scholarly journals Design of Efficient Floating-Point Convolution Module for Embedded System

Electronics ◽  
2021 ◽  
Vol 10 (4) ◽  
pp. 467
Author(s):  
Jiao Li ◽  
Xinjing Zhou ◽  
Binbin Wang ◽  
Huaming Shen ◽  
Feng Ran
Keyword(s):  
Embedded System ◽  
Critical Path ◽  
Floating Point ◽  
Great Success ◽  
Embedded Devices ◽  
Computing Systems ◽  
Limited Budget ◽  
Conversion Unit ◽  
Operation Unit ◽  
The Mean

The convolutional neural network (CNN) has made great success in many fields, and is gradually being applied in edge-computing systems. Taking the limited budget of the resources in the systems into consideration, the implementation of CNNs on embedded devices is preferred. However, accompanying the increasingly complex CNNs is the huge cost of memory, which constrains its implementation on embedded devices. In this paper, we propose an efficient, pipelined convolution module based on a Brain Floating-Point (BF16) to solve this problem, which is composed of a quantization unit, a serial-to-matrix conversion unit, and a convolution operation unit. The mean error of the convolution module based on BF16 is only 0.1538%, which hardly affects the CNN inference. Additionally, when synthesized at 400 MHz, the area of the BF16 convolution module is 21.23% and 18.54% smaller than that of the INT16 and FP16 convolution modules, respectively. Furthermore, our module using the TSMC 90 nm library can run at 1 GHz by optimizing the critical path. Finally, our module was implemented on the Xilinx PYNQ-Z2 board to evaluate the performance. The experimental results show that at the frequency of 100 MHz, our module is, separately, 783.94 times and 579.35 times faster than the Cortex-M4 with FPU and Hummingbird E203, while maintaining an extremely low error rate.

scholarly journals Design and Implementation ofReal-time Electrical Stimulation Artifact Suppression based on STM32

2021 ◽  
Vol 12 (5) ◽  
pp. 424-427
Author(s):  
Sangsoo Park, Hojun Yeom
Keyword(s):  
Central Nervous System ◽  
Nervous System ◽  
Electrical Stimulation ◽  
Embedded System ◽  
Muscle Contraction ◽  
Real Time ◽  
High Performance ◽  
Floating Point ◽  
Time Signal ◽  
Central Nervous System Damage

A biosignal is used as a control signal for electrical stimulation to restore weakened muscle function due to damage to the central nervous system. In patients with central nervous system damage, sufficient muscle contraction does not occur spontaneously. In this case, applying electrical stimulation can cause normal muscle contraction. However, it is necessary to remove the electrical stimulation artifact caused by the electrical stimulation. This paper describes a system design that removes electrical stimulation artifact in real time using a Cortex-M4-based STM32F processor. The STM32F is a very advantageous MCU for such DSPs, especially because it has a built-in floating point operator. Using STM32F's various high-performance peripherals (12-bit parallel ADC and 12-bit DAC, UART, Timer), an optimized embedded system was implemented.In this paper, the simulated and real-time results were compared and evaluated with the designed fir filter. In addition, the performance of the filter was evaluated through frequency analysis. As a result, it was verified that a high-performance 32-bit STM32F with floating point calculator and various peripherals is suitable for real-time signal processing

FPGA BASED OPC UA EMBEDDED INDUSTRIAL DATA SERVER IMPLEMENTATION

2013 ◽  
Vol 22 (08) ◽  
pp. 1350070 ◽  
Author(s):  
RAFAL CUPEK ◽  
ADAM ZIEBINSKI ◽  
MACIEJ FRANEK
Keyword(s):  
Embedded System ◽  
Process Data ◽  
Embedded Devices ◽  
Communication Interface ◽  
Industrial Data ◽  
Manufacturing Execution Systems ◽  
Control Devices ◽  
Increasing Demand ◽  
Underlying Processes ◽  
Server Architecture

Contemporary computer systems used in the industry are characterized by both an increase in the scale of supported industrial processes measured by the number of control devices and the increasing demand for information describing the underlying processes measured by the number of tags used in Supervisory Control and Data Acquisition (SCADA) systems and Manufacturing Execution Systems (MES). Classical industrial data servers based on the PC architecture are unreliable, expensive to operate and difficult to manage. The alternative is a new standard OPC UA communication interface that simplifies the communication protocol and increases the flexibility of the process data description which allow for the direct implementation of OPC UA servers in embedded devices. This paper presents an innovative approaches in the field of industrial data servers which are used for communication between control systems and SCADA or MES systems. The prototype industrial data server architecture has been implemented, run and tested on the platform of embedded system based on FPGA matrix with built-in FPGA Microblaze processor. The presented experimental results allow to evaluate the applicability of the proposed solution, the limits of the presented architecture and may be used for the improvement of the embedded industrial data servers' structure in subsequent implementations.

