mmWave radio spectrum, which will be critical for 5G rollout, has seen a lot of research attention in the last few years and shows great promise in early field tests. Through the field work, we have learned more about mmWave wireless systems as they pose their own unique challenges and our industry is working on innovative technology to address them. Although there is still some way to go before we see full-scale commercial deployments, mmWave based cellular systems are ready for large-scale network trials. The good news is that mmWaves have passed the reality test.
Why mmWave spectrum?
The explosion of mobile communications over the last two decades has been an incredible leap forward for communications and the world. But by 2020, we anticipate almost 30 billion sensors and mobile devices connecting to the network wirelessly. Thus, as we evolve to 5G networks between 2020 and 2030, we are coming up against some very real physical limits. Radio spectrum is getting scarcer. We are already suffering from a global bandwidth shortage, and 5G will only exacerbate the situation with its much wider range of use cases and related applications. Demand for wireless data traffic will grow 10,000-fold within the next 20 years, and without suitable new spectrum many of the promising use-cases and applications will never become a reality.
Regulatory bodies around the world are currently working towards opening up new spectrum bands from 6-100 GHz and Nokia Bell Labs researchers are working feverishly in developing new technologies to meet the challenges of these mmWave bands. The good news is that lately we have seen some very promising results with the technologies in these new spectrum bands. By nature these higher frequencies provide much more bandwidth than the current spectrum below 6GHz — currently in use for mobile communication. Very promising but, until recently, using mmWave frequencies has only been explored in labs and theory, not in the real world.
mmWaves require new technology
mmWave frequencies start at around 30 GHz with deployments up to 100 GHz. In low band mmWave systems with up to 40 GHz carrier frequency, larger bandwidths can also be achieved by aggregating multiple carriers. For example, 10 x 100 MHz carriers can be aggregated to achieve a bandwidth of 1 GHz. Bandwidths from 100 MHz to 2 GHz are envisioned to be realized at the higher end of the mmWave frequency range.
To eliminate any need for paired spectrum and address the elastic demand both in up- and downlink, Time Division Duplex (TDD) is the preferred duplexing method in these mmWave cells. Thanks to the higher frequencies of mmWave spectrum, high-dimension antenna arrays can be deployed in a very compact manner.
Figure 1: An example of a 12-element phased array showing how small the antenna arrays can be (from SiBeam Inc.)
There are, of course, challenges that mmWave systems must overcome. These high dimension arrays have narrow beamwidths and, because they operate at these higher frequency bands, they suffer from high penetration loss and diminished diffraction. mmWave systems also need to address other requirements, such as high peak and cell edge rates, which can be achieved using either Orthogonal Frequency Division Multiplexing - Cyclic Prefix (OFDM-CP) or single carrier, low-complexity modulation formats. Ultra-low latency can be obtained by scaling the transmission time interval (TTI) length (or slot) to a tenth of a 4G/LTE TTI. Given the availability of larger chunks of bandwidth at high mmWave bands (e.g. 71-76 GHz), a single carrier modulation is preferred because of its low peak to average power ratio (PAPR), which efficiently operates the power amplifiers (PAs), thus extending the expected range and also positively impacting out-of-band emissions.
mmWaves - Where are we today?
Nokia Bell Labs has been at the forefront of mmWave research since 2012 with channel measurements studies of mmWave bands with partner universities such as New York University, Aalto and Aalborg University. Particular attention to channel angular spread and coherence times has been the focus of measurement studies with Universidad F. Santa Maria in Chile. Nokia has demonstrated several experimental mmWave systems across bands ranging from 15 GHz to 73 GHz. In 2014, Nokia and DOCOMO assembled 15 participants from industry and academia, including ten 3GPP members, to initiate a study on 5G channel models. The scope of the study was primarily to address the channel model needed by standards bodies, such as 3GPP and ITU, with a focus on frequencies up to 100 GHz. At GLOBECOM 2015, the findings were published in a white paper.
