Hey everyone, check out my experiences at the Hong Kong Electronics Fair:
Hey everyone,
check out my experiences at the Hong Kong Electronics Fair:
check out my experiences at the Hong Kong Electronics Fair:
K
TAre you back in Belgium?
THey @Event Pings and @Middleware & OS get ready for today’s Engineering Hour on message queues and intertask communication in FreeRTOS! It starts in 30 minutes here: https://app.livestorm.co/devheads-community/devheads-weekly-office-hours-1
What questions do you have for Umesh?
What questions do you have for Umesh?
DevHeads IoT Integration Server
PLEASE JOIN AS A SPEAKER IF YOU WANT TO PARTICIPATE IN THE DISCUSSION!Join DevHeads meetups to connect with other DevHeads and riff on a variety of topics for 30-60 minutes. Join here: https:...
FHey guys,it's a pleasure to join your community.I'm embedded system engineer ,want to learn and share experience.
CWelcome @Fatogoma !
F@Camila_99$$ Thank you.
NThat's a quite widespread(or lets say broadband ;)) question - interesting play you're into.
As I understand you want to make a 1 bit SDR having PDM signal generated again from a 1-bit ADC. Now you want to mix it with a LO which was fine, but also want to take PDM signal as a mixing reference?
If wanting for a SDR, how about converting your PDM with a low-pass filter or a CIC filter in digital logic implementation.
As I understand you want to make a 1 bit SDR having PDM signal generated again from a 1-bit ADC. Now you want to mix it with a LO which was fine, but also want to take PDM signal as a mixing reference?
If wanting for a SDR, how about converting your PDM with a low-pass filter or a CIC filter in digital logic implementation.
KSo why this is interesting is it implies I could mix digitally before conversion - I already have a pipeline established that can do nyquist zone selection and down conversion - but I have to do a bit of downconversion before the data is at a rate that I can run filters on it - that part of the chain is not flexible
If I can digitally mix at the fpga input pin with this technique I can work around that and make the whole receiver more configurable
If I can digitally mix at the fpga input pin with this technique I can work around that and make the whole receiver more configurable
MAn arduino uno rr4 wifi to blynk platform , dynamically changing the wifi network name and password without altering the code is feasible , just that it requires the use of a wifi provisioning feature, somewhat like a blynk edgent library. Heard of Blynk.Edgent? Dat is designed for dynamic wifi management, it will wat wil make this work
AYes I heard about it and used it on esp32 but it doesn't support arduino uno r4 wifi
MYou sure about that ?
AYes I am sure, there is another library for embedded MCUs called blynk.ncp but I don't know how to use it if you know can you help me?
MThen you should use Arduino IoT Cloud
AIs it possible to change the wifi network through the arduino iot cloud application?
GHi, can anyone help me program a circuit? I'm a beginner and I have no idea how to do it
TWhat kind of circuit?
KNope. Still in Shenzhen.
MIt will allow u to manage device settings, u can add ur wifi credentials there too, and u wont have to be uploading new code each time you doin it. you can update the wifi network name SSID and password directly from the dashboard there
MHow do I resolve this error ?
ADo you mean that I can change it through the word network or is there another way?
G
GThe circuit has a buzzer, an LED, 1 button (which is like a sensor), 1 button which is like a reset, an LCD screen, a switch and a Potentiometer. The project has to work as follows, as soon as I start simulating the project (which will be in Tinkercad) the word count and the zero below will appear on the LCD screen, as soon as I press the button that serves as a sensor, the counter it will show 1, ring the buzzer and turn on the LED at the same time and so on continuously, if for example I press the button 5 times, each time it is pressed it will appear on the LCD, depending on the time that the potentiometer marks and the buzzer and LED turn on together. The Potentiometer will serve as time in seconds to carry out the count, so for example if the Potentiometer has a delay of 4000 and I press the button, the counter will only show on the LCD after 4 seconds and the buzzer and the LED will also ring (the Potentiometer goes from 0 to 4000). When I press the switch down, it will show the Potentiometer information on the screen with the word Potentiometer at the top and the count in seconds below and when I move it up the switch will show the counter on the screen again. The other button will reset the counter
ZNice plan, where are you currently having issues?
MYh that network section is where u should enter ur wifi information
MOnce you enter the SSID which is ur network name and then password, click that configure button to save the changes thrn try to connect to the new network
GI don't really know how to program what I want to do, I needed someone to help me program
ZAh I see, I doubt you will find someone to help with that,
However if you are learning how to program and are stuck let us know we may be able to help.
However if you are learning how to program and are stuck let us know we may be able to help.
GYes, I'm in the process, but I only know the basics, but I appreciate your attention
AThnx for help
AI don't know how to flash arduino uno r4 wifi can someone help me?
