6G Research in 2026: What Engineers Are Actually Working On

6G research is at the stage where the interesting work is happening in labs and standards bodies, not product announcements. Commercial deployment is targeted for 2030 in Japan and South Korea, with broader global rollout extending into 2032-2035. The research happening now will define what 6G is — there's still genuine room for decisions about which technology directions to pursue and which to abandon. Understanding the current state of 6G research means understanding both the ambitious targets and the engineering problems that make those targets non-trivial.

What 6G Targets That 5G Cannot Deliver

5G's theoretical maximums — 20 Gbps peak data rate, 1ms latency, 10^6 devices per km² — sound impressive. In practice, deployed 5G in most markets delivers 100-900 Mbps to users on sub-6GHz frequencies (the mmWave 5G that delivers multi-gigabit speeds has extremely limited coverage due to propagation characteristics). Network latency is typically 10-30ms end-to-end. This gap between theoretical and practical performance is partly a spectrum allocation story, partly an infrastructure density story, and partly a fundamental limit of the 5G radio interface.

6G targets aim to close this gap while extending performance dimensions that 5G doesn't address:

Terahertz (THz) frequency use: 6G research is exploring spectrum in the 100 GHz - 3 THz range, often called the "THz gap" because it was historically underutilized. Carrier frequencies in this range support data rates in the Tbps range because more bandwidth is available. The fundamental engineering challenge is propagation — THz waves have very high atmospheric absorption, which limits range to tens of meters in many scenarios.

Sub-millisecond latency: True 0.1ms air interface latency for industrial and haptic applications. This requires fundamentally different frame structure designs compared to 5G's 0.5-1ms slot durations.

Integrated sensing and communication (ISAC): 6G networks are being designed to simultaneously function as radar/sensing networks alongside communication networks. The same signal that delivers data also provides centimeter-resolution positioning and environmental mapping. This has applications in autonomous vehicles, industrial automation, and smart infrastructure.

AI-native network architecture: 5G added AI as an optimization layer on top of a conventionally designed network. 6G research is exploring making AI intrinsic to the physical layer — where resource allocation, beamforming, and channel estimation are performed by neural networks that are part of the radio interface standard.

Who's Leading Research

The 6G research landscape is distributed across standards bodies and national programs:

Samsung Research published a 6G white paper in 2020 and has been consistently the most active contributor to 6G technical discussions. Samsung's target is first 6G service in 2030, aligned with South Korea's national ambition to lead 6G deployment. Their research focus includes THz array antenna systems and AI-native PHY layer design.

NTT Docomo (Japan) has set a 2030 target for initial 6G service. Japan's 6G program includes the Beyond 5G promotion strategy with ¥500 billion (~$3.5B) in government funding. NTT's research includes THz backhaul for dense networks and orbital angular momentum multiplexing.

Nokia Bell Labs and Ericsson are the primary European contributors, working through the Hexa-X and Hexa-X-II projects funded by the EU's Horizon program. Their research emphasis includes the radio access network (RAN) aspects and the integration of satellite and terrestrial networks.

The US Next G Alliance, under the Alliance for Telecommunications Industry Solutions (ATIS), coordinates US industry 6G research. The US approach is more industry-led than government-directed compared to Asian programs. Academic contributions from MIT, NYU WIRELESS, and UT Austin are significant.

China has a parallel 6G program through the IMT-2030 (6G) Promotion Group under the Ministry of Industry and Information Technology. Huawei, ZTE, and the Chinese academic research institutes are active contributors. The US-China technology bifurcation (see: semiconductor geopolitics) creates uncertainty about whether 6G will develop as a single global standard or two parallel standards, as happened with some 5G components.

The THz Engineering Problem

Terahertz propagation physics are not cooperative. The atmospheric absorption at THz frequencies is primarily due to water vapor and oxygen, causing severe signal attenuation — at 300 GHz, attenuation is approximately 10 dB/km in typical atmospheric conditions, compared to fractions of a dB/km at sub-6GHz frequencies. At 1 THz, attenuation increases further and is sensitive to humidity.

This limits THz 6G links to short range: indoor (meters to tens of meters), short outdoor hops (street-level microcells), or satellite-to-ground in atmospheric windows (specific frequencies where absorption is lower). For dense urban environments, this is actually workable — very dense small cell deployments can use THz links for backhaul and serve users at street level. For coverage in rural or suburban areas, THz contributes nothing; sub-6GHz and mid-band frequencies will remain essential.

The antenna technology required for THz is qualitatively different from existing mmWave antennas. At 300 GHz, an antenna array with 64 elements fits in a fingernail-sized chip. Massive MIMO at THz frequencies is a materials and fabrication challenge — silicon CMOS transistors can operate at THz frequencies in laboratory demonstrations, but with the power efficiency required for commercial base stations and user devices, CMOS performance falls off. III-V semiconductor materials (InP, GaN) offer better THz performance but are more expensive and harder to integrate with the digital processing logic.

What 6G Might Enable

The capabilities 6G research is targeting would unlock specific applications that 5G cannot support:

Industrial haptic control — surgeons remotely operating from different continents with force feedback — requires round-trip latency under 1ms consistently, not just in ideal conditions. 5G's 10ms practical latency doesn't support this; 6G's 0.1ms target does.

Pervasive ISAC opens the possibility of infrastructure that simultaneously supports communication and environmental monitoring without additional sensor deployment. Smart city applications, precision agriculture, and autonomous vehicle coordination all benefit from sensing-as-infrastructure.

Tbit/s connections to fixed wireless access points would eliminate the last-mile problem for fiber-alternative internet access in dense environments.

The 2030 deployment target for first 6G service means the standards being written today in 3GPP working groups will shape what engineers are building with in 2030-2040. The research is not purely academic, but the translation from research paper to deployed network is where most of the uncertainty lies.