Diabetes is a life-threatening syndrome occurring around the world; it can have huge complications and is documented by large amounts of medical data. Therefore, attempts at early detection of this disease took a large area of research and many methods were used to deal with diabetes. In this paper, different types of KNN algorithm have been used to classify diabetes disease using Matlab. The dataset was generated by the criteria of the American diabetes association. For the training stage, 4900 samples have been used by the classifier learner tool to observe the results. Then, 100 of the data samples were used for the test. The results show that the KNN types (Fine, Weighted, Medium and Cubic) give high accuracy over the Coarse and the Cosine methods. Fine KNN is considered the most suitable according to its accuracy of classified samples. ABSTRAK: Penyakit kencing manis adalah sindrom penyakit ancaman nyawa yang berlaku di seluruh dunia dan ia mempunyai data perubatan yang besar serta komplikasi tinggi. Oleh itu, cubaan dalam mengesan awal penyakit ini mempunyai potensi luas dalam kajian dan banyak kaedah telah digunakan bagi mengkaji penyakit kencing manis. Dalam kajian ini, pelbagai jenis algoritma KNN telah digunakan bagi mengelas penyakit kencing manis menggunakan Matlab. Setdata dihasilkan berdasarkan kriteria Kesatuan Kencing Manis Amerika. Pada peringkat latihan, sebanyak 4900 sampel telah digunakan oleh pelatih alat pengelasan bagi memantau dapatan kajian. Kemudian, 100 daripada sampel data telah digunakan bagi ujian. Keputusan menunjukkan jenis KNN (Halus, Berat, Sederhana dan Kubik) lebih tepat berbanding kaedah Kasar dan Kosinus. KNN Halus di dapati lebih sesuai berdasarkan ketepatan sampel pengelasan.

No abstract

Today trade wireless communication systems are wanted to equip reliable communication and bigger data rates. Two considerable challenges in system design are the restricted fading and spectrum due to multipath components in the wireless system. Multiple transmission and reception can be used to form (MIMO) multiple input multiple output channels to raise the data rate and capacity. The advantages of utilizing multiple antennas are to get reliable performance and a higher data rate. In this paper, an attempt is made to design and simulate SISO, SIMO, MISO and MIMO systems. We have analyzed and compared the performance of these systems based on SNR and channel capacity in a different number of an antenna. The effect of the number of an antenna (4, 8, and 16) on the channel capacity over a range of SNR (0-30db) for the systems is examined. The system performance is simulated in Matlab. The results of the simulation show that as significant improvement in the capacity of systems reached dramatically to 214 (bits / Hz / Sec) when SNR = 30 and (16X16) MIMO is used.

On this work, the three-port converter derivative from the full-bridge converter is analysed and control. The full-bridge converter is made up of two parts, A and B. Each part contains two switches with independent input sources. The inputs drive the load through a transformer that boosts voltage and provides isolation. The controller is an important portion of the converter. It controls the pass of power between the three ports. The controller uses Fuzzy Logic Control to make decisions regarding the two control signals for each part. These decisions are based on the variance between the whole energy generated by the photovoltaic panel and the energy demand at the load. The proposed design is verified through MATLAB and Simulink.

In this paper, the power flow controller for multi-port DC-DC converter was modified, so it mediator a photocell, a storage element and an isolated load. The MOSFET switches are used to design the presented model to manage the flowing power of load voltage and two sources independently. Theoretical researches are implemented to prove the capability of the controllers to regulate the flowing of power for the photovoltaic generator and the storage element while maintaining the load in three cases. First, the photovoltaic power is greater, less and equal to load resistance and maintaining the switches in the best condition for each case. Finally, the power of the photovoltaic generator follows the direction proposed by the MOSFEST switcher's control.