An Integrated Architectural Clock Implemented Memory Design for Embedded System

2015 ◽  
Vol 4 (2) ◽  
pp. 129
Author(s):  
Ravi Khatwal ◽  
Manoj Kumar Jain
Keyword(s):  
Embedded System ◽  
Low Power ◽  
Propagation Delay ◽  
Internal Clock ◽  
Access Time ◽  
Architecture Design ◽  
Embedded Devices ◽  
Memory Design ◽  
Simulation Performance ◽  
Cache Simulation

Recently Low power custom memory design is the major issue for embedded designer. Micro wind and Xilinx simulator performs efficient cache simulation and high performances with low power consumption. SRAM efficiency analyzed with 6-T architecture design and analyzed the simulation performance for specific application. We have implemented clock based memory architecture design and analyzed internal clock efficiency for SRAM. Architectural clock implemented memory design that reduces access time and propagation delay time for embedded devices. Internal semiconductor material improvement increases simulation performance and these design implemented for application specific design architecture.

scholarly journals Self-Learning Pipeline for Low-Energy Resource-Constrained Devices

Energies ◽  
2021 ◽  
Vol 14 (20) ◽  
pp. 6636
Author(s):  
Fouad Sakr ◽  
Riccardo Berta ◽  
Joseph Doyle ◽  
Alessandro De De Gloria ◽  
Francesco Bellotti
Keyword(s):  
Embedded System ◽  
Nearest Neighbor ◽  
Energy Resource ◽  
K Nearest Neighbor ◽  
Resource Constrained ◽  
Confidence Measure ◽  
Embedded Devices ◽  
Remote Training ◽  
Iot Devices ◽  
Self Learning

The trend of bringing machine learning (ML) to the Internet of Things (IoT) field devices is becoming ever more relevant, also reducing the overall energy need of the applications. ML models are usually trained in the cloud and then deployed on edge devices. Most IoT devices generate large amounts of unlabeled data, which are expensive and challenging to annotate. This paper introduces the self-learning autonomous edge learning and inferencing pipeline (AEP), deployable in a resource-constrained embedded system, which can be used for unsupervised local training and classification. AEP uses two complementary approaches: pseudo-label generation with a confidence measure using k-means clustering and periodic training of one of the supported classifiers, namely decision tree (DT) and k-nearest neighbor (k-NN), exploiting the pseudo-labels. We tested the proposed system on two IoT datasets. The AEP, running on the STM NUCLEO-H743ZI2 microcontroller, achieves comparable accuracy levels as same-type models trained on actual labels. The paper makes an in-depth performance analysis of the system, particularly addressing the limited memory footprint of embedded devices and the need to support remote training robustness.

scholarly journals Software driven approach for Embedded Devices

2020 ◽  
pp. 01-08
Author(s):  
Hrvoje Dodig ◽  
Joško Šoda ◽  
Ivana Golub
Keyword(s):  
Embedded System ◽  
Working Hours ◽  
Software Components ◽  
New Paradigm ◽  
Embedded Devices ◽  
Electronic Circuitry ◽  
Communication Device ◽  
Embedded Design ◽  
Design Paradigm ◽  
The Embedded System

This paper presents the possible new design paradigm that emerged during the author’s design of an embedded communication device for Croatian Navy. Prior to codesign techniques that emerged in 1990's the traditional embedded design methodology involved problem specification, separate hardware and software specification, integration, and the system test as the final step in the embedded device design. Such an approach can potentially lead to numerous iterations and can increase the cost of the development cycle because there are no guarantees that separately developed software will work well with separately designed hardware. Codesign techniques, on the other hand, delay the decision to which components of hardware or software will be used for embedded system until late stages of embedded design process. At the time of the invention of the codesign techniques this seemed as perfectly balanced approach between design of hardware and software spending about equal time in the design of both hardware and software components. However, since the 1990’s the design of embedded devices has changed; nowadays the most working hours are spent in the design of software while the design of hardware requires less working hours due to extensive choice of IC’s and supporting electronic circuits, and due to advancement of EDA software tools. In favor of the software-driven approach presented in this paper, it should be noted that nowadays, there is a large number of freely-available software components and libraries which, if properly utilized, greatly expedite the development of the software part of the embedded system design. Therefore, perhaps it is a suitable time for a new paradigm shift where the design of the hardware is completely dictated by the design of software, and the design of the hardware is simply the matter of selecting proper IC’s and other electronic circuitry that supports the software. In this paper, we present an example of the embedded design using this software-driven design strategy. By the end of this paper, it is shown that software-driven design not only allows the rapid prototyping of embedded devices, but it reduces the possibility of design errors, as well. Keywords: Embedded design; Hardware-software codesign; Software driven design; ARM technology

