6G news and analysis for the week ending Friday, 3rd July, 2026.
US spectrum policy set the pace this week, which is quite apt given that it’s the Fourth of July weekend. The FCC announced plans to auction 160 megahertz of Upper C-Band spectrum, forming a contiguous 440 megahertz mid-band block. Chairman Brendan Carr framed it as a return to American wireless leadership, though the band serves broad mid-band capacity rather than 6G alone. The move came alongside warnings that the US trails China in mid-band holdings. The GSMA reinforced the spectrum theme from MWC Shanghai. It quantified the mid-band that future networks will require and pressed for early, harmonised decisions ahead of WRC-27, the 2027 World Radiocommunication Conference.
Another common thread this week concerns waveforms. The US Department of War awarded Cohere Technologies $28 million to build a multi-waveform sensing prototype using its Zak-OTFS technology. It is a pointed bet against the OFDM waveform that Ericsson and Nokia favour for integrated sensing. The award is prototype-stage, not a standards decision, but it gives OTFS fresh credibility. Grounding the enthusiasm, Dell’Oro Group projected $35 billion in AI RAN revenue to 2030, cautioning that AI RAN will not expand the overall RAN market.
Elsewhere, SK Telecom showcased quantum cryptography for the AI and 6G era. Four research papers round out the edition, covering semantic communications, uplink efficiency, gigantic MIMO and a unified security framework for AI-native 6G. Enjoy the weekend.
The FCC will vote on 22 July on an order to auction 160 megahertz of Upper C-Band spectrum (3.98–4.14 GHz) next year. Chairman Brendan Carr announced the plan, which would harmonise terrestrial operations across a 440 megahertz “super-band” spanning 3.70–4.14 GHz. The auction exceeds the 100 megahertz minimum that Congress mandated by July 2027. Winning bidders could begin services in top-75 markets by December 2030. The announcement came via an FCC press release. Carr said the order “sets the stage for the FCC to auction 160 megahertz of prime, mid-band spectrum next year.”
What’s new: The plan creates a single contiguous 440 megahertz block by combining the Lower and Upper C-Band. It would make 3,248 new flexible-use licences available. Winning bidders would fund incumbent transition costs and rebates for upgraded aircraft radio altimeters, following parallel FAA rulemaking.
Why this matters: Mid-band is the workhorse layer that balances coverage and capacity. This is the latest step in a US spectrum acceleration tied to the ‘One Big Beautiful Bill Act’. The timing reflects wider anxiety about China’s lead. A former US official, Paul Rosenzweig, warned in Newsweek that China holds nearly four times more mid-band spectrum, a claim reported by the state-owned Global Times, so treat the framing with caution. It is worth noting the C-Band is a general mid-band capacity resource, not exclusively a 6G band, and the political narrative here is pronounced.
Implications for 6G standards: Standards are technology neutral, but the holdings assembled now shape which carriers can afford early 6G spectrum. A large contiguous block supports the wide channel bandwidths that 6G will need and feeds mid-band planning ahead of the separate 7 GHz processes that will underpin FR3.
View source material to read more — FCC
View source material to read more — Global Times
The GSMA reported on its Policy Leaders Forum at MWC26 Shanghai. Two sessions examined how to monetise 5G while planning spectrum for 6G. Participants identified four 5G monetisation pathways: Fixed wireless access, enterprise and private networks, enhanced connectivity, and cloud and AI-enabled capabilities. On spectrum, forecasts showed Greater China will need 2.5 to 3.3 GHz of mid-band between 2035 and 2040. The global requirement is estimated at roughly 2 to 2.5 GHz.
What’s new: The quantified mid-band figures are the substantive element. Participants agreed that spectrum planning must begin years before commercial 6G deployment, as AI-driven applications become more data-intensive.
Why this matters: The 5G conversation has now shifted from technology rollout to commercial returns, with future growth expected from enterprise, industrial and AI services rather than consumer broadband. Capabilities such as network slicing and differentiated quality of service are creating value-based offerings. The forum stressed that policy, investment certainty and harmonised spectrum matter as much as technology. This complements the FCC item above, underlining mid-band as the contested resource ahead of WRC-27.
Implications for 6G standards: 3GPP specifications must accommodate whatever spectrum regulators release. The mid-band emphasis aligns with the upper-mid-band planning that runs through Release 21 discussions and vendor submissions to the 2025 3GPP 6G Workshop.
View source material to read more
Cohere Technologies has won a $28 million contract from the FutureG Office within the US Department of War, formerly the Defense Department. It funds a multi-waveform radio access network prototype for integrated sensing and communications (ISAC). The system pairs Cohere’s Zak-OTFS waveform with conventional OFDM. It aims to turn commercial cellular infrastructure into a sensing layer for drone detection while looking like ordinary traffic. The award expands a National Science Foundation VINES Phase 2 project, and was announced by Cohere and reported by several telecoms news outlets. Tom Rondeau, Principal Director for FutureG, said Cohere’s OTFS modulation “carries information directly in the sensing domain, delivering massive communications and sensing performance advantages in high-Doppler environments.”
What’s new: The award follows the FutureG prototype request we reported on 12 June, with proposals due 3 July. It signals a clear preference for OTFS over OFDM for sensing. Cohere claims OTFS is about four times more accurate than OFDM given the same spectrum, and the prototype must comply with the DoW-backed OCUDU (Open Centralized Unit Distributed Unit) platform.
Why this matters: OFDM (orthogonal frequency division multiplexing) underpins 4G and 5G and remains the likely 6G baseline. ISAC is a new service area where the industry is undecided, so a rival waveform has room to gain ground. Ericsson and Nokia have invested heavily in OFDM and argue their sensing trials already work. The IEEE ComSoc analysis stresses this is a prototype validation, not a path to standards adoption. Cohere’s Zak-OTFS featured in the 2025 3GPP 6G Workshop, and we have tracked OCUDU since October 2025.
