I want to solve the first example of the article, the logistic map. In this problem I have to generate the signal through the logistic map function, convert this signal (Time series) to a symbolic series (0 and 1), after this step, I need to divide the signal into 50 parts, with each part having a l= 1000 points, which would theoretically give me a matrix of 50 columns per 1000 points. The conversion base used is an L=6 (Symbol Tree). Using the formula l-(L-1) would have the number of possible new combinations, in this case 1000-(6-1) = 995 6-bit combinations. After these combinations I would convert to decimal numbers, obtaining 995 whole numbers, (All this for just one column), the next step would be to check the number of times each number is repeated and then I would have a graph, in all I would have 50 superimposed graphs, the which theoretically should be similar to figure 1 of the attached article.
----The time series has already been converted to symbolic series and separated into 50 columns per 1000 rows.
Step - subdivide the symbolic series into 50 parts. Each part will have l=1000 points;
or 'each column', Use an L-6, (means group the values in group of 6, following the logic of the image of the first post;
According to the article, using the formula l-(L-1) it is possible to find the number of possible combinations. For this case 995 combinations and consequently 995 integers, formed after conversion to decimal base;
after converting to decimal base, it is necessary to count the number of times each partition number is repeated. These values must be stored in order to generate a graph;
steps 5 and 6 must be performed for all 50 columns at the end of the process I will have 50 overlaid graphs generating the symbol spectrum graph.with the step by step, I hope I have helped to better understand the problem. Thank you for your help.
