Address Book


Texas Instruments


Microchip Technology Inc.

ON Semiconductor

Molex, LLC


Murata Manufacturing Co.,



ICP Deutschland GmbH


Yamaichi Electronics USA Inc.

EM Microelectronic

Murata Power Solutions Inc
Industry automation
Test & measurement
Solar energy
Training , Trade fairs, Evens
Interesting video
TDK announces world’s first MIPI standar
Portable SSD T7 Touch
Samsung Releases Portable SSD T7 Touch –
120W eighth brick DC-DC converters for r
AR236.B & AR232.B
APAR AR236.B & AR232.B radio loggers
JZ-500 and JZ-500-C black
JZ-500 and JZ-500-C black cables manufac
HF41F subminiature power relays from HON
GOLDTOOL toolsets in cases
Intel® Neural Compute Stick 2
RS Components announces availability of
MC4 solar connector sets
Multifunctional WIHA LED Torch

Artificial Intelligence on the Edge of Your Network

Artificial Intelligence on the Edge of Your Network

This is not the topic of the future; with AAEON hardware you can handle the deployment of artificial intelligence now. Read the article to find out how it all works and what you need for that.

Why now?

In the present state of technology we can train deep neural network (DNN) for a specific tasks like object and human face detection and recognition, speech recognition, language translation, games (chess, go, etc.), autonomous vehicle driving, sensor status monitoring and predictive machine maintenance decisions, evaluation of X-ray images in healthcare etc. For such specialized tasks, a DNN can reach or even exceeds human capabilities.

Why Use Artificial Intelligence on the Edge

For example, the modern building contains a lot of sensors, HVAC devices, elevators, security cameras, etc. connected to the internal network. For the security, latency or robustness reasons, it is more appropriate to run artificial intelligence tasks locally, on the edge of a local network, and send only anonymized data that are necessary for taking global decisions to the cloud.

Hardware on the Edge

To deploy DNN on the edge we need a device with enough computing power and simultaneously low power consumption. The current state of technology offers a combination of low power CPU and VPU accelerator (x86 CPU SBC+ Intel Myriad X VPU) or CPU + GPU accelerator (ARM CPU + Nvidia GPU).

The easiest way to start DNN is using UP Squared AI Vision X Developer Kit version B. It is based on UP Square SBC with Intel Atom®X7-E3950 processor with 8GB RAM, 64 GB eMMC, AI Core X module with Myriad X MA2485 VPU and USB camera with resolution 1920 x 1080 and manual focus. The kit is preinstalled with Ubuntu 16.04 (kernel 4.15) and OpenVINO toolkit 2018 R5.

Toolkit contains precompiled demo applications in /home/upsquared/build/intel64/Release and pre-trained models in /opt/intel/computer_vision_sdk/deployment_tools/intel_models. To see help for any demo application, run it in terminal with –h option. It is recommended to initialize OpenVINO environment before running demo application by command source /opt/intel/computer_vision_sdk/bin/

Besides UP Squared AI Vision X Developer Kit, Aaeon also offers:

1. Myriad X MA2485 VPU based modules: AI Core X (mPCIe full-size, 1x Myriad X), AI Core XM 2280 (M.2 2280 B+M key, 2x Myriad X), AI Core XP4/ XP8 (PCIE [x4] card, 4 or 8x Myriad X).
2. BOXER-8000 series based on Nvidia Jetson TX2 module.
3. BOXER-8320AI with Core i3-6100U or Celeron 3955U processor and two AI Core X modules.
4. Boxer-6841M series with the motherboard for Intel 6th / 7th generation of Core-I or Xeon processor for socket LGA1151 and 1x PCIe [x16] or 2x PCIe [x8] slots for GPU with power consumption max. 250W.

Hardware for Learning To train DNN we need high computing power. For example, on the ImageNet competition in 2012, the winning team used the convolutional neural network AlexNet. 1.4 ExaFLOP=1,4e6 TFLOP operations were required for learning. It took 5 to 6 days on two Nvidia GTX 580 GPUs, where each has the 1.5 TFLOP computing performance.

