5G air interface debate is revived as C Spire tests Cohere OTFS

Some mobile stakeholders have been frustrated that the first wave of 5G standards, 3GPP Release 15, have stuck with OFDM in the air interface, rather than adopting a more radical modulation technology.  However, additional standards are likely to creep into Release 16 standards, which will provide renewed opportunities for new approaches from giants like Huawei, to start-ups like Cohere Technologies, with its OTFS (Orthogonal Time Frequency Space) modulation scheme.

In a clear sign that OTFS is not off the table yet, US mobile operator C Spire has applied for permission to test the technology in its headquarters in Ridgeland, Mississippi, using spectrum in the 3650-3700 band. In its application, C Spire says it believes OTFS may deliver higher spectral efficiency, multipath diversity and throughput than current OFDM, as well as eliminating intra-cell interference, and being more resilient to signal fading.

Cohere CEO Shlomo Rakib learned about OTFS when studying at the University of Texas at Austin and then applied it to wireless, while also bringing his professor from Austin,  Ronny Hadani, on board as Cohere’s CTO. In another stamp of approval from US operators, Verizon’s former CTO, Dick Lynch, recently joined Cohere’s board.

C Spire is the sixth largest US operator by subscribers after the big four and US Cellular and aims to be a 5G frontrunner despite its relatively small size. It plans to launch 5G fixed wireless services in the first phase of deployments, starting in eight Mississippi markets, and says it will use “multiple technologies and spectrums depending on the options available in any given area” in order to cover 200,000 consumers and businesses in the state with 5G. It has already demonstrated 5G fixed wireless TV services, delivering C Spire Fiber TV content over Nokia-enabled wireless connections; and in common with Verizon, it has conducted a 5G test using equipment from Phazr.

If it goes ahead with its Cohere trial, it will be looking to increase the advanced services, and the cost efficiencies, it can support with 5G, to offset its regional nature and its lower economies of scale, compared to the national operators.

Cohere went public almost two years ago about OTFS, though it stresses that the technology can coexist with OFDM to support smooth migration from current networks, or coexistence between different 5G flavors. Indeed, OTFS is not an entirely new modulation scheme as it sits on top of OFDM, but is a “scheme for diversity that adds to MIMO with an equalization technique”, as Phil Marshall of Tolaga Research explained.

That integration could be critical to win acceptance for a technology from a small player. Many observers believe OFDM will be the most efficient option for sub-6 GHz spectrum and mobile broadband use cases, but that alternatives will be needed to optimize performance in millimeter wave bands, or for emerging IoT and ultra-low latency applications.

Cohere says OTFS creates a two-dimensional view of the delay and signal fading of a wireless channel. It spreads information across time and frequency, which allows signals to benefit from diversity in the channel while also penetrating through concrete and glass, and to achieve high levels of immunity to fading, multipath and other signal impairments.

In its early tests in 2015-16, it trialled its technology among tall buildings and mountains and in moving vehicles, and said that the OTFS radios never faded, even over distances above four kilometers, as OFDM ones would have done. A minimum of 4bps/Hz was achieved in 10 MHz of spectrum in the 3.5 GHz band during those initial trials, driving speeds of 120Mbps to 320Mbps. Those rates will be much increased as Cohere is incorporating Massive MIMO radios with 64×64 arrays into its plans, as well as implementation in millimeter wave bands.

“We solved some very fundamental problems around imperfect channel information and timing references that the wireless community has been struggling with forever,” Rakib told EETimes when the company emerged from stealth mode. “The good news is this is a thin layer above OFDM-based cellular silicon today, requiring small transforms in the modulation process that should require less than 10% in addition to today’s silicon.”

The best opportunity for smaller innovators like Cohere is likely to be to secure a big name supporter to fight its case for inclusion, though like many mmWave pioneers, it is developing its technology anyway, for particular applications, such as fiber extension, backhaul and campus small cell networks, which do not need to wait for mobile standards.

Rakib says the company has almost finished the architectural definition of its ASIC, probably to be made in 2018 to support gigabit speed wireless connections via a low cost home or office gateway. He added: “Today we are a systems company, but if we become part of the 5G standard, we will have to add on licensing models, and eventually we might have to build silicon.”

Larger companies are also hoping to get their favored air interface technologies into the standards in Release 16. For the mmWave bands, China Mobile has validated Huawei’s Filtered OFDM and ZTE’s FB-OFDM in its high frequency tests, but has provided no details of the results as yet. Last year, Huawei described a “unified” air interface which draws on several approaches at once and allows different sub-bands, within the baseband, to be configured individually for different purposes.

Its design is based on three key concepts – Filtered OFDM (fOFDM), Sparse Code Multiple Access (SCMA) and polar code. Each has a contribution to make to a standard which can be adaptable without compromising on the performance requirements, says Huawei.

fOFDM enables a flexible OFDM interface in which sub-bands are ‘filtered’ so that they can be configured differently from one another to support different use cases with different demands on the network. It can also, according to Huawei, provide twice the system throughput of LTE’s OFDM interface.

