IEEE Consumer Electronics Magazine - May/June 2022 - 8

Open Source
Currently, many electronic devices (both OS
and proprietary) are designed using OSS. For
instance, CASE tools like KiCad or LibrePCB,
allow designing printed circuit boards (PCBs).
Besides, all kinds of electronic devices can be
physically built using OSH, such as the open
CNCs and pick-and-place machines available at
Openbuilds, or the 3-D printers by Prusa which,
in turn, are controlled using the GRBL OSS.
The rest of this section focuses on the central
pieces of electronic devices, i.e., central processor
units (CPUs), graphic processor units (GPUs),
microcontroller units (MCUs), digital signal processors
(DSPs), application-specific integrated circuits
(ASICs), or field programmable gate arrays
(FPGAs). These computing and controlling hardware
devices, also known as industrial property
cores (IP cores), typically involve hardware and
software development (firmware, operating systems,
and applications) and, in some cases, also
communication capabilities.
Most IP cores rely on standard Hardware
Description Languages (HDL), such as Verilog or
VHDL, to specify their design, behavior, architecture,
and data flows. These standards are,
respectively, supported by OSS tools, such as
Verilator or GHDL. Recently, non C-based OS
HDLs are also emerging, such as PyMTL, based
on Python.
CPUs are the heart of our laptops and personal
computers. The lack of independent
instruction sets and the very few CPU manufactures
dominating the marketplace, makes it difficult
for new companies to become competitive.
Thus, there are few OSH projects related to
CPUs. Probably the most notable one is OpenSpark,
currently offering two CPUs. In spite of
this, there are well-known OSS toolchains and
operating systems allowing CPU management
and programming, being GNU/Linux the most
widely used today.
GPUs can be considered CPUs optimized for
processing huge amounts of information, performing
the same mathematical operation over
multiple data in a very efficient way. This is possible
thanks to the use of hundreds of cores running
in parallel (compared to the few cores
typically included in a CPU). This feature makes
GPUs excellent candidates for data-intensive
processes, such as those involved in computer
8
vision or deep learning. In this line, the NVIDIA
Jetson XavierNXGPU is being used in deep learning
OS projects, such as NVDLA. Unfortunately,
limitations like those described for CPUs regarding
proprietary cores also apply to GPUs, being
Miaow one of the few OSH GPUs available.
MCUs are general-purpose electronic devices
focused on information processing and control.
They are primarily made for embedded solutions
and can be adapted to a wide variety of applications.
Probably, the most relevant OSH project in
this area is Arduino. It includes OSH PCBs and also
OSS, such as boot-loaders and a set of C/Cþþ
libraries and wrappers aimed at easing hardwarerelatedprogramming.
TheArduinoC/Cþþ libraries
have significantly evolved thanks to the contributions
made by many vendors and OS developers.
Moreover, it currently supports many different
platforms from third-party manufactures. Especially
relevant are the low-cost System-on-a-Chip
platforms esp8266 and esp32, two of the OSH projects
that have gainedmore attention lately.
Most OSS for programming MCUs relies on
GCC/Make. However, some recent initiatives advocate
for the use of interpreted languages, such as
MicroPython, Lua, or JavaScript, although the limited
computational resources available at MCUs
makes their adoption quite challenging. The operating
systems used in most MCUs (e.g., Contiki or
FreeRTOS) significantly differ from those used in
PCs. Actually, most software engineers consider
them as a set of libraries providing a Hardware
Abstraction Layer.
Figure 1 displays some of the most wellknown
platforms, currently being used to prototype
CE: different types of Arduinos, esp8266s,
esp32s, Raspberry Pis, a Beagle-Bone, and a
Sipeed RISC-V k210. Among them, only the last
one (a 64-b microcontroller with an AI-dedicated
unit) can be really considered OSH. In the background,
some of the CE devices developed in our
laboratory are also displayed, such as an UAV
using Ardupilot, an electronic smart lock based
on esp32, a small robot based on Arduino and a
game console based on Raspberry Pi. Most of our
projects are based on OS, being some of them
available as full OS solutions (e.g., a voice
assistant used as a medicine reminder or a control
system for home boilers), whereas others
are being developed and commercialized by
IEEE Consumer Electronics Magazine
IEEE Consumer Electronics Magazine - May/June 2022

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - May/June 2022
