Signal Separation using Multichannel/Multimodal Side Information

信号分离的目的是从一个或多个信号混合中恢复源信号。这些混合可以来自许多领域，包括所谓的鸡尾酒会问题的语音混合，多种乐器和/或声乐的音乐记录，来自大脑电活动的脑电图（EEG）信号，用于胎儿监测测量表面电位时的母体/胎儿心电图（ECG）混合，来自肌肉运动单位动作电位混合的肌电图（EMG）信号，各种天文数据集，包括弱天文源与背景的混合，以及雷达或声纳系统。很多时候，源分离是一个归纳推理问题，这个问题是高度不确定的，因此存在有限的信息，希望找到一个唯一的解决方案。源分离问题可能伴随着不同级别的信息，包括单通道混合、多通道混合或混合，这些混合伴随着可能有助于分离的侧元数据。例如，音乐表演可以由单个麦克风（单通道）录制，也可以由分布在房间各处的多个麦克风（具有空间多样性的多通道）录制，或者可以访问乐谱和乐器类型（侧元数据信息）。

The aim of signal separation is to recover source signals from one or more signal mixtures. These mixtures can come from a multitude of domains, including speech mixtures for the so-called cocktail party problem, musical recordings of multiple instruments and/or vocals, electroencephalograph (EEG) signals from the electrical activity of the brain, maternal/fetal electrocardiograph (ECG) mixtures when measuring surface potentials for fetal monitoring, electromyography (EMG) signals from mixtures of motor unit action potentials in muscles, various astronomical datasets with mixtures of weak astronomical sources versus background, and radar or sonar systems. Quite often, source separation is an inductive inference problem where the problem is highly underdetermined such that limited information exists for hoping to find a unique solution. Source separation problems may come with varying levels of information, including single channel mixtures, multichannel mixtures, or mixtures that are accompanied with side metadata that may help with the separation. For example, a music performance could be recorded by a single microphone (single channel), by multiple microphones distributed around the room (multichannel with spatial diversity), or access may be had to the musical score and instrument types (side metadata information). The focus for this proposal is to research develop advanced source separation techniques that multichannel and multimodal metadata as side information to improve the quality in source separation. The signal processing research will concentrate on incorporating multichannel and side information into the source separation problem and in evaluating the impact of such side metadata information. Of particular interest is when there are lack of time alignment information wit the metadata, small variations in time-frequency or sparsity information between the mixtures and the metadata, effects of short-time non-stationarities in the source signals being separated, and use of sparsity constraints in multiple domains. These signal separation characteristics will be research and studied under various applications, including speech, vocals/instrument separation, and EMG signal decomposition, but, the key research focus will be on the core components of source separation in multichannel mixtures with side metadata information. Some of the anticipated outcomes are as follows. First, in the vocal/instrument separation space, we expect to be able to take a musical recording, likely as a stereo multichannel recording, along with side metadata, which could include the musical score, sound samples of the instruments in the performance, an example performance of the piece by a secondary musician, a list of lyrics, a MIDI file, or other types of metadata, and be able to separate the individual instruments in the recording to either isolate or suppress individual instruments. In the EMG signal decomposition space, we expect to be able to take multichannel EMG recordings and determine firing times of individual motor units in the muscle and the corresponding shape of the motor unit action potentials. In the fetal ECG separation space, we expect to be able take a multichannel mixtures containing the stronger maternal ECG and weaker fetal ECG, and be able to suppress the stronger maternal ECG signal in an improved fashion. The proposed core research in signal separation will have wider applicability, where these examples would be used to validate the techniques and determine other limitations that may present themselves in one domain but not another.