The table below summarizes the theoretical peak performance of the hardware.

For comparison, the high-end Intel Xeon Platinum 8180 processor
? has 28 cores with 2 AVX-512 & FMA units per core
? AVX-512 turbo frequency 2.3GHz if all cores are active
? costs USD 10 000

It offers theoretical peak performance: # of cores * frequency in GHz * AVX-512 DP FLOPS/Hz * # of AVX-512 units * 2 = 2060.8 GFLOPS in double precision (DP) › 4.1216 TFLOPS in single (32-bit).

As you can see from the table above, GPU provides far more performance for neural networks learning. It is necessary to note that the number of operations per second is not the only parameter that affects learning performance. Factors as RAM size, the data transfer rate between CPU and RAM, GPU and GPU RAM or between GPU also influence the learning speed.


The OpenVINO (open visual inference and neural network) toolkit is free software that allows quick deployment of applications and solutions that emulate human vision.

The OpenVINO toolkit:
? Uses CNN (convolution neural network)
? Can split computation between Intel CPU, integrated GPU, Intel FPGA, Intel Movidius Neural Compute Stick and vision accelerators with Intel Movidius Myriad VPUs
? Provides an optimized interface to OpenCV, OpenCL, and OpenVX
? Supports Caffe, TensorFlow, MXNet, ONNX, Kaldi frameworks,

TensorFlow is an open-source library for numerical computation and machine learning. It provides a convenient front-end API for building applications in Python programming language. However, the TensorFlow-generated application itself is converted to optimized C ++ code that can run on a variety of platforms such as CPUs, GPUs, local machine, a cluster in the cloud, embedded devices at the edge and the like.

Other Useful Software
Jupyter Lab / Notebook

How it Works

Simplified Neuron Model
Simple neuron model – the perceptron was first time described by Warren McCulloch and Walter Pitts in 1943 and it is still the standard of reference in the field of neural networks.

Activation function f () adds nonlinearity to perceptron. Without non-linear activation function in the neural network (NN) of perceptrons, no matter how many layers it had, it would behave just like a single-layer perceptron, because summing these layers would give you just another linear function. The most often used activation function is rectified linear unit – ReLU.

y = f(x) = max (0, x), for x < = 0, y = 0, for x ? 0, y=x

Inference (Forward pass)

The figure above shows a deep neural network (DNN) because it contains multiple layers between input and output layers. As you can notice, DNN requires matrix multiplications and additions. Special hardware optimized for this task, such as GPU (graphics processing unit) and VPU (vision processing unit), is much faster than general-purpose CPU (central processing unit, processor) and has lower power consumption.

Learning (Backward pass)

Let’s say we want to teach DNN to recognize orange, banana, apple, and raspberry (object classes) in the image.

1. We prepare a lot of images of fruits above and divide it into the training set and validation set. The training set contains images and correct, required outputs for these images. DNN will have 4 outputs. The first output provides a score (probability) that fruit in the picture is orange, the second provides the same for a banana, etc.

2. We set initial values for all weights w_i and biases b_i. Random values are typically used.

3. We pass the first image through DNN. The network provides scores (probability) on each output. Let’s say that the first image depicts orange. Network outputs can be y= (orange, banana, apple, raspberry) = (0.5, 0.1, 0.3, 0.1). Network “says” that input is orange with a probability of 0.5.

4. We define a loss (error) function that quantifies the agreement between the predicted scores and the correct scores for each class. Function E = 0.5*sum (e_j) ^2, where e_j = y_j - y_real_j and j is number of images in training set is often used. E_1_orange = 0.5*(0.5-1)^2=0.125, E_1_banana =.0.5*(0.1-0)^2 = 0.005 E_1_apple = 0.5*(0.3-0)^2 = 0.045, E_1_raspberry = 0.5*(0.1-0)^2 = 0.005 E_1 = (0.125, 0.005, 0.045, 0.005)

5. We pass all the remaining images from the training set through DNN and calculate E (E_orange, E_banana, E_apple, E_raspberry) loss function value for the whole training set.

