The first type: directly use the content keyword+"overview, survey" to directly search in major databases and google.

The advantage of this method is simplicity.

There are two disadvantages:

First, sometimes the content of the search is more, and it is easy to fall into the ocean of literature;

Second, there are no words such as overview and investigation in some review documents, so it is easy to miss such documents.

The second type: directly use content keywords to search in SCI.

Because the documents retrieved by SCI are classified as review documents when the value of the cited documents is greater than 1 (I don't know how many). When we use content keywords to search, we can use the literature analysis function of SCI, and select the literature type as the survey to search for the second time, and then we can get the comprehensive literature about a certain content. Moreover, we can also judge the weight of literature according to the number of citations.

The third type: search in the journals that publish review documents.

Some journals are specialized in publishing review articles, and we can directly find the needed review documents through the way of "content keywords+designated journals". The advantage of this method is that the literature quality is better. Compared with general journals, this kind of review articles published in review journals are higher in both the author of review documents and the professional level of review.

In the field of communication, suitable for beginners, specialized journals that publish review documents are:

1.Proceeding There are many contents in this journal, and communication is only one of them.

2.IEEE Signal Processing Magazine-Just like the name of the magazine, this magazine mainly focuses on signal processing.

3.IEEE communication investigation and research. The course, a review journal of communication, can find the required review literature through the way of "content keywords+designated journals" in IEEE search.

Step 1

Read the summary paper, read the abstract first, then read the conclusion. If you can read the summary and conclusion, start reading the introduction.

Step two?

After reading the introduction, if you are still interested, you can continue to read the text; Otherwise, if you are not interested, then you don't need to read the text. When reading the Introduction, try to form some framed structures in your mind, such as what is the structure of the question and the organizational structure of the article.

Step three?

After reading a review article, if you can understand more than 60% of the content of this article and are interested in trying to do it in this direction, all you need to do is download the references mentioned in introduction from IEEE and read them (note that they are references in introduction, not all references). This is the introduction of a good comment that I didn't notice before. The cited documents are often symbolic documents in this field, which are usually called classic documents.

Fourth step

After reading a certain number of references (about 8 articles), you will have a little understanding of the most advanced research status in this field. The next thing to do is to ask yourself if you have any new ideas in this direction, which can be the improvement of existing algorithms, the adjustment of existing assumptions and so on. Unless you are a great person or have a solid basic knowledge, you usually don't have much creativity in the first period after starting a new field (because you know too little). At this time, all you need to do is keep reading the references you have read, the references in the references, the references in the references … until you can put forward new ideas and ideas. Usually, if you can ensure that you have read all the references in a review, and the understanding degree of each article is above 70%, or if you have read all six articles, the understanding degree of each article is above 90%, I believe you will naturally have new ideas.

Extended content:

Literature review is a kind of academic paper that collects a large number of relevant materials in a certain field, a certain major or a certain discipline, problem or research topic, and makes a comprehensive introduction and elaboration by analyzing, reading, sorting out and refining the latest progress, academic viewpoints or suggestions of the current discipline, problem or research topic.

Literature review is a comprehensive analysis, induction and comment on the current research situation (including main academic viewpoints, previous research achievements and research level, debate focus, existing problems and possible reasons, etc.). ), new levels, new trends, new technologies, new discoveries, development prospects, etc. Based on extensive reading and understanding of the literature in the research field involved in the topic selection, I put forward my own opinions and research ideas. It requires the author not only to comprehensively sort out and state the main points of the materials he has consulted, but also to make a more professional, comprehensive, in-depth and systematic discussion and corresponding evaluation of the comprehensively sorted documents according to his own understanding and understanding, rather than just "piling up" academic research in related fields.

References:

Literature Review Baidu Encyclopedia

Blog "How to find and read summary articles"