Introduction to Random Processes With Applications to Signals and ...
Building AI Applications for Signals and Time Series Data
Transcript of Building AI Applications for Signals and Time Series Data
![Page 1: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/1.jpg)
Building AI applications for Signals and Time-
Series Data
Esha Shah, MathWorks
Francis Tiong, MathWorks
![Page 2: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/2.jpg)
ARTIFICIAL INTELLIGENCE
MACHINE LEARNINGSupervised and Unsupervised Statistical Models…
DEEP LEARNINGNeural networks, GANs, Autoencoders….
Machine Learning and Deep learning have grown rapidly over the
last decade
2
![Page 3: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/3.jpg)
ARTIFICIAL INTELLIGENCE
MACHINE LEARNINGSupervised and Unsupervised Statistical Models…
DEEP LEARNINGNeural networks, GANs, Autoencoders….
Machine Learning and Deep learning have grown rapidly over the
last decade
2
![Page 4: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/4.jpg)
4
Use of AI in signal processing applications is growing rapidly
![Page 5: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/5.jpg)
Modulation Classification of RF waveforms
4
![Page 6: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/6.jpg)
Modulation Classification of RF waveforms
4
TRANSMITTER
(Software
Defined Radio)
RECEIVER
(Software
Defined Radio)
![Page 7: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/7.jpg)
Modulation Classification of RF waveforms
4
TRANSMITTER
(Software
Defined Radio)
RECEIVER
(Software
Defined Radio)
![Page 8: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/8.jpg)
Modulation Classification of RF waveforms
4
TRANSMITTER
(Software
Defined Radio)
Modulation
Type
RECEIVER
(Software
Defined Radio)
![Page 9: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/9.jpg)
Modulation Classification of RF waveforms
4
TRANSMITTER
(Software
Defined Radio)
Modulation
Type
RECEIVER
(Software
Defined Radio)
![Page 10: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/10.jpg)
AI-driven system design
5
![Page 11: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/11.jpg)
AI-driven system design
Data cleansing and
preparation
Simulation-
generated data
Human insight
Data Preparation
5
![Page 12: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/12.jpg)
AI-driven system design
Model design and
tuning
Hardware
accelerated training
Interoperability
AI Modeling
Data cleansing and
preparation
Simulation-
generated data
Human insight
Data Preparation
5
![Page 13: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/13.jpg)
AI-driven system design
Model design and
tuning
Hardware
accelerated training
Interoperability
AI Modeling
Data cleansing and
preparation
Simulation-
generated data
Human insight
Data Preparation
Enterprise systems
Embedded devices
Edge, cloud,
desktop
Deployment
5
![Page 14: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/14.jpg)
AI-driven system design
Model design and
tuning
Hardware
accelerated training
Interoperability
AI Modeling
Data cleansing and
preparation
Simulation-
generated data
Human insight
Data Preparation
Enterprise systems
Embedded devices
Edge, cloud,
desktop
Deployment
5
![Page 15: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/15.jpg)
Preparing and labelling data
Data cleansing and
preparation
Simulation-generated
data
Human insight
Data Preparation
![Page 16: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/16.jpg)
16
Preparing and labelling data
Q. How to label collected data?Data cleansing and
preparation
Simulation-generated
data
Human insight
Data Preparation
![Page 17: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/17.jpg)
17
Preparing and labelling data
Q. How to label collected data?
Q. What if it is not possible to collect
data?