AI don't know how to do step 4
Z@32bitSaviour @Ming Is yocto just
westbut for embedded linux. As in does this just help me keep a handle on my builds rather than a million scripts?MIn essence - yes.
You still have a million scripts, they are just packaged nicely and called The Yocto Project!
You still have a million scripts, they are just packaged nicely and called The Yocto Project!
ZAha that's insane that embedded linux is that huge, that the build tool is a project, as opposed to being shipped with a project.
MEmbedded Linux is that huge. A typical distribution contains thousands of files - but then a Linux system is meant to be for more heavyweight tasks.
MIf I wanted ‘small’ then I would go hand-rolled or just use something like zephyr.
MI think so. It is just scale that differs.
Zwell you want a good build tool rather than a small one
Zwhat is the scale difference would you say, I'm still looking at things very
baremetalMI mean scale in terms of features, entirely based off, what each of them offers - yocto would be just more complex.
For example, I think it would take a lot to have layering in zephyr but is it even needed.
And look at the learning curve on yocto.
For example, I think it would take a lot to have layering in zephyr but is it even needed.
And look at the learning curve on yocto.
ZI know this has been explained before, but can you explain
layering to a layman?MLayering is simply having groups of related build instructions bundled in a 'layer'. You can then stack those layers on top of each other, e.g:
You can then build up your final system by adding in layers that contain the useful stuff you need (and therefore also don't include the layers of non-useful stuff you don't need)
This allows for complex systems to be almost pieced together in a way that is manageable.
You can then build up your final system by adding in layers that contain the useful stuff you need (and therefore also don't include the layers of non-useful stuff you don't need)
This allows for complex systems to be almost pieced together in a way that is manageable.
MThe scale is not clear cut.
For 'simple things' where for example you are 'just' reading sensor data and then either acting upon it to control something or pass that data upstream, a full blown Linux type system would probably be an overkill.
While I don't know RTOS's that well, but for more complex systems, where you may want to have task isolation (in Linux every task is isolated from each other, so that an issue in user-space with one task should not affect any other tasks) or if you just need more computational power, then Linux makes it easier, as the OS handles a lot of it for you.
The other things would be familiarity with the system (I'm always going to be biased towards Linux, as I 'think' in Linux and can see the solution faster in Linux) or if an essential library is only available for one particular OS.
For 'simple things' where for example you are 'just' reading sensor data and then either acting upon it to control something or pass that data upstream, a full blown Linux type system would probably be an overkill.
While I don't know RTOS's that well, but for more complex systems, where you may want to have task isolation (in Linux every task is isolated from each other, so that an issue in user-space with one task should not affect any other tasks) or if you just need more computational power, then Linux makes it easier, as the OS handles a lot of it for you.
The other things would be familiarity with the system (I'm always going to be biased towards Linux, as I 'think' in Linux and can see the solution faster in Linux) or if an essential library is only available for one particular OS.
MWhen you are building your bike each layer is a like a specific group of things that serve a distinct purpose (grouping things into a function), so a layer could be the engine or custom seats. All the layers stack up and you have your bike. Building from layering allows you to build (and easily modify) your bike. Swapping the engine would be as simple as replacing or modifying the engine layer. This modular system gives you flexibility to create what you need while making it easy to modify without complexity.
Zso in this kind of world you don't need to rebuild layers that have not changed. I.e my OS and Drivers havent changed but my application has changed so just use a cache for the former 2 and re build my application.
ZAlmost like releases in elixir, but that's a bundle as opposed to a build system.
MIn a Yocto Project - the caching actually happens at the recipe level. It is more like a
make system - it only rebuilds the package if the recipe has changed.MFor layers - think of them more as bundles of functionality.
I have a great CLI system and I now want to add a GUI. All I need to do is add a Graphical layer, say XFCE and I can then select which recipes from that layer I actually want to use in my project.
Or, I have a great system from a RaspberryPi but now I want to build it for an ODroid. All I need to do is swap my RaspberryPi layer for the ODroid layer - keeping all the other layers in place.
I have a great CLI system and I now want to add a GUI. All I need to do is add a Graphical layer, say XFCE and I can then select which recipes from that layer I actually want to use in my project.
Or, I have a great system from a RaspberryPi but now I want to build it for an ODroid. All I need to do is swap my RaspberryPi layer for the ODroid layer - keeping all the other layers in place.
ZMiiiing, what in the world is a recipe. Is this like a list of dependecies and their versions?
MA Recipe is the instructions in a Yocto Project for the how tos:
- How to download the source
- How to configure the source
- How to compile the code
- How to install the resulting files into the final image
The collection of files that actually end up in the final image are then apackage.
