Investigation and summary investigation should be fully investigated before preparing to do a certain direction (topic) The main purpose of the research is to understand the development process of this field, grasp the research status at home and abroad, make yourself stand on the shoulders of giants, take the overall situation into account and be condescending, so as to clearly understand the key problems to be solved in this field and choose a specific problem, which is novel and mainstream. It's better to find a well-researched topic or a small problem that is difficult to break through than a blind eye.

The main method of investigation is to find information: related papers, platforms, tools and applications. Reading "thesis" is to find out the specific problems to be studied, learning "platform" and "tools" is to find out the solutions to problems, and reading "application" is to face the actual needs and avoid behind closed doors.

SIGCOMM、MOBICOM、SENSYS、NSDI、MOBISYS、ICNP、INFOCOM

Ieee trans。 Network, wireless network, information theory, etc.

These papers basically represent the highest research level in this field at present. If you often read these papers, your appreciation will gradually improve. 2) The citation rate is high. Some papers are not very well-born, but they have great influence, and the citation rate (he cited) is as high as 1000 times. 3) Author Niu. The writer is a community bull. Usually 1), 2) and 3) are equivalent. 4) timeliness. With the rapid development of network technology, the views of the first three years may have become products long ago and have no research value. But there are also some good papers that have accumulated and become the basis of research.

According to the above definition, it is not difficult to find valuable papers in two ways: 1.

Find the relevant papers of the top conferences in recent three years, find the high-level papers he cited, and find the relevant papers published by its authors and members of the research group in recent years. 2.

Looking for the latest review articles and reading reviews are shortcuts to understand the development of this field. You can find the high-level papers cited in the review, and then find the high-level papers since the publication of the review paper. Also pay attention to the level of summarizing the article. But even a high-level paper, reading a summary paper is like eating chewed steamed bread. Therefore, reading a summary paper cannot replace reading the original text.

Read these papers carefully, and other papers only read abstracts to understand their basic ideas for reference.

Don't intensively read papers other than "valuable papers", which will not only waste time, but also catch its bad habits. The so-called "near Mexico is black, near Zhu Zhechi is black".

Read it briefly first to find out the general idea and whether it is related to your topic.

Read the abstract first.

Read the introduction again.

Finally, look at the conclusion. 2.

Summary is a by-product of research, which can also be said to be a research report. In the process of research, record your own ideas, summarize, analyze and compare them, and organize them into research reports. If the research is done well, you can put forward the specific problems you want to solve. The research report can be used as the opening report. If well written, on the one hand, it can be published as a small paper, on the other hand, it can be used as the content of the first chapter of the graduation thesis in the future.

As researchers, we can start to work and solve the problems raised. When you encounter some problems at this stage, you often need to check the information again. Learning without thinking is useless, thinking without learning is dangerous,

Only by reading more, thinking more and doing more can we gradually improve our scientific research level and working ability.