Andy Pye explores how 5G will undoubtedly evolve our wireless networks to heights never before imagined and the challenges such advancement result in.
Researchers must not only address the requirements of unprecedented wireless data rates but also find solutions for network latency and responsiveness while accommodating a thousand-fold increase in capacity. And if all that weren’t enough, service operators are demanding that these advancements consume less energy than existing infrastructure.
So how do we begin to solve these complex challenges? The answer lies in prototypes and, more specifically, the kind of 5G prototypes that enable wireless researchers to test experimental ideas using real systems in real-world scenarios. When done right, these 5G prototypes can lay the foundation for rapidly increasing an organisation’s time-to-market schedule.
Setting a New Standard
Recognising the large amount of speculation regarding 5G networks, the world’s standardisation bodies, including the 3GPP, have recently begun to transition concepts into reality. Not surprisingly, the vision painted by IMT-2020, the NGMN, and the 3GPP is expansive. 5G researchers must now build the framework that will redefine our very existence — from vehicles and transport systems to manufacturing, energy, healthcare monitoring and more.
To do this, researchers are adopting new design approaches to help with the challenging task of defining, developing and deploying 5G technologies within a random access network. Most acknowledge that conventional approaches to vetting 5G technologies take too long and incur significant costs. Therefore, building a prototype and a proof of concept earlier in the process enables faster commercialisation. Noting the importance of research-enabling testbeds, National Science Foundation Programme Director Thyaga Nandagopal, said, “A viable prototype is increasingly the critical element in determining success or failure of a particular concept.”
To expedite the time it takes to produce a working prototype, many researchers have adopted a platform-based design approach that embraces a unified design flow. It starts with maths and simulation and then maps the algorithm in a system and working hardware.
Full dimentsion and massive MIMO
Consider Samsung, which has built one of the world’s first demonstrators of multi-antenna technology with a base station (BTS) that includes 32 antenna elements called Full-Dimension MIMO or FD-MIMO. FD-MIMO uses a 2D grid of antennas to create a 3D channel space. With FD-MIMO, service operators can place antenna grids at elevated positions such as buildings or poles and aim the antenna beams at users on the ground or in adjacent buildings to consistently deliver enhanced data rates.
Researchers at Lund University in Sweden have taken this multi-antenna concept to the next level with their Massive MIMO prototype. Massive MIMO increases the number of antennas in a cellular BTS to hundreds. Composed of low-cost technology, the grid of antenna elements focuses the energy directly at the user while enabling the hundreds of antennas to more easily detect weak signals from mobile devices. Additionally, Massive MIMO uses linear coding techniques to simplify the processing at the BTS.
As more BTS antennas enhance the mobile user data experience, we can see how theory confirms that Massive MIMO may also dramatically reduce the power consumed by both the BTS and mobile devices. Because multiple low-cost BTS antennas transmit lower aggregate power than a monolithic approach, the power consumed by the BTS may be reduced by a factor of 10 or more.
Fundamentally, enhanced data rates and increased capacity are constrained by spectrum according to Shannon’s theory on channel capacity. More spectrum yields higher data rates, which help service operators accommodate more users. As such, service operators around the world have paid billions of dollars for spectrum to service their customers, yet the currently available spectrum below 6 GHz is almost tapped out. Researchers are now investigating the possibility of deploying cellular networks above 6 GHz, specifically in the millimeter wave (mmWave) bands.
Worth noting is that the mmWave spectrum is plentiful and lightly licensed, meaning it is accessible to service operators around the world. Professor Ted Rappaport at New York University Wireless has been investigating mmWave as a possible evolutionary path for mobile networks since the mid-2000s. His pioneering channel measurement work has led researchers all over the world to reconsider their assumptions that mmWave mobile networks are either impractical or unfeasible.
Furthermore, researchers at Nokia Networks are also investigating mmWave technologies, and the preliminary results are encouraging. In 2015 alone, Nokia Networks demonstrated a fully working mmWave prototype that delivers the fastest rates ever recorded for mobile access. The Nokia Networks prototype composed of a BTS and a UE consistently streamed data at a rate of over 10Gbps at 73.5GHz. Nokia Networks’ achievement paints a promising future for mmWave and 5G.
5G offers many exciting new developments ultimately to improve our lives through enhanced connectivity and unlock tremendous economic value. But for us to reap these benefits, researchers need a faster path to prototype. Traditional approaches have grown too expensive and time-consuming, and the world is growing impatient. A platform-based design approach can deliver these new developments faster, just as the researchers at places like Lund University, Nokia Networks, NYU Wireless, and Samsung are already demonstrating.
Enhanced version of VirtualBench
National Instruments (NI) now has a high-performance model of VirtualBench, the software-based all-in-one instrument which combines a mixed-signal oscilloscope, function generator, digital multimeter, programmable DC power supply and digital I/O. With 350 MHz of bandwidth, four analogue channels and Ethernet connectivity, the new version offers increased functionality for engineers characterising and debugging new designs or automated test systems, leading to more efficient circuit debugging and validation. For higher current applications, there is a higher wattage programmable DC power supply with up to 3A for the 6V output channel and up to 1A for the +25V and -25V channels.
“With this next iteration, we have incorporated greater performance and more features that open up even more possibilities for engineers interacting with benchtop test equipment,” says Stefanie Breyer, vice president of R&D, data acquisition at NI.
Users still interact with VirtualBench through free upgradable software applications that run on PCs or iPads for all five instruments. Because VirtualBench uses contemporary computing platforms, engineers and scientists interact with their instruments using multi-touch displays, multi-core processors, wireless connectivity and intuitive interfaces. VirtualBench integrates seamlessly with LabVIEW system design software