Data cleansing and
preparation
Simulation-generated
data
Human insight
Data Preparation
![Page 18: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/18.jpg)
Labeling Signals with Signal Labeler App
7
![Page 19: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/19.jpg)
Generate Synthetic Data for various applications in MATLAB
8
![Page 20: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/20.jpg)
Generate Synthetic Data for various applications in MATLAB
8
Simulate data using Simulink models
![Page 21: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/21.jpg)
Generate Synthetic Data for various applications in MATLAB
8
Simulate data using Simulink models Generate wireless waveforms
![Page 22: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/22.jpg)
Generate Synthetic Data for various applications in MATLAB
8
Generate Radar Returns
Simulate data using Simulink models Generate wireless waveforms
![Page 23: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/23.jpg)
Generate Synthetic Data for various applications in MATLAB
8
Generate Radar Returns Generate and Augment Audio Data
Simulate data using Simulink models
text2speech
23
Generate wireless waveforms
![Page 24: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/24.jpg)
9
Generation of wireless communication waveforms with impairments
![Page 25: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/25.jpg)
9
•Modulate digital baseband signals using built-in functions•BPSK, QPSK, 8PSK, FM, DSB-AM, SSB-AM, GFSK,PAM4
Generation of wireless communication waveforms with impairments
![Page 26: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/26.jpg)
9
•Modulate digital baseband signals using built-in functions•BPSK, QPSK, 8PSK, FM, DSB-AM, SSB-AM, GFSK,PAM4
•Easily account for various impairments•RF / Hardware impairments (Frequency/ Phase Offsets etc. )
• Channel Impairments (Multipath Fading Channels)
Generation of wireless communication waveforms with impairments
![Page 27: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/27.jpg)
9
•Modulate digital baseband signals using built-in functions•BPSK, QPSK, 8PSK, FM, DSB-AM, SSB-AM, GFSK,PAM4
•Easily account for various impairments•RF / Hardware impairments (Frequency/ Phase Offsets etc. )
• Channel Impairments (Multipath Fading Channels)
• Generate Datasets for Deep Learning
• 5000 frames generated for each modulation type
• 80% data – Training; 10% data – Validation; 10% data - Test
Generation of wireless communication waveforms with impairments
![Page 28: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/28.jpg)
Feature Extraction
Data cleansing and
preparation
Simulation-generated
data
Human insight
Data Preparation
![Page 29: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/29.jpg)
Feature Extraction
Q. Can I use raw data?Data cleansing and
preparation
Simulation-generated
data
Human insight
Data Preparation
![Page 30: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/30.jpg)
Feature Extraction
Q. Can I use raw data?
Q. How do I extract the right features
for my data?
Data cleansing and
preparation
Simulation-generated
data
Human insight
Data Preparation
![Page 31: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/31.jpg)
Use of raw data for AI models
11
![Page 32: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/32.jpg)
Use of raw data for AI models
11
IQ waveform
I waveform
Q waveform
![Page 33: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/33.jpg)
Use of raw data for AI models
11
High
Dimensionality
Need for more data
Need for
specialized models
IQ waveform
I waveform
Q waveform
Challenges with Raw Data
![Page 34: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/34.jpg)
Feature extraction with signal processing techniques
12
![Page 35: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/35.jpg)
Feature extraction with signal processing techniques
12
![Page 36: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/36.jpg)
Feature extraction with signal processing techniques
12
![Page 37: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/37.jpg)
Feature extraction with signal processing techniques
12
![Page 38: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/38.jpg)
Feature extraction with signal processing techniques
12
![Page 39: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/39.jpg)
Building the AI models
Model design and
tuning
Hardware
accelerated training
Interoperability
AI Modeling
![Page 40: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/40.jpg)
Building the AI models
Q. How do I select the right model for
my application:Model design and
tuning
Hardware
accelerated training
Interoperability
AI Modeling
![Page 41: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/41.jpg)
Building the AI models
Q. How do I select the right model for
my application:• If I do not have enough data?
• If I do not have domain expertise?
• If I need an easily interpretable model?
…..
