Access Official Technical PDF

Share Summary:LinkedInEmail

A 39 GHz phased-array transceiver developed by Tokyo Institute of Technology and NEC Corporation demonstrates how integrated calibration can improve beamforming accuracy in millimeter-wave wireless equipment. Presented in 2019, the design combines 64 antenna elements with a standard 65-nanometer CMOS process, targeting 5G infrastructure and other wireless devices. Its results are relevant to industrial automation decision-makers evaluating the communications technologies that may support connected equipment, IoT deployments, and future wireless systems.

The work addresses a practical challenge in phased-array radio design: coordinating the signals sent and received by many antenna elements so the array can steer a beam in the intended direction. The researchers reported low phase error and gain variation after calibration, alongside a 400 MHz over-the-air transmission demonstration. These findings describe a research-stage transceiver, rather than a deployed industrial product, but they show how calibration and silicon integration can contribute to scalable radio hardware.

Why develop a 39 GHz phased-array transceiver?

Millimeter-wave frequencies are an important part of 5G development. While much earlier research had focused on 28 GHz systems, the researchers identified 39 GHz as another band relevant to 5G in many parts of the world. Operating at these frequencies requires antenna arrays and beamforming to direct radio energy, rather than relying on a single antenna with broad coverage.

A phased array uses multiple antenna elements whose signal phases can be adjusted electronically. By changing those phase relationships, the system can steer its radiation pattern. Beamforming is the signal-processing approach that controls this pattern. Accurate steering matters because phase or gain mismatches among elements can distort the intended beam and increase unwanted radiation.

For industrial users, the broader relevance is connectivity: 5G and wireless IoT applications may include devices and infrastructure that need wireless links. However, the reported work does not establish performance in a factory environment. Its results focus on radio-frequency hardware and controlled over-the-air testing.

Built-in calibration targets array accuracy

The transceiver uses a 4-by-16 configuration, giving it 64 phased-array elements. Its built-in gain and phase calibration is designed to reduce differences in the signals associated with those elements. The system uses a local-oscillator (LO) phase-shifting approach for calibration, according to the project description.

The researchers reported a root-mean-square (RMS) phase error of 0.08 degrees. They described this as an order of magnitude lower than previous comparable results. They also measured a maximum gain variation of 0.04 dB across the full 360-degree tuning range. The whitepaper notes that transceivers developed to that point typically had gain variation exceeding 1 dB.

These measurements indicate how closely the transceiver maintained its phase and gain settings while tuning. In practical terms, more consistent control across the array can help the system form its intended beam. The figures should be understood as reported results for this design, not as guarantees for every implementation or operating condition.

CMOS integration and output characteristics

The design was fabricated using a standard 65-nanometer CMOS process, with a reported chip area of 12 mm2. CMOS is a widely used method for manufacturing integrated circuits. The researchers identified silicon-based components and the standard process as a route suited to cost-effective mass production, an important consideration for scaling 5G equipment.

The transceiver’s maximum equivalent isotropic radiated power (EIRP) was reported as 53 dBm. EIRP expresses the effective radiated power in a given direction relative to an ideal isotropic antenna. In this case, it provides an indication of the output from the 64-antenna array; it should not be confused with the power of a single antenna element.

Integration and array size are connected design considerations. The project team noted that increasing the array scale could enable greater communication distance. At the same time, the researchers identified adapting the transceiver for devices such as smartphones and base stations as a development challenge.

What the over-the-air test demonstrated

Testing took place indoors under anechoic-chamber conditions, in a non-reflective environment designed to limit echoes. The team conducted a one-meter over-the-air measurement and demonstrated wireless transmission of a 400 MHz signal using 64QAM modulation.

This result verifies a specific transmission capability under controlled test conditions. It does not by itself establish range, reliability, throughput, or performance in a working base station, mobile device, or industrial facility. Those deployment questions would require evaluation in the intended system and environment.

Implications for automation and connectivity planning

For system integrators and plant technology leaders, the study is best viewed as an example of radio-component development rather than a ready-to-deploy connectivity solution. It highlights several design factors to consider when assessing future wireless infrastructure:

  • Array calibration: Phase and gain consistency influence how accurately a phased array can steer its beam.
  • Manufacturing approach: A standard CMOS process can support integration and the researchers’ stated goal of cost-effective scaling.
  • Evidence context: Bench and anechoic-chamber measurements provide useful component-level evidence, but do not replace testing against site-specific coverage and operational requirements.
  • Application readiness: The researchers identified development for smartphones and base stations as a challenge, underscoring the distinction between a demonstrated transceiver and a finished system.

When evaluating 5G or wireless IoT plans, technical decision-makers should therefore distinguish between promising radio characteristics and validated application performance. The relevant questions include whether the eventual equipment supports the required environment, coverage, integration, and operational needs—details not established by this component demonstration.

Conclusion

The 39 GHz phased-array transceiver from Tokyo Tech and NEC combined 64-element beamforming with built-in calibration in a 65-nanometer CMOS design. Its reported 0.08-degree RMS phase error, 0.04 dB maximum gain variation, 53 dBm maximum EIRP, and controlled 400 MHz transmission test illustrate the potential of calibrated, integrated millimeter-wave hardware. For automation stakeholders, the practical takeaway is to track such developments as part of the evolving 5G landscape while requiring application-specific validation before drawing conclusions about industrial deployment.

Access Official Technical PDF

About The Author