Abstract: With the continuous expansion of aquaculture scale and the improvement of its intelligent level,traditional methods such as manual inspection and water quality sampling struggle to meet the fine management requirements of modern aquaculture due to their intrusive nature and lack of real-time capability. Passive Acoustic Monitoring(PAM) technology can accurately analyze the behavioral characteristics of aquatic organisms without disturbing them. Centered on acoustic signals,this technology has established an analytical framework covering data collection,signal processing,feature extraction,and pattern recognition ,demonstrating strong adaptability in actual aquaculture environments. Studies have shown that PAM technology has obvious advantages in low -light,deep -water,and turbid environments,and has exhibited application potential in feeding monitoring,reproduction identification,water quality early warning,and other aspects. However,the further development of this technology is restricted by issues such as equipment noise interference,lack of cross - species databases, and insufficient algorithm generalization. Future development should focus on advancing noise reduction and enhancement,multimodal fusion, establishing a standardized data system,and strengthening interdisciplinary collaboration to promote industrialization.

Key words: aquaculture;passive acoustic monitoring, acoustic signal

摘要： 随着水产养殖规模不断扩大和智能化水平提升 ,传统人工巡视和水质采样因干扰性强、实时性不足 ,难以满足现代养殖的精细化管理要求 。被动声学监测(PAM)技术可在不干扰水生生物的情况下 ,精准解析其行为特征 。该技术以声信号为核心 ,构建了涵盖数据采集、信号处理、特征提取与模式识别的分析框架 ,在实际水产养殖环境中展现出较强的适应性 。研究表明 ,PAM 技术在低光照、深水和浑浊环境中具有明显优势 , 已在摄食监控、繁殖识别和水质预警等方面展现应用潜力 。然而 ,该技术的进一步发展受到设备噪声干扰、跨物种数据库缺乏和算法泛化不足等问题的制约 。未来发展应着重推进降噪增强与多模态融合 ,建立标准化数据体系 ,并强化跨学科协作以推动该技术的产业化落地。

关键词: 水产养殖, 被动声学监测, 声学信号