The current activity in mobile computing and wireless device connectivity (aka Internet of Things) is a source of continuing excitement for many engineers and software developers and a sizeable number of consumers as well.
Not for me, though. What has grabbed my attention is the underlying network infrastructure that makes this possible and the revolutionary new approaches emerging there.
Despite the current IoT buzz, not much is new in the mobile and IoT market since the year 2000 when trends and capabilities established then have now begun to take hold. That was when the first of several generations of powerful and low-power processor technology emerged to drive the mobile device revolution. It was also when IPv6 and a few years later its wireless 6LoWPAN extension rolled out and removed the limit on the number of URL available. This has made it possible to give unique Internet addresses not only to every human being on the planet, but all of their things as well. It is only recently that developers and their customers are beginning to fully realize the possibilities.
But none of this would not have been possible without the underlying wired and optical network infrastructure that delivers increasingly multimedia-rich content at 2.5 to 5 gigabits/second to the information superhighway’s wireless off-ramps, where it is managed and directed by next generation cellular base stations for delivery to our mobile and wireless devices.
It is because of its key role in enabling the wireless/cellular mobile/IoT environment that one of the most technically interesting market segments to me is the cloud-enabled radio access networks (RANs) being developed using a new generation of femto- and pico-cell wireless 3G/4G, multiple input-multiple output (MIMO) base stations.
The challenges of building RANs that meet the new demands are considerable. RANs reside between devices such as mobile phones, computers, or remotely controlled machines (or things) and provide connection with the core internet backbone network.
RANs are implemented on multiple standalone base stations (BTS), each covering a small area and which together cover broad geographic areas. While such base stations are increasingly costly to build and operate, new technologies are emerging to deal with this as well as the challenges of managing mobile users moving from place to place, bursty data transmission, and varying utilization rates, among many other complex coordination tasks.
Included in this week’s Tech Focus newsletter on “Building the mobile/wireless network infrastructure” are a range of design articles and technical papers on these challenges and as well as new hardware and software architectures that are emerging to deal with them. Of these, my Editor’s Top Picks are:
Designing an ARM-based Cloud RAN cellular/wireless base station. Barak Ullman, Chief Architect, ASOCS Ltd., describes a novel architecture for baseband processing using ARM’s Cortex A57 processors for use in mobile wireless base stations. It is used with a real-time, reconfigurable platform for use in cloud RAN applications.
Implementing effective floating point support in wireless base station designs . Avi Gal, Dmitry Lachover, and Itay Peled of Freescale Semiconductor describe how floating point-based DSP can be used to deal with the complex matrix inversion problems faced by designers of LTE wireless MIMO base stations.
Making wireless MIMO equalization more efficient with a multiprocessor DSP SoC. The authors demonstrate how Freescale’s its Qonverge B4860 SoC with StarCore SC3900 DSP cores was used to implement a 4×4 LTE MIMO equalizer.
Using SerDes in 4th Generation Wireless Infrastructure. Ajinder Singh discusses the challenges facing developers of the network equipment infrastructure for 4G cellular where there will be an unprecedented demand for high serial data rates between the main control radio equipment and the various distributed base stations.
Some other useful design articles for assessing the approaches a developer may have to take in this fast moving embedded infrastructure market segment include:
Beyond the complex network environment of these mobile and wireless delivery platforms, the challenges facing developers in this multicore design environment include not only the number of cores, but the fact that this is a heterogeneous mix of special accelerators, RISC based CPUs, and floating point digital signal processors.
According to Wim Rouwet, in “Building a Linux-based femtocell base-station using software performance engineering, ” this will require the use of sophisticated techniques and tools to be sure that the most efficient, low cost, and high performance design emerges. If you find his arguments compelling you can read more about it in a four part series on the principles of software performance engineering by Freescale’s Robert Oshana.
Embedded.com Site Editor Bernard Cole is also editor of the twice-a-week Embedded.com newsletters as well as a partner in the TechRite Associates editorial services consultancy. He welcomes your feedback. Send an email to firstname.lastname@example.org, or call 928-525-9087.