ĐỊNH HƯỚNG NGUỒN TÍN HIỆU TRÊN CƠ SỞ THUẬT TOÁN SMOOTH MUSIC KẾT HỢP KỸ THUẬT CHỌN LỌC PHỔ ĐẶC TRƯNG CỦA TIẾNG ỒN TỪ CÁC MỤC TIÊU BIỂN CÓ CHÂN VỊT

0 Người đánh giá. Xếp hạng trung bình 0

Tác giả: Ngọc Đông Nguyễn, Thanh Chinh Nguyễn, Khắc Hoan Phạm

Ngôn ngữ: vie

Ký hiệu phân loại:

Thông tin xuất bản: Journal of Science and Technique: Section on Information and Communication Technology, 2024

Mô tả vật lý: tr.7

Bộ sưu tập: Báo, Tạp chí

ID: 462506

Thêm vào giỏ Liên kết toàn văn

Bài báo đề xuất giải pháp cho thuật toán định hướng (DOA) trong các hệ thống sonar thụ động trên cơ sở kết hợp thuật toán Smooth MUSIC và chọn lọc phổ đặc trưng (BS) của các mục tiêu biển có chân vịt, gọi là BS-SMUSIC. Các hài của tần số trục và tần số cánh được xem là các thông tin có ích để phát hiện mục tiêu và ước lượng các thông số của chúng, bao gồm bài toán định hướng. Ghi và phân tích tần số thấp (LOFAR) là kỹ thuật được chọn để thực thi BS. Bằng cách phân tích thuật toán trên cơ sở MUSIC kết hợp với BS, các thuậttoán định hướng trong sonar thụ động được cải thiện đáng kể. Các kết quả mô phỏng được thực hiện với cả hai tình huống: các nguồn không tương quan và các nguồn tương quan. Các kết quả này cho thấy thuật toán BS-SMUSIC có hiệu năng tốt nhất so với ba thuật toán khác trên cơ sở thuật toán MUSIC (MUSIC, Modified MUSIC và Root MUSIC) trong tất cả các tình huống khảo sát: các tín hiệu băng hẹp có tỉ số tín/tạp (SNR) bằng 5 dB
tín hiệu băng rộng có SNR bằng -5 dB. Hơn nữa, thuật toán BS-SMUSIC cũng đạt được sai số trung bình bình phương (RMSE) nhỏ nhất trong dải khảo sát của SNR (-5 dB đến +20 dB). Trong dải này, RMSE của BS-SMUSIC giữ ổn định và chỉ bằng 0.3◦ tại SNR bằng -5 dB. Hiệu năng này khẳng định khả năng làm việc ổn định, siêu phân giải và độ chính xác cao của BS-SMUSIC cho các hệ thống sonar thụ động.This article demonstrates a proposal for a direction of arrival (DOA) estimation algorithm for passive sonar systems based on the combination of the smooth multiple signal classification (Smooth MUSIC) algorithm and frequency bin selection (BS) of the sound of marine propeller-equipped targets, so-called BS-SMUSIC. The harmonics of propeller shaft and blade frequencies are considered useful information for detecting targets and estimating their parameters, including DOA. Low frequency analysis recording (LOFAR) is utilized to implement BS. By analyzing MUSIC-based algorithm together BS, passive DOA estimation algorithms have been significantly improved. Simulation results have been given in both cases: uncorrelated and correlated sources. Outcomes demonstrate that the BS-SMUSIC algorithm has the best performance in comparison to three MUSIC-based algorithms (MUSSIC, Modified MUSIC and Root MUSIC) in all investigation cases: narrow band signals at 5 dB of signal to noise (SNR)
wideband signals at -5 dB of SNR. Besides, BS-SMUSIC algorithm also archives the lowest root mean square error (RMSE) in SNR range -5 dB to 20 dB. In the SNR range, the RMSE of BS-SMUSIC algorithm remains stable and is about 0.3◦ at -5 dB of SNR. This performance reflects its ability to ensure stable operation, super resolution and high accuracy of BS-SMUSIC algorithm for the passive sonar systems.

1. Đọc trực tuyến

1. Đọc trực tuyến

Tạo bộ sưu tập với mã QR