Table of Contents
What is the integration revolution of phased‑array radar T/R chips?
The integration revolution of phased‑array radar T/R (Transmit/Receive) chips refers to the technological evolution from discrete, function‑specific chip architectures based on gallium arsenide (GaAs) to monolithic, multi‑function integrated chips based on gallium nitride (GaN). The core drivers are the material leap from GaAs to GaN and the paradigm shift in chip design from “multiple discrete chips combined” to “single‑chip multi‑function integration.” This ultimately delivers breakthroughs across four dimensions: power density, integration level, reliability, and system cost.
TL;DR
- What it is: The integration revolution of phased‑array radar T/R chips is the transition from discrete multi‑chip solutions using GaAs to monolithic multi‑function integrated solutions using GaN, encompassing both material upgrade (GaAs → GaN) and architectural upgrade (discrete → multi‑function integrated).
- How it works: GaN HEMTs (High Electron Mobility Transistors), with their wide bandgap (3.4 eV), high breakdown field (≈3.3 MV/cm), and high 2DEG density, deliver 5–10 times higher output power density than GaAs on the same chip area. Meanwhile, MMIC (Monolithic Microwave Integrated Circuit) technology integrates power amplifiers, low‑noise amplifiers, phase shifters, attenuators, T/R switches, and control logic onto a single chip, enabling amplitude/phase control and beamforming via on‑chip digital interfaces.
- Why it matters: The integration revolution shrinks T/R module volume by over 60%, increases power density by 5×, and reduces system cost by 30%–50%. This drives active phased‑array radar from high‑end military equipment into widespread civilian applications, including spaceborne, airborne, naval, 5G base stations, satellite internet, and automotive radar.
1. Definition and Core Elements
A phased‑array T/R chip is a critical RF front‑end device integrated within a phased‑array radar T/R module. It performs essential functions such as transmit power amplification, receive low‑noise amplification, phase shifting, attenuation, and T/R switching. The chip’s performance directly determines the overall radar system’s parameters and capabilities.
This “integration revolution” consists of two core elements:
Element 1: Material Revolution – from GaAs to GaN
Traditional GaAs devices offer a power density of about 1–2 W/mm, whereas GaN HEMT devices typically achieve 10 W/mm—5–10 times higher. GaN’s wide bandgap (3.4 eV vs. GaAs’s 1.43 eV) provides higher breakdown voltage (over 600 V) and superior high‑temperature operation. GaN devices combine high output power and low noise figure at microwave frequencies.
Element 2: Architectural Revolution – from Discrete to Multi‑Function Integrated
Traditional T/R modules use a “chip‑set” approach—multiple separate dies (power amplifier, low‑noise amplifier, phase/amplitude control, limiter, switch, etc.) are individually packaged and assembled on a PCB. The new generation of multi‑function integrated chips combines phase shifters, attenuators, driver amplifiers, serial‑to‑parallel conversion control logic, and power modulation onto a single MMIC. Going further, multi‑channel integrated chips (e.g., 4‑channel phase/amplitude control MMICs) can handle multiple transmit/receive channels simultaneously on one die.
2. Working Principle
Phased‑array T/R chips operate in transmit and receive modes, with beam steering as the core functionality.
Transmit Mode
In transmit mode, the RF excitation signal enters the T/R chip, first passing through the phase shifter and attenuator for amplitude and phase adjustment—this is the physical basis of beamforming. The adjusted signal is pre‑amplified by a driver amplifier and then finally amplified by the power amplifier (PA), delivering output power from several watts to tens of watts (X‑band GaN PA MMICs already achieve 10–30 W). The amplified RF signal is radiated into space through the antenna element.
Receive Mode
In receive mode, the weak echo signal captured by the antenna (as low as microwatt level) first passes through a limiter to protect subsequent circuits from large signals, then enters the low‑noise amplifier (LNA) for low‑noise amplification. A good T/R chip LNA can achieve a noise figure as low as about 0.5 dB. The amplified signal is then phase‑adjusted by the phase shifter and output to the back‑end receiver.
Beam Steering Principle
The core advantage of phased‑array radar is electronic scanning rather than mechanical rotation. The T/R chip precisely adjusts the phase and amplitude of each antenna element’s transmit/receive signal using digitally controlled phase shifters and attenuators. This causes the electromagnetic waves radiated by each element to constructively interfere in a specific direction and destructively interfere elsewhere, enabling beam steering with instantaneous electronic agility.