Implications for 6G standards: OTFS (orthogonal time frequency space) could be proposed as an IMT-2030 Radio Interface Technology candidate in the 2027 to 2029 window. The realistic outcome keeps OFDM as the 6G New Radio baseline, with OTFS targeted at high-mobility, non-terrestrial and ISAC niches rather than universal use.
View source material to read more — Cohere Technologies
View source material to read more — IEEE ComSoc Technology Blog
Dell’Oro Group projects cumulative AI-RAN revenue of $35 billion over the next five years, 2026 to 2030. However, it does not expect AI-RAN to expand the overall RAN market. GPU RAN projections were revised upward to a $1 billion-plus market by the end of the forecast. Near-term deployments will centre on AI-for-RAN, single-purpose non-GPU architectures and 5G. The report was published by Dell’Oro Group. Stefan Pongratz, Vice President, said: “AI-RAN is already happening and will scale ahead of 6G. At the same time, these tools will enhance the RAN, but they are unlikely to expand the overall RAN market.”
What’s new: The upward GPU RAN revision is the notable data point. The firm expects AI-RAN to generate little incremental RAN revenue by the end of the forecast, even as suppliers introduce software subscription models.
Why this matters: This tempers the AI-RAN enthusiasm we have tracked across NVIDIA, ODC and other vendor announcements. Incumbents are well-positioned in the initial phase, driven by AI-for-RAN upgrades on existing hardware. The top five RAN suppliers held roughly 96 per cent of 2025 RAN revenue. The finding is consistent with Dell’Oro’s evolutionary thesis, which we followed through its January and June reports. Remember that AI-RAN is a technology enabler, not a market expander.
Implications for 6G standards: AI-native RAN capabilities are becoming foundational for Release 21, but the economics rest on efficiency rather than new revenue. That sharpens the pressure on 3GPP to deliver measurable efficiency gains, not just new features.
View source material to read more
SK Telecom unveiled next-generation quantum cryptography technologies at Quantum Korea 2026 in Seoul, running from 2 to 4 July. The showcase covered PIC-based quantum key distribution (QKD) and quantum random number generation (QRNG). It also included wireless and satellite QKD, plus two security products named Q-HSM and Q-SSE. SKT frames the work as preparation for security in the AI and 6G era. Ryu Tak-ki, Head of Network Technology at SK Telecom, said the company would “contribute to meeting security demands in the AI and 6G era.”
What’s new: SKT has implemented a 10 Gbps-class QRNG on a chip measuring 10 by 10 millimetres. It is developing an integrated QKD chip and a system for wireless links up to 30 kilometres. Its Q-HSM module applies quantum cryptography to edge devices such as drones and robots on 6G networks.
Why this matters: 6G security is increasingly treated as foundational rather than bolted on. Post-quantum cryptography (PQC) was named one of three top priorities by the 3GPP security group SA3, as we reported in September 2025. SKT’s ATHENA white paper embedded quantum cryptography and PQC in its zero-trust architecture, covered in February. The 6G link here is forward-looking rather than a deployment, and this is an exhibition showcase, so read it as direction of travel.
Implications for 6G standards: PQC is already active on the 3GPP SA3 agenda for 6G. QKD sits largely outside 3GPP scope, but it informs the quantum-safe architecture that operators such as SKT and KT advocate in their submissions.
View source material to read more
This survey argues that the heterogeneous security landscape of AI-native 6G cannot be addressed by isolated, technology-specific fixes. It develops a cross-layer threat taxonomy spanning infrastructure, network, AI, privacy and security management. It then maps threats to countermeasures, including standards harmonisation treated as a security function.
Why this Matters: Security in AI-native 6G is fragmented across technologies and standards efforts. A cross-layer framework that treats harmonisation as a security function aligns with the foundational approach that 3GPP SA3 is taking. It complements the SK Telecom quantum item elsewhere in this edition.
View source material to read more
The paper introduces the Quantized Semantic Age of Information (QSAoI). It is a metric that captures the trade-off between freshness and semantic efficiency under short-packet, low-latency conditions. A foundation-model-based framework then dynamically optimises quantization precision and physical blocklength to minimise QSAoI over fading channels.
Why this Matters: Semantic communications and edge computing are candidate 6G techniques, but the semantic and physical layers remain poorly connected. This work bridges that gap for latency-constrained transmission, addressing a practical barrier to real-time 6G services that carry meaning rather than raw bits.
View source material to read more
The authors propose a semantic-aware multiple access scheme for uplink-dominant 6G use cases, such as cooperative vehicular streaming. Nearby vehicles first share observations over vehicle-to-vehicle (V2V) links to find spatial redundancy. They then transmit only non-redundant, semantically important image patches to the base station.
Why this Matters: Uplink capacity is a persistent constraint, and cooperative vehicular streaming generates heavy visual data. Exploiting overlapping fields of view among nearby devices could ease uplink pressure, supporting more resource-efficient and sustainable 6G systems.
View source material to read more
To preserve 5G massive MIMO apertures at FR3, the upper-mid band, 6G base stations may need four to eight times more antennas. This gives rise to “gigantic MIMO.” The paper adapts the WAX framework to explore decentralized processing under hard-boundary constraints, using non-cooperating hardware modules.
Why this Matters: FR3 (roughly 7 to 24 GHz) is the primary new band for early 6G, but scaling antenna counts raises processing and interconnection challenges. Decentralized architectures could manage that complexity, informing practical base station design for the upper-mid band.
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.