{"id":2081,"date":"2019-05-13T15:45:15","date_gmt":"2019-05-13T15:45:15","guid":{"rendered":"https:\/\/blogs.mathworks.com\/deep-learning\/?p=2081"},"modified":"2021-04-06T15:50:46","modified_gmt":"2021-04-06T19:50:46","slug":"deep-learning-for-signal-processing-applications","status":"publish","type":"post","link":"https:\/\/blogs.mathworks.com\/deep-learning\/2019\/05\/13\/deep-learning-for-signal-processing-applications\/","title":{"rendered":"Deep Learning for Signal Processing Applications"},"content":{"rendered":"I\u2019d like to introduce Frantz Bouchereau, development manager for Signal Processing Toolbox who is going to dive deep into insights on deep learning for signal processing, including the complete deep learning workflow for signal processing applications. It's a longer post than usual, but jam packed with actionable information. Enjoy!<\/em>\r\n
\"\"<\/h6>\r\nIntroduction<\/strong><\/span>\r\n
<\/h6>\r\nToday we will highlight signal processing applications using deep learning techniques. Deep Learning developed and evolved for image processing and computer vision applications, but it is now increasingly and successfully used on signal and time series data.\r\n\r\n \r\n\r\nDeep learning is becoming popular in many industries including (but not limited to) the following areas:\r\n
<\/h6>\r\n
\"\"<\/h6>\r\nThe unifying theme in these applications is that the data is not images but signals coming from different types of sensors like microphones, electrodes, radar, RF receivers, accelerometers, and vibration sensors. The collection of large signal datasets is enabling engineers to explore new and exciting deep learning applications.\r\n
<\/h6>\r\n

Signal Processing vs. Image Processing<\/strong><\/span><\/h3>\r\nDeep learning for\u00a0signal\u00a0data\u00a0typically\u00a0requires\u00a0preprocessing, transformation, and feature extraction steps that\u00a0image\u00a0processing\u00a0applications often do not. While large high-quality image <\/em>datasets can be created with relatively low-cost cameras, signal sets are harder to obtain and will usually suffer from large variability caused by wideband noise, interference, non-linear trends, jitter, phase distortion, and missing samples.\r\n
<\/h6>\r\nWhile it is likely that you will be able to successfully train a deep model with a large set of high-quality raw images fed directly into a network, you will probably not succeed doing the same with a limited size, low SNR signal data set.\r\n
<\/h6>\r\n

Knowledge vs. Data<\/strong><\/h3>\r\nWe don\u2019t always have the luxury of having large signal data sets. Signals may be difficult to measure, or the application may not be easily observable. For example, consider the case of predicting remaining useful life of a machine. Fortunately, we live in a time where most machines work properly, so failure data will be less prevalent than healthy data.\r\n
<\/h6>\r\nBeing successful with deep learning for signal processing applications depends on your dataset size, your computational power, and how much knowledge you have about the data. You can visualize this in the figure below:\r\n\r\n\"\"\r\n
<\/h6>\r\nThe larger and higher quality the dataset, the closer you can get to being able to perform deep learning with raw signal data. Similarly, if you have higher computational power, you will be able to train more complex networks that will improve performance.\r\n
<\/h6>\r\nOn the other hand, if you understand your data, you may be successful extracting key features to feed into a traditional machine learning algorithm without the need of a large dataset and of a deep network. This requires more signal processing knowledge, but results can be impressive.\r\n
<\/h6>\r\nBetween the two extremes of raw signal data and crafting a customized machine learning approach are some interesting considerations:\r\n