Mehmet Fidan

Randomness analysis of Antimycielski number generator

Random number generation is one of the important issues of cryptography. Based on the efficient Mycielski predictor, we propose a new random number generation algorithm, denoted by Anti-mycielski, which generates a new data at the end of a sequence by making it orthogonal to the prediction. By taking the initial sequence as a key, it is possible to use the algorithm for encryption purposes by masking. The algorithm works on binary sequences and each 8 bit block is converted to integers. Since the complexity of the algorithm is of non-polynomial order, it is necessary to occasionally chop the history of the predictor. The effect of history length and the length of the key is analyzed in terms of the quality of the generated random sequence.

Mycielski approach for synthetic wind speed data generation

In this paper, a novel Mycielski based approach for wind speed data generation is developed and presented. The efficincy of the proposed approach is tested using hourly wind speed data obtained from İzmir region. To test the efficiecy of the approach, the four year-long measured data are seperated into two parts: data belonging to first three years are used for training whereas the remaining one-year data are used for testing and accuracy comparison purposes. In order to compare the efficiency of the proposed generation method, the same data are used in artificial wind speed generation with the classical method of first order Markov chains. Results indicate that the proposed Mycielski based algorithm produces better quality artificial data as compared to previous methods.

Mycielski Based 2d-Predictive Image Coding Algorithm

The Mycielski method is a prospering prediction algorithm which is based on searching and finding largest repeated binary patterns. It uses infinite-past data to devise a rule based prediction method on a time series. In this work, a novel two-dimensional (image processing) version of the Mycielski algorithm is proposed. Since the dimensionality definition of “past” data increases in two-dimensional signals, the proposed algorithm also needs to handle how the boundaries of the pixel cliques are iteratively extended in the neighborhood of a current pixel. The clique extension invokes novel similarity search strategies that depend on the chosen physical distance metric. The proposed prediction algorithm is used for predictive image compression and performance comparisons with other predictive coding methods are presented.

Wind speed prediction with Mycielski algorithm

Wind speed prediction is an important issue in wind related engineering studies. However, the wind data has random behavior like other meteorological events. Therefore, it is difficult to apply conventional statistical approaches. On the other hand, wind speed data has an important feature; large fluctuations from a wind state to a significantly different level is relatively seldom. This feature leads to some patterns that should be exemined in detail. In this study, a novel approach for wind speed modeling using Mycielski algorithm that considers this important future is demonstrated. Developed procedure, predicts future values of wind data by analysing repeatings in the history of data and assumes that the history will be structurely repeated in the future. The prediction capability of the procedure is tested using wind speed data obtained from 2 cities in Turkey: İzmir and Antalya. Reasonable prediction results are obtained and analysis results are reported.

Effects of Temperature and Pressure Information in a Hybrid (Fourier Series / Neural Networks) Solar Radiation Model

Solar radiation modeling is a critical step in efficient management of solar energy. In this study, a novel solar radiation modeling procedure is developed with the a-priori information of temperature and pressure values, which are naturally dependent on solar radiation via indirect atmospheric phenomena. Firstly, daily behavior of hourly solar radiations is considered in frequency domain. Initial nine Fourier series coefficients are calculated for each day. Secondly, various neural networks models are built for prediction of these nine Fourier coefficients using the input data gathered from early morning hours and previous day. Apart from the solar radiation readings, temperature and pressure data are also used for developing a more accurate model. It is concluded that, the support of temperature and pressure data of the region improves the solar radiation model. Finally, differences between the performances of the proposed models reveal correlative relationships between atmospheric parameters and solar radiation.