A Scalable And Low Power Dcnn For Multimodal Data Classification

• doi: 10.1109/RECONFIG.2018.8641702
• title: A Scalable and Low Power DCNN for Multimodal Data Classification
• publisher: IEEE
• isbn: 978-1-7281-1969-4
• issn: 2325-6532
• rank: 912
• access_type: LOCKED
• content_type: Conferences
• abstract: This paper presents SensorNet which is a scalable and low power embedded Deep Convolutional Neural Network (DCNN), designed to classify multimodal time series signals. Time series signals generated by different sensor modalities with different sampling rates are first converted to images (2-D signals) and then DCNN is utilized to automatically learn shared features in the images and perform the classification. SensorNet is scalable with respect to different types of multi-channel time series data, and does not require expert knowledge for extracting features for each sensor data. Additionally, it can achieve very high detection accuracy for different case studies, and has a very efficient architecture which makes it suitable to be employed at IoT and wearable devices. A custom low power hardware architecture is also designed for the efficient deployment of SensorNet at embedded real-time systems. SensorNet performance is evaluated using three different case studies including Physical Activity Monitoring, stand-alone Tongue Drive System (sdTDS) and Stress Detection and it achieves an average detection accuracy of 98%, 96.2% and 94% for each case study, respectively. We implement SensorNet using our custom hardware architecture on Xilinx FPGA (Artix-7) which on average consumes 0.3 mJ energy per classification while meeting all applications time requirements. To further reduce the power consumption, SensorNet is implemented using ASIC at the post layout level in 65-nm CMOS technology which consumes approximately 8$\times$ lower power compared to the FPGA.
• article_number: 8641702
• pdf_url: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8641702
• html_url: https://ieeexplore.ieee.org/document/8641702/
• abstract_url: https://ieeexplore.ieee.org/document/8641702/
• publication_title: 2018 International Conference on ReConFigurable Computing and FPGAs (ReConFig)
• conference_location: Cancun, Mexico
• conference_dates: 3-5 Dec. 2018
• publication_number: 8637084
• is_number: 8641689
• publication_year: 2018
• publication_date: 3-5 Dec. 2018
• start_page: 1
• end_page: 6
• citing_paper_count: 5
• citing_patent_count: 0
• download_count: 266
• insert_date: 20190214
• index_terms:
  • ieee_terms:
    • Convolution
    • Time series analysis
    • Monitoring
    • Feature extraction
    • Tongue
    • Biomedical monitoring
    • Stress
  • dynamic_index_terms:
    • Scalable
    • Low Power
    • Deep Convolutional Neural Network
    • Multimodal Data Classification
    • Neural Network
    • Time Series
    • Time-series
    • Deep Neural Network
    • Detection Accuracy
    • Time Series Data
    • Time-series Data
    • Power Consumption
    • Average Accuracy
    • Internet Of Things
    • Internet-of-Things
    • Wearable Devices
    • Wearable Technology
    • Internet Of Things Devices
    • Internet-of-Things Devices
    • IoT Devices
    • High Detection Accuracy
    • Hardware Architecture
    • Physical Activity Monitoring
    • Stress Detection
    • Convolutional Layers
    • Convolutional Block
    • Number Of Filters
    • Channel Data
    • Dynamic Time Warping
    • Fully-connected Layer
    • Fully Connected Layer
    • Feature Maps
    • Inertial Measurement Unit
    • Inertial Sensors
    • Softmax Layer
• isbn_formats:
  • format: Print on Demand(PoD) ISBN, value: 978-1-7281-1969-4, isbnType: New-2005
  • format: Electronic ISBN, value: 978-1-7281-1968-7, isbnType: New-2005
• authors:
  • Author Name: Ali Jafari
    Affiliation: Navigation R&D Division, Korea Aerospace Research Institute, Daejeon, South Korea
    Author URL: https://ieeexplore.ieee.org/author/37085622965
    ID: 37085622965
    Order: 1
    Author Affiliations:
    • Navigation R&D Division, Korea Aerospace Research Institute, Daejeon, South Korea
  • Author Name: Morteza Hosseini
    Affiliation: School of Electronic and Electrical Engineering, Hongik University, Seoul, South Korea
    Author URL: https://ieeexplore.ieee.org/author/37086110143
    ID: 37086110143
    Order: 2
    Author Affiliations:
    • School of Electronic and Electrical Engineering, Hongik University, Seoul, South Korea
  • Author Name: Houman Homayoun
    Affiliation: Navigation R&D Division, Korea Aerospace Research Institute, Daejeon, South Korea
    Author URL: https://ieeexplore.ieee.org/author/37398927100
    ID: 37398927100
    Order: 3
    Author Affiliations:
    • Navigation R&D Division, Korea Aerospace Research Institute, Daejeon, South Korea
  • Author Name: Tinoosh Mohsenin
    Affiliation: Department of Electronic Engineering, Hanyang University, Seoul, South Korea
    Author URL: https://ieeexplore.ieee.org/author/37392295000
    ID: 37392295000
    Order: 4
    Author Affiliations:
    • Department of Electronic Engineering, Hanyang University, Seoul, South Korea

Image Sensor

• sensor_type: CMOS
• resolution: Not Specified
• dynamic_range: Not Specified
• pixel_size: Not Specified
• dark_current: Not Specified

Optical Data

• focal_length: Not Specified
• aperture: Not Specified
• field_of_view: Not Specified
• distortion: Not Specified

Performance Metrics

• frame_rate: Not Specified
• signal_to_noise_ratio: Not Specified
• sensitivity: Not Specified
• shutter_speed: Not Specified
• power_consumption: Not Specified
• noise: Not Specified

Applications & Benefits

• cell_imaging: Not Specified
• benefits: Not Specified

Supporting Organizations

• supported_by: Not Specified

Manuscript Details

• publication_date: 3-5 Dec. 2018

Relevancy Score

• score: 2
• missing_fields:
  • sensor_type
  • resolution
  • dynamic_range
  • pixel_size
  • dark_current
  • focal_length
  • aperture
  • field_of_view
  • distortion
  • frame_rate
  • signal_to_noise_ratio
  • sensitivity
  • shutter_speed
  • power_consumption
  • noise
  • cell_imaging
  • benefits
  • supported_by
  • publication_date

Processed JSON Filename

request_31b0965c-44f5-40c0-9c48-4e840f98bdb2-a_scalable_and_low_power_dcnn_for_multimodal_data_classification.json