6. To modify all weights w_i and biases b_i for the next training pass (epoch), we need to know the influence of each parameter w_i and b_i to loss function for each class. If the increase of parameter value causes the increase of loss function value, we need to decrease this parameter and vice versa. But how to calculate required parameters increase or decrease?

Let’s try a simple example.

We have three points with coordinates (x, y): (1, 3), (2.5, 2), (3.5, 5). We want to find a line y = w.x + b for which loss function E = 0.5*sum (e_j) ^2 , where e_j = y_j – y_real_j, j=1, 2, 3 is minimal. To make task as simple as possible, let’s say that w=1.2 and we need to find optimal value for b only. We pick the initial value for b=0.

Let’s calculate loss function: E = 0.5*sum (e_j) ^2 = 0.5*(e_1^2 + e_2^2 + e_3^2), e_1=1.2*1 + b -3, e_2 = 1.2*2.5 + b – 2, e_3 = 1.2*3.5 + b – 5.

Loss function E is simple, we can find a minimum of E by solving equation: ?E/?b = 0. It is a compound function, to calculate ?E/?b we apply the chain rule.

?E/?b=0.5*((?E/?e_1)*(?e_1/?b) + (?E/?e_2)*(?e_2/?b) + (?E/?e_3)*(?e_3/?b)) = 0.5*(2*e_1*1 + 2*e_2*1 + 2*e_3*1) = (1.2*1 + b – 3) + (1.2*2.5 + b – 2) + (1.2*3.5 + b – 5) = 0 => b = 0.53333

In practice, where number of parameters w_i and b_i can reach a million or more, it is not practical to solve equation ?E/?b_i = 0 and ?E/?b_i = 0 directly, iterative algorithm is used instead of that.

We started with b = 0, next value will be b_1 = b_0 – ?*?E/?b, where ? is learning rate (hyper-parameter) and -?*?E/?b is step size. We stop learning when step size reaches the defined threshold, in practice 0.001 or smaller.

For ? = 0.3, b_1 = 0.48, b_2 = 0.528, b_3 = 0.5328 and b_4 = 0.53328 and b_5 = 0.5533328. After 5 iterations the step size dropped to 4.8e-5 and we stop learning here. The value of b obtained by this algorithm is practically the same as the value obtained by solving the equation ?E/?b=0.

This method is called gradient descend.

Learning rate is an important hyper-parameter. If it is too small, it takes a lot of steps to find a minimum loss function; if it is high, the algorithm can fail. In practice, improved variants of an algorithm such as  Adam are used.

7. We repeat steps 5 and 6 until loss function value decreases to the required value.

8. We pass the validation set through trained DNN and evaluate accuracy.

In the present time, DNN learning is a highly experimental work. Many DNN architectures are known, each of them is well suited for a particular range of tasks. Every DNN architecture has its own set of hyper-parameters that influence the behavior of DNN. Arm yourself with patience and the result will come soon.

For further information about AAEON products, please do not hesitate to contact us at

