel estimation research.
Zuo Cheng opened the simulation platform, entered his account, and a red prompt popped up:
"Your account does not have sufficient privileges to access this database. Please contact the administrator."
He frowned. Last week, Professor Lin had clearly granted him full access to the lab, and he had used the database before without any problems when running small-scale tests.
He found Song Yuwei and asked, "Who's responsible for database access control?"
Song Yuwei paused for a moment, her tone slightly ambiguous: "Senior Brother Ma."
Zuo Cheng didn't speak, but opened the lab's internal communication group and sent a message to Ma Hao: "Senior Brother Ma, it seems the database permissions for the simulation platform have been adjusted. My account can't access it. Could you please take a look?"
Ma Hao returned fifteen minutes later.
"We're currently upgrading the database security and the access control system has been readjusted. Your account is at the undergraduate level, so you can only access the basic dataset for now. Access to the full database requires core member privileges from the research group, which will take about a week to process."
A week.
Zuo Cheng looked at the message and tapped his fingers lightly twice on the table.
Database security upgrade? No one in the entire research group had mentioned this, and Song Yuwei looked completely bewildered. Moreover, the term "undergraduate level" was quite interesting—the lab's access control system didn't have any academic level classifications; there were only "members" and "visitors."
This isn't a security upgrade; it's a deliberate attempt to block him.
The one-week approval cycle precisely hit the critical juncture of his multi-scenario testing. Without a complete database, his comparative tests couldn't run; without running the tests, the solution couldn't undergo final performance verification; without completing the verification, the thirty-day deadline would be at risk.
Ma Hao didn't openly sabotage him; in fact, he spoke very politely. But the knife was hidden in cotton, stabbing without drawing blood.
Zuo Cheng leaned back in his chair and closed his eyes for a few seconds.
Then he stood up and went straight to Lin Zhiyuan's office.
"Professor Lin, there's something I'd like to confirm with you." He explained the situation simply, without embellishment or blaming anyone, just stating the facts—the database access had been adjusted, and he couldn't access it now, while multi-scenario testing was a crucial part of the project.
Lin Zhiyuan's expression didn't change much after hearing this; he simply took off his reading glasses and slowly wiped them.
"Log in directly with my administrator account." He took a slip of paper from the drawer, on which was written the username and password. "I haven't approved the database permission adjustments; I'll look into it."
It's that simple.
After obtaining administrator privileges, Zuo Cheng returned to the lab and logged into the simulation platform, where he had unrestricted access to the complete database.
He noticed that Ma Hao's seat was empty—he had probably received some news and left early.
Song Yuwei walked over, placed a glass of water on Zuo Cheng's desk, and whispered, "Teacher Lin just made a phone call. She wasn't speaking loudly, but I heard her say, 'Ma Hao, come to my office.'"
Zuo Cheng nodded but did not reply.
That's the end of it. He won't pursue this further, nor does he need to. Professor Lin isn't a fool; Ma Hao's days in the research group will only get harder from today onward—not because of what Zuo Cheng did, but because Ma Hao himself crossed the line.
He opened the simulation platform and started running multi-scenario tests.
Data was output in batches, and the results were even better than expected. The algorithm's channel estimation accuracy in the three core scenarios of densely populated urban areas, highways, and indoor coverage all exceeded the current industry benchmark. The improvement was most significant in the high-speed mobile scenario—the estimation error at 150 km/h was 37% lower than that of Blue Bay Communications' existing solution.
Zuo Cheng sent a screenshot of the results to Chen Hao and Yu Ying.
Chen Hao replied with a string of exclamation marks.
Yu Ying's reply was more restrained, but more informative: "I've carefully reviewed the data for the high-speed scenario. The switching logic of your anomaly detection module has a slight jitter during the signal recovery phase at the tunnel exit. It doesn't affect overall performance, but it would be perfect if it could be optimized away. I have an idea; let's discuss it tomorrow."
Zuo Cheng replied with a single word: "Okay."
He enjoyed working with Yu Ying. It wasn't the kind of insincere flattery; it was a genuine, unspoken understanding where they could truly engage in technical debate. She could spot problems and dared to point them out—a quality even rarer in academia than technical ability itself.
When they returned to the dormitory that evening, Zhang Lei excitedly came over.
"Brother Cheng! We've found him!"
"Who is it?"
"Didn't you tell me to keep an eye out for embedded systems developers?" Zhang Lei pulled out his phone and found a post on the campus forum. "A third-year computer science student named Fang Ze, his forum ID is 'Compiler Maniac.' This guy recently posted an open-source project on the forum—he wrote his own microkernel for an embedded real-time operating system. The code isn't large, but the design is incredibly brilliant. A bunch of graduate students from the computer science department are asking about the technical details in the comments."
Zuo Cheng took the phone and looked at some of the code and design documents posted in the thread.
The code style is clean and concise, variable naming is standardized, and comments are succinct. What he cares about most is the architecture design—this microkernel task scheduler uses an unconventional priority inheritance scheme, cleverly avoiding the classic priority inversion problem.
Someone who can write something like this in their junior year of college is either exceptionally talented or has put in a lot of hard work. Either way, they are the kind of people Studio 402 needs.
"Can you get in touch with me?"
"I've already messaged him privately. He said we could talk, but he's not short of money and isn't interested in just taking on freelance work."
"Then let's not talk about taking orders." Zuo Cheng handed the phone back to Zhang Lei. "I'll arrange to meet him and we can talk about something else."
Zhang Lei chuckled: "Brother Cheng, are you trying to poach someone?"
"It's not about poaching people, it's about making friends." Zuo Cheng lay back on the bed, hands behind his head. "Talented people don't like being used as tools. We need to make them feel that being with us is meaningful and that they can grow; money is secondary."
Zhang Lei nodded, seemingly understanding but not quite.
Zuo Cheng closed his eyes, but his mind kept racing.
Fang Ze – Embedded Development – Real-time Operating Systems
The light gray fourth blade on the system panel is called "Embedded System Design".
If we can have an in-depth conversation with Fang Ze, the technology radar might come in handy.
Of course, this is on the premise that his skills are indeed worth replicating.
But Zuo Cheng had a gut feeling that this person was not simple.
My phone rang; it was a message from Yu Ying.
"I just got back from the lab and reviewed your test data again. I made a discovery—your algorithm performs exceptionally well in low signal-to-noise ratio environments, almost abnormally so. I checked the data twice and found no errors, but if this result is true, it means that your adaptive tracking mechanism may be producing some kind of unexpected synergistic effect under extreme conditions. This is worth exploring further; it could be a new discovery."
Zuo Cheng suddenly sat up.
The unusual performance at low signal-to-noise ratios—if it is indeed a characteristic of the algorithm itself rather than a data error, then the value of this discovery may far exceed the research topic itself.
He replied to Yu Ying with three words: "Let's talk tomorrow."
Then he lay back down, staring at the ceiling, and the corners of his mouth unconsciously curved into a smile.
The project is accelerating, the team is taking shape, and unexpected discoveries are emerging.
The person who was lurking in the shadows and causing trouble had already been called away by the old professor with a single phone call.
This game is getting more and more interesting.