Wearable Monitoring and Predicting System for Knee Joint Fatigue Based on Curvature and Pressure Sensing (Preprint)

2020 ◽  
Author(s):  
Jiawei Xin ◽  
Jialun Chen ◽  
Xuanyu Huang ◽  
Xiaodong Pan ◽  
Tengyue Zou
Keyword(s):  
Knee Joint ◽  
Embedded System ◽  
Knee Injuries ◽  
Human Perception ◽  
Table Tennis ◽  
Embedded Devices ◽  
Excessive Exercise ◽  
Fitting Method ◽  
Curve Fitting Method ◽  
The One

BACKGROUND Knee injury is always a trouble for people in daily life. It not only threatens the career of an athlete but also affects a normal engineer through morning running. The injury of the knee joint is found to be directly related to the fatigue caused by excessive exercise. OBJECTIVE The aim of the study was to explore the use of wearable embedded devices to monitor and predict the fatigue degree of the knee joint during exercise, so as to prevent the knee joint from being injured. METHODS An economical embedded system with a designed acceleration-weighted curve fitting method was developed to estimate and predict the knee fatigue state. Then the warning message and recommended lasting time were sent to users to avoid excessive exercise. 24 healthy volunteers were involved in the experiments to verify the effectiveness of the system compared to human perception. RESULTS Only using human perception to prevent knee joint fatigue had a risk of failure while the designed wearable system could protect knee successfully. It was also found that the knee of female was more likely to be injured than the one of male in intense exercises and a high BMI value could influence the risk of knee injuries during sports. However, a short break in sports could significantly extend the healthy time for knee. CONCLUSIONS Early warning from the specially designed embedded system can successfully help people avoid knee joint fatigue and injuries during exercises, such as running, badminton, table tennis and basketball.

scholarly journals Deployment of IoT Edge and Fog Computing Technologies to Develop Smart Building Services

2018 ◽  
Vol 10 (11) ◽  
pp. 3832 ◽  
Author(s):  
Francisco-Javier Ferrández-Pastor ◽  
Higinio Mora ◽  
Antonio Jimeno-Morenilla ◽  
Bruno Volckaert
Keyword(s):  
Fog Computing ◽  
Edge Computing ◽  
Climate Control ◽  
Embedded Devices ◽  
Computing Systems ◽  
Smart Building ◽  
Computing Paradigm ◽  
Novel Approach ◽  
Edge And Fog Computing ◽  
Building Designs

Advances in embedded systems, based on System-on-a-Chip (SoC) architectures, have enabled the development of many commercial devices that are powerful enough to run operating systems and complex algorithms. These devices integrate a set of different sensors with connectivity, computing capacities and cost reduction. In this context, the Internet of Things (IoT) potential increases and introduces other development possibilities: “Things” can now increase computation near the source of the data; consequently, different IoT services can be deployed on local systems. This paradigm is known as “edge computing” and it integrates IoT technologies and cloud computing systems. Edge computing reduces the communications’ bandwidth needed between sensors and the central data centre. Management of sensors, actuators, embedded devices and other resources that may not be continuously connected to a network (such as smartphones) are required for this method. This trend is very attractive for smart building designs, where different subsystems (energy, climate control, security, comfort, user services, maintenance, and operating costs) must be integrated to develop intelligent facilities. In this work, a method to design smart services based on the edge computing paradigm is analysed and proposed. This novel approach overcomes some drawbacks of existing designs related to interoperability and scalability of services. An experimental architecture based on embedded devices is described. Energy management, security system, climate control and information services are the subsystems on which new smart facilities are implemented.

Study on Embedded TCP_IP Protocol and ARM Implementation

2014 ◽  
Vol 556-562 ◽  
pp. 6046-6049 ◽  
Author(s):  
Jin Hai Zhang
Keyword(s):  
Embedded System ◽  
Rapid Development ◽  
Embedded Devices ◽  
Information Product ◽  
Embedded Internet ◽  
Control Functions ◽  
Development Technology ◽  
And Performance ◽  
Application Requirements ◽  
Internet Networks

Rapid development of Internet networks and backbones technology makes communication based on packet-switching technology, quality and performance have been steadily improved. Existing embedded Internet scenarios, there are high costs, insufficient to achieve complex control functions such as. Embedded devices in the application of limited hardware resources, depending on application requirements simplified TCP/IP protocol. Research of embedded system based on TCP/IP to embedded information product development technology will undoubtedly have an important value.

Sign in / Sign up

Export Citation Format

Share Document