| Budget Amount *help |
¥43,654,000 (Direct Cost: ¥33,580,000、Indirect Cost: ¥10,074,000)
Fiscal Year 2022: ¥5,330,000 (Direct Cost: ¥4,100,000、Indirect Cost: ¥1,230,000)
Fiscal Year 2021: ¥5,954,000 (Direct Cost: ¥4,580,000、Indirect Cost: ¥1,374,000)
Fiscal Year 2020: ¥7,670,000 (Direct Cost: ¥5,900,000、Indirect Cost: ¥1,770,000)
Fiscal Year 2019: ¥9,880,000 (Direct Cost: ¥7,600,000、Indirect Cost: ¥2,280,000)
Fiscal Year 2018: ¥14,820,000 (Direct Cost: ¥11,400,000、Indirect Cost: ¥3,420,000)
|
|---|
| Outline of Final Research Achievements |
First, we succeeded in constructing a QNN system that adaptively and accurately classifies ground vegetation in PolSAR. We also developed the theory of PQNN by integrating phase information, and realized neural networks that reliably process PolInSAR data and ground penetrating radar data. Then, we constructed a QNN system that enables channel prediction of mobile communications with practically effective accuracy by introducing polarization information, and demonstrated its usefulness clearly. At the same time, we proceeded with the construction and engineering systematization of the QNN framework that handles electromagnetic and optical polarization information. Based on these experimental results, we made progress in summarizing and systematizing the theoretical framework of QNN and PQNN.
|