2019102102 / 24.10.2019 / Various / SOS electronic s.r.o. /

How does the EMC Filter work?
How does the EMC Filter work?
Swiss company Schurter offers EMC filters, common mode chokes, as well as filters integrated with inlet, switch and fuse. Typical EMC filter contains common mode choke and X and Y class capacitors. But how does it work?
Zhaga. Smart Lighting for Smart Cities
Thanks to sensors and communication modules, modern LED luminaires will also be able to provide smart services. A key part of these luminaires are connectors, such as the LUMAWISE Endurance S series. They can be found in our product range under the TE Connectivity brand, a member of the Zhaga Consortium - the organization for standardization of LED lighting components for public lighting.
Zhaga. Smart Lighting for Smart Cities
The world’s first GSM/NB2 dual-mode module 
The world’s first GSM/NB2 dual-mode module
Quectel Wireless Solutions, the leading global supplier of Internet of Things (IoT) Modules, launched its first LTE Cat NB2/GSM dual-mode module BC92, which supports switchover between GSM and NB-IoT networks and ensures utilizing network coverage of both technologies.
Get a lot of connectivity for the buck
WizFi360 is fast, affordable and simple plug&play WiFi module WIZnet designed for various IoT applications, mobile and wearable devices.
Get a lot of connectivity for the buck
Infrared cameras seeing 4x more details than their predecessors
Infrared cameras seeing 4x more details than their predecessors
The new TG267 and TG297 infrared cameras significantly accelerate the thermal diagnostics process. The features like 160x120px resolution, MSX combined image, new laser, larger display or easy data transfer to mobile devices are now available for the better price than you would ever expect.
FIS5084 Easily Detects HFC Leakage
A brief guide on how the FIS module for refrigerant leads detection works, a summary of its features, and technical details you cannot find in a publicly available datasheet.
FIS5084 Easily Detects HFC Leakage
Robust enclosures with elegant design
Robust enclosures with elegant design
The new Hammond 1557 Series offers a surprisingly attractive industrial waterproof enclosure solution with three mounting options, for different control panels or monitoring units.
3 reasons for the TRI Series DC/DC converters
Reinforced insulation system, high common-mode transient immunity, and compact DIP-24 package. With the latest, IT security certificates, the TRACO POWER TRI Series is the perfect choice for demanding applications in industrial, transportation and instrumentation sectors.
3 reasons for the TRI Series DC/DC converters
Smart dual Nano-SIM card socket
Smart dual Nano-SIM card socket
Two SIM cards give us the choice of an operator with better coverage as well as the possibility to use 4G and LPWA technologies. Attend series 115S holder for 2 nano-SIM cards gives several proven benefits in comparison with the competition.
Kontakt Trio - set for perfect contact treatment
A strong trio of Kontakt Chemie sprays for quality and reliable contacts, in a discounted package.
Kontakt Trio - set for perfect contact treatment
Get the best results with MEC switches
Get the best results with MEC switches
Whether you choose the current MEC button generation 5E/5G or the previous one, 3A/3C/3E/3F, it’s always gonna be a great decision. MEC is a guarantee of reliability, so both generations have their place on the market.
Quartz Crystals and Quartz Crystal Oscillators Ansen
Without crystal oscillators, there would be no wireless communication, GPS receivers or Ethernet. Therefore, we are expanding the portfolio of suppliers with the renowned ANSEN manufacturer. Although new technologies such as MEMS oscillators exist, crystal oscillators remain the dominant source of accurate and stable frequency.
Quartz Crystals and Quartz Crystal Oscillators Ansen
6x6 new enclosures for your devices
6x6 new enclosures for your devices
If you haven't found your "size" between Hammond Series 1554 or 1555 industrial enclosures, you'll very likely find it now. Hammond offers 6 new sizes and 36 different combinations of popular plastic enclosures.
EA PlugS - small OLED display with added value
Graphics feature support, USB interface, capacitive touch panel, aesthetic glass frame and easy to install - all the pluses for smooth and fast application development.
EA PlugS - small OLED display with added value
MSM LA Series – Vandal resistant latching switches
MSM LA Series – Vandal resistant latching switches
The range of vandal resistant momentary push-button switches is relatively wide but there are fewer latching switches. We extended our stock range with basic models of vandal resistant latching switches from the MSM LA 19 Series from SCHURTER.
Apacer Brings New Technologies to SSD
New Apacer’s SSDs improve already high data integrity, longevity, security, and survivability. Let’s have a look at how they did it.
Apacer Brings New Technologies to SSD
Get more space on PCB with TEL 10WI
Get more space on PCB with TEL 10WI
TRACO POWER 10 Watt DC/DC converter for space-limited applications.
Artificial Intelligence on the Edge of Your Network
This is not the topic of the future; with AAEON hardware you can handle the deployment of artificial intelligence now. Read the article to find out how it all works and what you need for that.
Artificial Intelligence on the Edge of Your Network
Company of the week

Interesting video

How Intel Makes Chips: Concept to Customer

How Do You Minimize Bubbles in a Two Part Epoxy System?

