Your browser is not supported. Please update it.

17 May 2022

Artemis upgrades pCell and business model, targeting private 5G  

Only a small number of start-ups in the mobile industry achieve instant breakthrough, especially if they focus on chips or equipment. This is not the world of apps or web services, where a small investment and some clever marketing can create a sensation, and attract a skyhigh valuation. Companies focused on wireless infrastructure are more likely to have to adopt the old motto – ‘if at first you don’t succeed, try, try and try again’. 


An example is Artemis Networks, whose CEO, Steve Perlman, first unveiled the pCell (personal cell) back in 2014, claiming it would make conventional 4G redundant. Bold claims persisted – in 2020 he claimed the pCell enabled networks that were faster than 5G. But behind the big talk, there has been a process, familiar to many start-ups, of tweaking the business case and the target applications to try to move beyond trials or small-scale, esoteric roll-outs, and hit the big time. 


Now Artemis is back in the spotlight with an upgraded product, geared towards virtualized RAN; and a new go-to-market strategy focused on private networks and CBRS spectrum. Last week, it unveiled the pCell Multi-Gigabit LTE/5G vRAN, which promises to deliver 10 times the capacity of conventional cellular networks in the same amount of spectrum.  


This is the latest in a series of high profile attempts to disrupt the cellular market. Having dabbled with 4G, fixed wireless and TV white space spectrum, but without commercial breakthrough, Artemis resurfaced in 2020 with its eye on the USA’s newly opened CBRS band, in which licensed and shared spectrum schemes are both supported. This was clearly attractive to a company that is focused on spectral efficiency, and also provided the prospect of many new deployers of 4G and 5G, which would potentially drive an alternative ecosystem to that of the large operators, with their reliance on established vendors and foundational technologies. 


The latest step forward is a distribution partnership with Federated Wireless, a leading light in the CBRS ecosystem thanks to its spectrum access system (SAS) services and private wireless platforms. The pCell has also powered a real-world deployment at the SAP Center sports stadium in San Jose. 


Some analysts, which had gone quiet on Artemis, are reawakening their interest. Wall Street firm MoffettNathanson is one, and the company’s Craig Moffett was quoted by Artemis saying: “pCell is the most innovative technology I’ve seen in wireless in a very long time, upending everything we’ve been taught about wireless. We’ve all grown up thinking that spectrum is a scarce resource, but pCell breaks that equation, delivering over 10 times the performance of existing LTE and 5G technology.” 


Not only does Artemis see new potential in CBRS – and in other flexible spectrum schemes that are starting to emerge outside the USA – but it has adjusted its business model too. It originally planned to sell its technology directly to large equipment makers and MNOs, but its biggest publicly discussed engagement, in trials with Dish Wireless in its H-band spectrum, did not lead to a 5G relationship (although Dish chairman Charlie Ergen name-checked Artemis in a recent earnings call); and high profile work with Nokia was put on the back burner, reportedly when its main sponsor, former CTO Hossein Moiin, left the company. 


Now, in keeping with the shift of innovation towards private and enterprise cellular, Artemis will mainly sell to venues and private network deployers. 


This is the source of hope for many challenger vendors that have found it hard to break into the charmed circle of the large MNOs’ supply chains. Enterprise networks are based largely on small cells and are usually greenfield, so the ecosystem is likely to evolve differently to that of macro networks, with different deployers, performance requirements and cost assumptions. In that context, there is no reason to assume that the same vendors and ecosystems will succeed in enterprise networks as in macro public RANs, or that a new set of customers will automatically want the same solutions as the tier 1 telcos. 


“We decided we could not go in the front door, so we went in the back door,” Perlman commented, citing the very long buying cycles of the MNOs. Even large companies have despaired of quick results in macro 5G networks and decided to focus on private wireless for now – HPE, for instance, complained about the difficulty of getting large operators to adopt a challenger solution in the core, and has focused its 5G core efforts on the enterprise. 


“The pCell vRAN’s unique ability to multiply spectrum capacity by over 10 times (and growing) will ultimately transform the lives of the billions of people who use mobile networks every day,” said Perlman in a statement, having lost none of his capacity for hyperbole in the past seven years.  


However, he puts technical detail behind the grand words, together with case studies such as the SAP Stadium. All this adds up to claims that the pCell, in its latest form, can deliver “10x the capacity of conventional LTE/5G networks in the same amount of spectrum. While existing LTE/5G networks average less than 100Mbps in 20 MHz of spectrum, a network powered by pCell with equivalent spectrum averages over 1Gbps, and over 7.5Gbps in 150 MHz, faster than any mobile network in the world.”  


Like many innovative approaches to spectral efficiency, pCell aims to exploit interference between cells in dense networks or user populations. The pCell claims a breakthrough in densification because, rather than avoiding interference like conventional wireless technologies, or waiting in line to use spectrum, the technology exploits interference, combining the interfering radio waves to create the unshared personal cell for each device, providing the full wireless capacity to each user at once, even in extremely high density situations. 


Perlman explained that data rates remain the same, in a pCell, even when more users are added, because “we work in the opposite way to most cells, which try to avoid interference … These antennas deliberately interfere with each other and where the radio waves run into each other, they add up.” This creates an individual signal around each user or group, the foundational idea that launched pCell back in 2014. 


John Sculley, former CEO of Apple, said Artemis pCell is “one of the largest technology breakthroughs I’ve seen in my career. However, the real genius is getting around longstanding barriers to entering the market”. This would be achieved by delivering “vastly higher performance” at lower total cost of ownership than conventional networks, targeting private networks in CBRS.  


“For the first time, large mobile customers including venues, campuses and enterprise can choose mobile infrastructure based on performance, quality and economics, and most significantly, enabling future applications,” Sculley said in the Artemis press release. 


Subject to availability of components, pCell systems will be increasingly available through 2022, the company said. 


In a white paper released after its debut, Artemis detailed how it achieved multi-gigabit services with sub-millisecond latency and it also ran a video demonstration transmitting at 800Mbps to multiple phones next to each other in a 20 MHz piece of spectrum. The system, as described then, required a single virtualized LTE hub, which acted as a software defined radio, driving up to 32 radio antennas which could be placed just about anywhere in the target area. That architecture works well in stadiums and conference centers, where Nokia conducted trials in 2015, and where the latest iteration of pCell is finding its biggest initial deployments.  


In 2017, the start-up introduced the pWave Mini, a 15mm-wide base station that could be daisy-chained into cables, for deployment on rooftops and along buildings or streetlights. This dramatically shrunk the size and cost of its technology, to slash the cost of entry to mobile networks for new entrants such as cablecos, or smaller ISPs. 


That iteration of pWave expanded support to more protocols and to unlicensed or licensed bands from 600 MHz to 6 GHz. The technology promised to deliver over 50 times better spectral efficiency than current LTE on smartphones and to support costs of under $100 each at volume, potentially ushering new entrant service providers into the mobile world with far lower cost of entry.