Hasil Pencarian  ::  Simpan CSV :: Kembali

"This text discusses various applications of space time adaptive processing, including applications in OTH-radar, ground target tracking, STAP in real world clutter environments, jammer cancellation, supper-resolution, active sonar, seismics and communications. The book provides a unique overview of the broad field of space-time processing and is divided into two parts: the first dealing with the classical adaptive suppression of airborne and space-based radar clutter, and the second comprising miscellaneous applications in other fields such as communications, underwater sound and seismics."

"This third edition provides a detailed introduction to the fundamentals of space-time adaptive processing, with emphasis on clutter suppression in airborne or space based phased array radar, covering specifically the principles of airborne or space based MTI radar for detection of slow moving targets for use in the fields of earth observation, surveillance and reconnaissance, with special attention paid to clutter rejection techniques. The book includes topics such as signal processing, clutter models, array processing, bandwidth effects, non-linear antenna arrays, anti-jamming techniques, adaptive monopulse, bistatic radar configurations, SAR and ISAR, and sonar. After the success of the first and second editions, this third edition has been extensively updated and extended to reflect the numerous advances in the field. A completely new chapter has been added on the impact of the radar range equation, which is of particular importance for radar system designers."

"Banjir di Jakarta merupakan masalah yang kompleks yang dipengaruhi oleh kombinasi faktor geografis, sosial, ekonomi, dan lingkungan. Studi ini berfokus pada prediksi banjir dengan membandingkan data stasiun darat Automatic Rain Gauge (ARG) dan data satelit Climate Hazards Group InfraRed Precipitation (CHIRPS) menggunakan Adaptive Neurofuzzy Inference System (ANFIS) yang terintegrasi dengan Principal Component Analysis (PCA). Dataset mencakup pengukuran curah hujan dari ARG dan CHIRPS, serta data ketinggian air dari tahun 2014 hingga 2020. ARG menyediakan data curah hujan lokal yang akurat, sementara CHIRPS menawarkan cakupan curah hujan regional yang luas. Teknik praproses seperti imputasi rata-rata, normalisasi data, dan metode interquartile range (IQR) digunakan untuk meningkatkan kualitas data. Model ANFIS-PCA, yang mengintegrasikan logika fuzzy dan pelatihan jaringan saraf tiruan, diterapkan dengan pembagian data 80:20 untuk pelatihan dan validasi. Ketika dilatih dengan data stasiun darat ARG dan pengukuran ketinggian air, model ANFIS-PCA menunjukkan akurasi yang superior, dengan root mean square error (RMSE) sebesar 0,13, mean absolute error (MAE) sebesar 0,12, dan R² sebesar 0,82. Sebaliknya, model ANFIS tanpa PCA menghasilkan kesalahan yang lebih tinggi, dengan RMSE 6,3, MAE 6,2, dan R² 0,74. Pelatihan dengan data satelit CHIRPS menghasilkan kesalahan yang jauh lebih tinggi (RMSE 30,14, MAE 24,05, R² 0,42). Sedangkan hasil ANFIS – PCA menghasilkan akurasi yang lebih bagus (RMSE 4,8, MAE 2,0 dan R² 0,55) . Hasil penelitian menunjukkan bahwa ANFIS-PCA memiliki kinerja yang lebih baik dibandingkan model ANFIS tanpa PCA, terutama ketika dilatih dengan data dari stasiun darat. Integrasi PCA berhasil mengurangi dimensi data, meningkatkan efisiensi komputasi dan akurasi model. Selain itu hasil ini juga menegaskan keunggulan pengukuran curah hujan data ground station untuk prediksi banjir, mempunyai angka presisi yang lebih tinggi dan kerentanan yang lebih rendah terhadap kesalahan dibandingkan data satelit. Sementara itu data satelit CHIRPS menawarkan cakupan spasial yang lebih luas.

---

Flooding in Jakarta is a complex issue influenced by a combination of geographical, social, economic, and environmental factors. This study focuses on flood prediction by comparing ground station data from Automatic Rain Gauges (ARG) and satellite data from the Climate Hazards Group InfraRed Precipitation (CHIRPS) using the Adaptive Neuro-Fuzzy Inference System (ANFIS) integrated with Principal Component Analysis (PCA). The dataset includes rainfall measurements from ARG and CHIRPS, as well as water level data from 2014 to 2020. ARG provides accurate local rainfall data, while CHIRPS offers broad regional precipitation coverage. Preprocessing techniques such as mean imputation, data normalization, and the interquartile range (IQR) method were employed to enhance data quality.

The ANFIS-PCA model, which integrates fuzzy logic and neural network training, was implemented using an 80:20 data split for training and validation. When trained with ARG ground station data and water level measurements, the ANFIS-PCA model demonstrated superior accuracy, achieving a root mean square error (RMSE) of 0.13, mean absolute error (MAE) of 0.12, and R² of 0.82. In contrast, the ANFIS model without PCA yielded higher errors, with RMSE of 6.3, MAE of 6.2, and R² of 0.74. Training with CHIRPS satellite data resulted in significantly higher errors (RMSE 30.14, MAE 24.05, R² 0.42). Meanwhile, the ANFIS-PCA model trained on combined datasets showed improved performance, achieving RMSE of 4.8, MAE of 2.0, and R² of 0.55.

The results indicate that the ANFIS-PCA model outperforms the ANFIS model without PCA, particularly when trained with ground station data. The integration of PCA successfully reduced data dimensionality, improving computational efficiency and model accuracy. Furthermore, the findings reaffirm the superiority of ground-based measurements for flood prediction due to their higher precision and lower susceptibility to errors compared to satellite-derived data, while CHIRPS satellite data offers wider spatial coverage."