Model design and
tuning
Hardware
accelerated training
Interoperability
AI Modeling
![Page 42: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/42.jpg)
Start by using published literature and MATLAB examples
14
![Page 43: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/43.jpg)
Start by using published literature and MATLAB examples
14
![Page 44: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/44.jpg)
Start by using published literature and MATLAB examples
14
![Page 45: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/45.jpg)
Understanding tradeoffs for model selection
15
![Page 46: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/46.jpg)
Understanding tradeoffs for model selection
15
Da
ta V
olu
me
Time Required
![Page 47: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/47.jpg)
Understanding tradeoffs for model selection
15
Da
ta V
olu
me
Signal Processing /
Domain Knowledge
![Page 48: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/48.jpg)
There are three ways to build AI models in MATLAB
16
![Page 49: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/49.jpg)
There are three ways to build AI models in MATLAB
16
Writing code
fitcauto/fitrauto
![Page 50: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/50.jpg)
There are three ways to build AI models in MATLAB
16
Interactively Design Models with
Apps
Writing code
fitcauto/fitrauto
![Page 51: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/51.jpg)
There are three ways to build AI models in MATLAB
16
Interactively Design Models with
Apps
Use Transfer Learning
for Deep Learning
Writing code
fitcauto/fitrauto
![Page 52: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/52.jpg)
Iterate to find the best model with Experiment Manager App
17
![Page 53: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/53.jpg)
Iterate to find the best model with Experiment Manager App
17
![Page 54: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/54.jpg)
Iterate to find the best model with Experiment Manager App
17
Find optimal
training options
![Page 55: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/55.jpg)
Iterate to find the best model with Experiment Manager App
17
Find optimal
training options
Compare the
results of using
different data sets
![Page 56: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/56.jpg)
Iterate to find the best model with Experiment Manager App
17
Find optimal
training options
Compare the
results of using
different data sets
Compare the
results of using
different models
![Page 57: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/57.jpg)
Selecting the Right Model : Understanding Tradeoffs
18
Da
ta V
olu
me
Signal Processing /
Domain Knowledge
![Page 58: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/58.jpg)
Selecting the Right Model : Understanding Tradeoffs
18
Da
ta V
olu
me
Signal Processing /
Domain Knowledge
![Page 59: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/59.jpg)
Continuous Wavelet Transform is used to extract the Time-
Frequency maps
19
![Page 60: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/60.jpg)
Continuous Wavelet Transform is used to extract the Time-
Frequency maps
19
•One line of code for generating wavelet time-
frequency visualization in MATLAB. Works for any
signal >> cwt(inputSignal)
![Page 61: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/61.jpg)
Continuous Wavelet Transform is used to extract the Time-
Frequency maps
19
•One line of code for generating wavelet time-
frequency visualization in MATLAB. Works for any
signal >> cwt(inputSignal)
•Localizes sharp transients and slowly varying
oscillations simultaneously
![Page 62: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/62.jpg)
Continuous Wavelet Transform is used to extract the Time-
Frequency maps
19
•One line of code for generating wavelet time-
frequency visualization in MATLAB. Works for any
signal >> cwt(inputSignal)
•Localizes sharp transients and slowly varying
oscillations simultaneously
• Works with complex data
![Page 63: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/63.jpg)
Using time-frequency maps as inputs to a pretrained CNN
20
![Page 64: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/64.jpg)
Using time-frequency maps as inputs to a pretrained CNN
20
![Page 65: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/65.jpg)
Using time-frequency maps as inputs to a pretrained CNN
20
![Page 66: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/66.jpg)
Transfer Learning with Deep Network Designer App
21
![Page 67: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/67.jpg)
Train and Test Deep Network
22
![Page 68: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/68.jpg)
Test Deep Network
23
![Page 69: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/69.jpg)
Testing network with connected hardware
24
![Page 70: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/70.jpg)
Testing network with connected hardware
24
![Page 71: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/71.jpg)
Testing network with connected hardware
24
![Page 72: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/72.jpg)
Testing network with connected hardware
24
![Page 73: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/73.jpg)
AI-assisted system design
Model design and
tuning
Hardware
accelerated training
Interoperability
AI Modeling
Data cleansing and
preparation
Simulation-
generated data
Human insight
Data Preparation
Enterprise systems
Embedded devices
Edge, cloud,
desktop
Deployment
25
![Page 74: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/74.jpg)