SCMA supports huge numbers of connections and also boosts system throughput by optimizing power allocation across multiple SCMA layers. Huawei said that, in its test, SCMA delivered three times more uplink connections per link and increased downlink system throughput by 80%. Meanwhile, polar code is a new coding scheme which claims to deliver increased gain, compared to the turbocoding used in LTE. Polar techniques claim to support channel coding of any code rate with an appropriate code construction to fit any future service requirements.

Huawei says it has trialled these three concepts in early-stage tests on outdoor macrocells, which also used two of the most hotly tipped technologies to underpin 5G – advanced multiuser MIMO and full duplex radio (the latter allowing for simultaneous transmit and receive on the same frequency, doubling spectral efficiency). The MU-MIMO implementation supported up to 24 users, and up to 24 parallel layers of transmission on the same time-frequency resources. The tests took place under the auspices of China’s IMT-2020 Promotion Group, an umbrella initiative for 5G R&D.

Some of the items included in the study list for Release 16 are the use of unlicensed spectrum for 5G NR; integrated access/backhaul in the same millimeter wave channels, for small cells; non-terrestrial networks (using 5G NR for satellite communications, especially in underserved areas); enhanced vehicle communications; and use of non-orthogonal multiple access.

The departure from OFDM will come, then, in the second wave of 5G standards, and will be mainly targeted at low power IoT applications which require non-scheduled network access for uplinks. That would allow devices to wake up, transmit and immediately sleep again, without having to wait to be scheduled, which would save energy and signalling overhead, and could also be important for urgent mission critical messages.

Alan Carlson, European head of InterDigital, a major contributor to 3GPP standards, wrote in a blog post: “Two licensed band cellular definitions are expected below 6 GHz. The second will be especially tailored for IoT support. It is less likely in my opinion that this will be based on OFDM. OFDM is great for video, but in the IoT world of trillions of randomly occurring access events, its strict synchronization needs render a severe handicap.”

Qualcomm’s own pitch has been non-orthogonal RSMA (resource spread multiple access) technology, plus a new multiplexing technique which would allow traffic requiring very low latency to take priority automatically and to use RSMA. This technology uses time and frequency spreading and overlaps users in a way that aims to improve network efficiency and power consumption. It can support mobility and downlink meshing, as well as network-assisted mesh on the uplink.

Cohere is not alone in thinking that some of this radical thinking should have been done for Release 15. Last fall, the 3GPP NR Release 15 study item agreed to support CP-OFDM (Cyclic Prefix OFDM) for the downlink in enhanced mobile broadband applications, with the complementary technology, DFT-Spread OFDM, working alongside CP-OFDM on the eMBB uplink. This has some continuity with LTE, which uses OFDM in the downlink and DFT-S OFDM in the uplink, but increases the importance of CP-OFDM.

Cohere’s VP, Anton Monk, believes that, in the haste to get first standards out and pre-empt operators which might deploy pre-standard kit, 3GPP has failed to consider all its options fully. “Some big vendors wanted to get something done fast and slap a 5G name on it — we’ll see how widely it gets deployed,” he said.

He claims that the first set of specs – whose non-standalone version will be frozen at the end of this year, six months ahead of the original schedule – add little to what is also being developed for LTE-Advanced Pro. They are “LTE with Massive MIMO and beamforming — nothing really new except for including Huawei’s polar codes in the control channel and using LDPC (low density parity check) everywhere else,” he added.

But Qualcomm argues that DFT-S OFDM provides link budget benefits while CP-OFDM supports MIMO spatial multiplexing advantages. It believes there are advantages to using both technologies on the uplink, which can adaptively switch between them to get the best of both worlds.

Ericsson has also been a big supporter of the CP-OFDM waveform and has welcomed its inclusion in the first phase of 5G NR specs. Senior researcher Ali Zaidi and master researcher Robert Baldemair explained in a recent blog post why Ericsson backs CP-OFDM over the many waveform options presented to 3GPP.

“The trend has been to tweak OFDM in any possible way – sub-carrier wise filtering or pulse shaping, filtering of groups of sub-carriers, allowing successive symbols to overlap in time, dropping cyclic prefix, replacing cyclic prefix with nulls or with another sequence,” they said. “Various waveforms became strong contenders for 5G—numerous research publications showed CP-OFDM being outperformed. At one point, we felt that every multicarrier waveform was going to be part of 5G, except CP-OFDM.”

Ericsson Research said its tests showed CP-OFDM performing best on the indicators that matter most to operators – compatibility with multi-antenna technologies, high spectral efficiency and low implementation complexity.

“Moreover, CP-OFDM is well-localized in time domain, which is important for latency critical applications and TDD deployments,” the researchers wrote. “It is also more robust to oscillator phase noise and Doppler than other multicarrier waveforms. Robustness to phase noise is crucial for operation at high carrier frequencies (e.g. mmWave band).”