With massive growth in mobile data traffic, traditional cellular networks are in danger of exceeding capacity. But Artemis, a Silicon Valley start-up, may have just turned the problem on its head. Jacob Harris explains.
We are at the beginning of era where smart mobile devices form an integral part of our day-to-day lives. A rapid increase in the use of smart phones, tablets and connected cars, coupled with data-intensive applications such as HD video streaming, is creating an almost exponential increase in mobile data demand worldwide – and it’s a trajectory that shows no sign of letting up.
According to the Cisco Visual Networking Index: Global Mobile Data Traffic Forecast Update, 2015–2020 white paper, global mobile data traffic will increase nearly eightfold between 2015 and 2020. Mobile data traffic will grow at a compound annual growth rate (CAGR) of 53% from 2015 to 2020, reaching 30.6 exabytes per month by 2020.
“Asia Pacific will account for 45% of global mobile traffic by 2020, the largest share of traffic by any region by a substantial margin. North America, which had the second-largest traffic share in 2015, will have only the fourth-largest share by 2020, having been surpassed by Central and Eastern Europe and Middle East and Africa. Middle East and Africa will experience the highest CAGR of 71%, increasing nearly 15-fold over the forecast period. Asia Pacific will have the second-highest CAGR of 54%, increasing nearly nine‑fold over the forecast period.”
Many industry bodies believe this growth will see our finite spectrum running out of data capacity. But Silicon Valley start-up, Artemis, may have created a solution: pCell technology, a wireless system the company claims can achieve mobile data rates over 50 times the capacity of current systems. And all while maintaining compatibility with legacy LTE devices.
Current cellular networks work by each tower transmitting a radio signal that forms a large cell. Each cell must avoid interference with other cells. Every user within a cell’s radius shares the cell’s capacity with every other user, effectively taking turns so they don’t interfere with each other. According to Artemis, even with more spectrum and smaller cells, demand is outpacing capacity and we will soon hit a physics upper limit.
“We’re running out of spectrum and current technologies that are using densification through small cells in 4G LTE networks are not enough for satisfying such a skyrocketing growth in data traffic,” says Artemis principal scientist and co-founder Antonio Forensa.
pCell, the company says, provides a solution to this problem by increasing spectral efficiency. Instead of dodging interference, pCell exploits it by combining radio signals to synthesise small, personal cells, or pCells, that follow each device around. This means that instead of users taking turns – sharing the capacity of one large cell – each user has access to full wireless capacity at all times. Instead of cell towers, the technology uses pWave radios: small transmitters about the size of a lunchbox (connected to data centres) that create signals that interfere with each other and are calibrated to synthesise pCells about 1cm in size around every device in range.
“When we compare the different wireless technologies, we see that 3G HSPA+ achieves 1.2bps/Hz per sector average down-link spectral efficiency as opposed to a 4G LTE 1.7bps/Hz – so we can see there’s only a marginal gain. But with pCell technology we have achieved spectral efficiency of up to 59.3 bps/Hz with 16, 4G LTE devices clustered in 1m2 (see table). So we’re achieving these results in a worse than a worst case scenario,” says Antonio.
To explain how pCell achieves such a high spectral efficiency gain, Antonio compares a conventional cellular layout against a pCell architecture.
“A conventional cellular system has base stations (indicated by the blue dots in diagram…) that are placed based on a very specific cell plan. They can transmit a higher or lower power depending on whether it is a macro cell or a small cell (often small cells will overlap a macro cell umbrella).
“Performance varies dramatically for users (indicated by the red dots) depending on their location. So as a user moves from the cell centre towards the cell edge, the signal quality degrades because of path loss and shadowing as well as inter-cell interference, particularly in a frequency use one system such as LTE networks.
“On the other hand we have pCell (the blue dots indicate the position of the access point). In its simplest form it is single antenna transceiver that can be placed anywhere. They can transmit at higher power and they can also be placed in much higher density than a cellular layout because the interference from all these access points is what we exploit in order to create a high signal quality around the user’s location.”
The technology has started to pique the curiosity of some big players in the US. Late last year, Nokia Networks and Rearden LLC (Artemis’ parent company) signed a memorandum of understanding (MoU) to jointly test pCell in 2016 with wireless operators.
The test will initially take place in large indoor venues and other high density areas, and will be offered as pCell Proof-of-Concept deployments to selected Nokia Networks customers. The collaboration may also be extended to consider further advanced features that could be enabled by pCells, such as precise 3D location positioning.
Another company that is helping to test pCell technology is Webpass, a provider of high speed internet to residential and business customers in several US cities. Webpass has installed Artemis antennas in buildings it provides internet to.
What makes this partnership particularly interesting is that Webpass has recently been acquired by Google Fibre, giving the tech giant a first-hand look at pCell in action.