AI-assisted system design
Model design and
tuning
Hardware
accelerated training
Interoperability
AI Modeling
Data cleansing and
preparation
Simulation-
generated data
Human insight
Data Preparation
Enterprise systems
Embedded devices
Edge, cloud,
desktop
Deployment
25
![Page 75: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/75.jpg)
Deep Learning can be used in each step of the AI workflow
26
![Page 76: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/76.jpg)
Deep Learning can be used in each step of the AI workflow
26
Labeling assistance
Manually Correct
Neural Network
Auto Label
Inspect
![Page 77: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/77.jpg)
Deep Learning can be used in each step of the AI workflow
26
Labeling assistance
Manually Correct
Neural Network
Auto Label
Inspect
classifySound (YAMNet),GoogLeNet,
fitcecoc(ResNet18)
![Page 78: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/78.jpg)
Deep Learning can be used in each step of the AI workflow
26
Labeling assistance Synthetic Data Generation
Manually Correct
Neural Network
Auto Label
Inspect
classifySound (YAMNet),GoogLeNet,
fitcecoc(ResNet18)
![Page 79: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/79.jpg)
Deep Learning can be used in each step of the AI workflow
26
Labeling assistance Synthetic Data Generation
Manually Correct
Neural Network
Auto Label
Inspect
classifySound (YAMNet),GoogLeNet,
fitcecoc(ResNet18)
Generative Adversarial Networks
(GANs)
![Page 80: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/80.jpg)
Deep Learning can be used in each step of the AI workflow
27
![Page 81: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/81.jpg)
Deep Learning can be used in each step of the AI workflow
27
Feature Extraction
![Page 82: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/82.jpg)
Deep Learning can be used in each step of the AI workflow
27
Feature Extraction
![Page 83: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/83.jpg)
Deep Learning can be used in each step of the AI workflow
27
Feature Extraction
vggFeatures, waveletScattering
![Page 84: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/84.jpg)
Deep Learning can be used in each step of the AI workflow
27
Feature Extraction
Differentiable Signal Processing
vggFeatures, waveletScattering
![Page 85: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/85.jpg)
Deep Learning can be used in each step of the AI workflow
27
Feature Extraction
Differentiable Signal Processing
vggFeatures, waveletScattering
![Page 86: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/86.jpg)
Deep Learning can be used in each step of the AI workflow
27
Feature Extraction
Differentiable Signal Processing
vggFeatures, waveletScattering
dlstft (Differentiable STFT)
![Page 87: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/87.jpg)
AI-driven system design
Model design and
tuning
Hardware
accelerated training
Interoperability
AI Modeling
Data cleansing and
preparation
Simulation-
generated data
Human insight
Data Preparation
Enterprise systems
Embedded devices
Edge, cloud,
desktop
Deployment
28
![Page 88: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/88.jpg)
Deploy to any processor with best-in-class performance
29
![Page 89: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/89.jpg)
Deploy to any processor with best-in-class performance
Preprocessing, Feature
Extraction, AI Model
29
![Page 90: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/90.jpg)
Deploy to any processor with best-in-class performance
CPU
Code
Generation
Preprocessing, Feature
Extraction, AI Model
29
![Page 91: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/91.jpg)
Deploy to any processor with best-in-class performance
CPU
GPUCode
Generation
Preprocessing, Feature
Extraction, AI Model
29
![Page 92: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/92.jpg)
Deploy to any processor with best-in-class performance
FPGA
CPU
GPUCode
Generation
Preprocessing, Feature
Extraction, AI Model
29
![Page 93: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/93.jpg)
Deploying complete AI algorithms to embedded processors, GPUs
and FPGAs
30
![Page 94: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/94.jpg)
Deploying complete AI algorithms to embedded processors, GPUs
and FPGAs
30
![Page 95: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/95.jpg)
Deploying complete AI algorithms to embedded processors, GPUs
and FPGAs
30
![Page 96: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/96.jpg)
Deploying complete AI algorithms to embedded processors, GPUs
and FPGAs
30
![Page 97: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/97.jpg)
Deploying complete AI algorithms to embedded processors, GPUs
and FPGAs
31
![Page 98: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/98.jpg)
MATLAB supports the entire AI-driven system design
32
![Page 99: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/99.jpg)
MATLAB supports the entire AI-driven system design
Signal Processing apps
Feature Extraction Techniques
Generate Data Quickly build models
Accelerate training
Deploy to targets with
code generation
32
![Page 100: Building AI Applications for Signals and Time Series Data](https://reader031.fdocuments.us/reader031/viewer/2022022118/62127497ba4ded22e875b90c/html5/thumbnails/100.jpg)
mathworks.com
© 2021 The MathWorks, Inc. MATLAB and Simulink are registered
trademarks of The MathWorks, Inc. See
www.mathworks.com/trademarks for a list of additional
trademarks. Other product or brand names may be trademarks or
registered trademarks of their respective holders. 33