Adafruit Partners with Panasonic

Computer-On-Modules are world-leading platforms for embedded system designs

Ghostbuster's Theme on eight floppy drives
Address Book


Texas Instruments


Microchip Technology Inc.

ON Semiconductor

Molex, LLC


Murata Manufacturing Co.,



ICP Deutschland GmbH


Yamaichi Electronics USA Inc.

EM Microelectronic

Murata Power Solutions Inc

Keysight Technologies

XP Power Limited

Fischer Connectors SA


TE Connectivity

akYtec GmbH

SOS electronic s.r.o.


Taiwan Semiconductor

ROHM Semiconductor

Pickering Interfaces Ltd


Mouser Electronics

Foremost Electronics

RS Components GmbH
embedded world 2020, Nuremberg, 25.2.-27.2.2020
BATTERY JAPAN, Tokyo, 26.2.-28.2.2020
Light + Building 2020, Frankfurt am Main, 8.3.-13.3.2020
AMPER 2020, Brno, CZ, 17.3.-20.3.2020
intersec Saudi Arabia, Riyadh, 23.3.-25.3.2020
ExpoElectronica, Moscow, 14.4.-16.4.2020
SECUTECH 2020, Taipei, 22.4.-24.4.2020
SMTconnect 2020, Nuremberg, 5.5.-7.5.2020
PCIM Europe, Nuremberg, 5.5.-7.5.2020
LED Expo Mumbai 2020, 7.5.-9.5.2020
SPS Italia, Fiere di Parma, 26.5.-28.5.2020
CWIEME Berlin, 26.5.-28.5.2020
Wire & Cable Guangzhou, China, 27.5.-29.5.2020
Automechanika Dubai, 7.6.-9.7.2020
hardware+tools, Dubai, UAE, 7.6.-9.6.2020
ELEKTRO 2020, MOSCOW, Russia, 8.6.-11.6.2020
PCIM Asia, Shanghai, 1.7.-3.7.2020
SPS Automation Middle East, Dubai, 20.8.-21.8.2020
Intersec Buenos Aires, 26.8.-28.8.2020
Automechanika Frankfurt, DE, 8.9.-12.9.2020
BATTERY JAPAN, Osaka, 9.9.-11.9.2020
IMTS2020, 14.9.-19.9.2020, Chicago, IL
Interlight Russia, Moscow, 14.9.-17.9.2020
Prolight + Sound NAMM, Moscow, 17.9.-19.9.2020
NEPCON Nagoya, Japan, 21.10.-23.10.2020
Prolight + Sound Shanghai, 28.10.-30.10.2020
SPS 2020, Nuremberg, 24.11.-26.11.2020
NEPCON Japan, Tokio, 20.1.-22.1.2021
ISE 2021, Barcelona, Spain, 2.2.-5.2.2021
DistribuTECH, 9.2.-11.2.2021, San Diego, CA
Interesting video
The ISS Design Challenge ...
Interesting video
Mouser Electronics Warehouse Tour with Grant Imahara
naše portály dle jazyka:

česko/slovenská jazyková verze:

anglická jazyková verze:

polská jazyková verze:

ruská jazyková verze:
naše portály dle zaměření:




MALUTKI media s.r.o.
Těrlická 475/22
735 35 Horní Suchá
tel. 00420-603531605

All trademarks are the property of their respective owners.
ISSN 1801-3813