Two Main Architecture Modes:
- Analog Beamforming: Amplitude/phase control is implemented at RF or IF level using analog phase shifters/attenuators. This is currently the mainstream approach, with T/R chips performing the core amplitude/phase control.
- Digital Beamforming (DBF): Amplitude and phase parameters for each channel are computed in real time at the digital baseband using FPGAs or dedicated chips, allowing multiple independent beams to be generated simultaneously. DBF architecture is a major future trend for T/R modules.
3. Key Advantages
Power Density Leap
GaN HEMT power density reaches 10 W/mm, compared to only 1–2 W/mm for traditional GaAs devices—a 5‑ to 10‑fold improvement. For the same output power, GaN chip area shrinks to 1/5 to 1/10 of the GaAs solution.
Efficiency Improvement
Domestic GaN modules have achieved 45% power‑added efficiency (PAE) in the 6–18 GHz band. Fraunhofer IAF‑developed GaN devices exceed 80% efficiency below 3 GHz. In comparison, conventional GaAs power amplifiers typically operate at 30%–40% PAE.
Size and Weight Reduction
Using 3D packaging, dual‑channel T/R modules can be integrated into a 30×70 mm footprint—a 60% volume reduction and 300% integration improvement. MMIC solutions drastically reduce T/R module dimensions and weight, enabling more T/R channels within the same antenna aperture.
Cost Reduction
After domestic production, GaN chip costs can drop by 30%–50%. Silicon‑based GaN (GaN‑on‑Si) solutions are about 40% cheaper per chip than traditional GaAs. With GaN chip penetration rising from 52% in 2025 to an expected 78% in 2030, T/R channel costs are projected to decline 8%–10% annually.
Thermal Stability and Reliability
GaN devices operate stably at 200°C ambient temperature, while GaAs devices degrade significantly at high temperatures. GaN’s intrinsic robustness makes it a natural choice for demanding environments like spaceborne applications.
4. Comparative Analysis: GaAs Discrete Solution vs. GaN Multi‑Function Integrated Solution
| Comparison Dimension | GaAs Discrete Solution | GaN Multi‑Function Integrated Solution |
|---|---|---|
| Power Density | 1–2 W/mm | 10 W/mm (5‑10× improvement) |
| Power‑Added Efficiency (PAE) | 30%–40% | 45%–80% (depending on band) |
| Chip Count | 4–6 discrete dies (PA+LNA+phase shifter+attenuator+limiter+switch) | 1 multi‑function MMIC die (all functions integrated) |
| T/R Module Volume | Baseline (e.g., 100×65×10 mm) | Reduced by 60% (e.g., 30×70 mm) |
| Maximum Junction Temperature | <150°C | >200°C |
| Breakdown Voltage | ~15–20 V | >600 V |
| Per‑Channel Cost | Baseline (~$48 average per die) | 30%–50% lower than GaAs (domestic); GaN‑on‑Si 40% lower |
| Applicable Scenarios | Medium/low‑power, traditional radar where cost is less sensitive | High‑power, spaceborne/airborne/naval, 5G base stations, satellite internet |
| Technology Maturity | Mature (decades of use) | Rapidly maturing (52% penetration in 2025, expected 78% by 2030) |
5. Step‑by‑Step Implementation: T/R Chip Selection and Integration Process
The following is a standardised implementation process for T/R chip selection and integration in phased‑array radar system design:
Step 1: Define System Specifications and Constraints
First determine the operating frequency band (e.g., X‑band: 8–12 GHz; Ku‑band: 12–18 GHz), required transmit output power, receive noise figure target, array size (number of T/R channels), and platform constraints (size, weight, power—i.e., SWaP). An airborne phased‑array radar typically integrates 1,000–2,000 T/R channels.
Step 2: Select Material Process Route
Choose the material route based on power requirements:
- Output power < 5W: GaAs or SiGe solutions are viable, with lower cost.
- Output power 5–10W: GaAs can meet the requirement, but GaN offers better efficiency and miniaturisation.
- Output power > 10W: GaN is mandatory.