The measurements indicate that smaller wavelengths introduce increased sensitivity in the propagation models to the scale of the environment and show some frequency dependence with regard to path loss, as well as increased occurrence of blockage. Further, the penetration loss is highly dependent on materials and tends to increase with frequency. The shadow fading and angular spread parameters are larger and the boundary between line-of-sight and non-line-of-sight depends, not only on antenna heights, but also on the local environment. The small-scale characteristics of the channel, such as delay spread and angular spread, and the multipath richness of the channel is somewhat similar over the frequency range, which is encouraging for extending the existing 3GPP models to the wider frequency range. The standardization of 5G channel models in 3GPP is nearing completion and primarily based on the GLOBECOM white paper.
Several demonstrations were developed to prove the results of the theoretical work. A first version shown at Mobile World Congress ’15, using a steerable LENS antenna demonstrating tracking of moving users, operated @73 GHz as a bidirectional system with a peak throughput rate of 2.3 Gbps using SISO (Single Input, Single Output) in 1 GHz bandwidth. The system supported IP bearer data with a one-way latency less than 1 ms. It was field-tested with DOCOMO and Verizon in outdoor and indoor environments, both in Japan and the US. A maximum range of 160–200 meters with peak throughput exceeding 2 Gbps was achieved. At Mobile World Congress ’16, Nokia Bell Labs demonstrated the next version, a unidirectional 15 Gbps system incorporating 2-stream MIMO (Multiple Input, Multiple Output) in the 2 GHz bandwidth.
As a next step, the antennas were miniaturized.
To that end, fully integrated 16-Element 90GHz phased array ASICs were developed at Nokia Bell Labs along with a packageless integration of the ASICs directly on an organic printed circuit board (PCB) substrate housing the patch antennas. The transmitter demonstrated a measured equivalent isotropically radiated power of 34dBm at 90GHz and a receiver noise figure of 7dB per element. The system can establish multi-gigabit wireless links at a distance of tens of meters. This low-cost integrated solutions does not require the use of any waveguide components or expensive material and follows a traditional die-on-PCB assembly process.
At the 2016 Brooklyn 5G Summit, Nokia Bell Labs demonstrated beamscanning using a phased array @60 GHz with the 1GHz bandwidth system. Because the half lambda element spacing decreases with higher frequencies (~ 2.5mm @60 GHz), an 8 x 8, half-lambda, spaced antenna array could easily fit into an area smaller than 20 square mm. As an example, the size of a 12 element phased @ 60 GHz is illustrated in Figure 1.
Lab prototypes are one thing, trial systems running on commercial platforms bringing 5G closer to commercial reality are a much more complicated undertaking. Nokia has demonstrated the world’s first 5G radio commercial product – our AirScale/AirFrame demo @15 GHz, which was also showcased at Mobile World Congress ‘16. The Proof of Concept system will be available @28 GHz too, using Massive MIMO and beamforming, including phased array technology. At 5G World 2016 in London, the World’s first 5G-ready network was demonstrated; multiple field trials are planned worldwide in the 2016–17 timeframe with the top 32 operators based on AirScale/AirFrame products.
Nokia Bell Labs is also pushing the borders of spectral effiency to get the most capacity out of the spectrum. It successfully demonstrated spectrum efficiencies as high as 100 bps/Hz in the 28 GHz millimeter waveband. This will make wireless transmission of large data files far more economical than what is currently achievable. The prototype, using a novel physical layer technology, including massive MIMO, guarantees huge system capacity and spectral efficiency. It achieved a peak transmission rate of over 50 Gbps for a short range while supporting applications demanding latency requirements as short as 250μsec. At that speed, a 50 GB movie can be downloaded in less than eight seconds.
mmWave systems have passed their reality check. The next step is commercial deployment, with first trials of specific 5G use-cases in early 2017. Our research will continue with the goal of creating an elegant small cell solution. Ideally, it will be simple, low-cost, using a fully flexible baseband technology, massive MIMO and phased arrays covering bands up to 100 GHz. This solution will go a long way to achieving the promise of 5G and the many new applications and use cases it supports. To learn more about mmWave and 5G technologies and how they impact the way forward, read The Future X Network: A Bell Labs Perspective.
 These include 8k video streaming, augmented reality, different ways of data sharing and various forms of machine-type applications, including vehicular safety, different sensors and real-time control requiring ultra-low latency.