Also evaluate the trade‑off between GaN‑on‑SiC (high power density, up to 8 W/mm) and GaN‑on‑Si (cost advantage, 40% lower cost).
Step 3: Evaluate Chip Integration Level
Choose between “discrete chip‑set” and “multi‑function single‑chip”:
- Discrete: high flexibility, individual chips can be replaced, but larger volume and more complex assembly.
- Multi‑function single‑chip: smaller size, higher reliability, simpler assembly—the future trend.
Step 4: Thermal Design and Cooling Assessment
GaN chips can generate heat flux exceeding 1 kW/cm² (hot spot under the gate). Evaluate whether thermal solutions—such as micro‑channel cooling, thermal interface materials, etc.—meet the system’s thermal budget.
Step 5: System Integration and Test Verification
After die attach, wire bonding, and assembly, perform full‑channel amplitude/phase consistency calibration, transmit power testing, receive noise figure testing, and overall array pattern validation.
6. Common Mistakes
Mistake 1: Assuming GaN Always Increases System Cost
Many designers assume GaN solutions are inherently more expensive than GaAs. In reality, while a single GaN die may have higher wafer cost, its power density is 5–10 times that of GaAs—the required die area for the same output power is much smaller. Additionally, multi‑function integration reduces chip count and assembly steps. Overall, domestic GaN production can lower system‑level costs by 30%–50%. Confusing individual die cost with total system cost is the most common pitfall in selection.
Mistake 2: Overlooking GaN’s Thermal Management Challenges
GaN’s high power density is a double‑edged sword—it delivers high power in a small area, but also creates extremely high heat flux (hot spots can exceed 1 kW/cm² under the gate). Many engineers carry over cooling designs from the GaAs era, leading to premature performance degradation or failure due to excessive junction temperatures. GaN solutions require dedicated thermal design (micro‑channel cooling, high‑thermal‑conductivity substrates, thermal interface materials, etc.)—you cannot simply reuse the old cooling approach.
7. FAQ
Q1: What exactly is the integration revolution of phased‑array radar T/R chips—from GaAs to GaN, from discrete to multi‑function?
This revolution has two dimensions: Material: moving from gallium arsenide (GaAs) to gallium nitride (GaN), raising power density from 1–2 W/mm to 10 W/mm and efficiency from 30%–40% to over 45%. Architecture: moving from multiple discrete dies (PA, LNA, phase shifter, attenuator, limiter, switch, etc. separately packaged) to a single multi‑function MMIC chip, reducing T/R module volume by 60% and increasing integration by 300%. Together, these drive phased‑array radar towards higher performance, smaller size, and lower cost.
Q2: How much can GaN T/R chips improve detection range over GaAs?
For the same array aperture and number of channels, GaN output power is at least 5 times higher than GaAs for the same die area. Radar detection range scales with the fourth root of transmit power (radar equation: R ∝ P_t^(1/4)). A 5× power increase theoretically improves range by about 49% (5^0.25 ≈ 1.495). In practice, system noise figure and antenna gain also matter, but the magnitude is significant.
Q3: What percentage of phased‑array radar cost is attributable to T/R chips?
T/R modules typically account for 50%–70% of the total phased‑array antenna cost. In satellite applications, the antenna subsystem represents about 52.5% of the total satellite cost, and T/R modules make up 50% of the antenna subsystem cost—so T/R modules are roughly 26% of the satellite cost. As the core component of the T/R module, T/R chips have a direct impact on overall system cost.
Q4: What is the difference between digital beamforming (DBF) and analog beamforming? What role does the T/R chip play?
Analog beamforming achieves amplitude/phase control at RF or IF using analog phase shifters/attenuators inside the T/R chip—this is the current mainstream. Digital beamforming computes amplitude and phase parameters in the digital baseband using FPGAs or dedicated chips. DBF offers the advantage of generating multiple independent beams simultaneously (e.g., some early‑warning radars can track 500 targets at once) and provides larger dynamic range. In DBF architectures, the T/R chip evolves from an “amplitude/phase control executor” to an “RF digitalisation front‑end,” with even higher integration requirements.
Q5: Where do SiGe and CMOS processes stand in phased‑array T/R chips?
SiGe BiCMOS and silicon‑based CMOS processes are inferior to GaAs and GaN in transmit power, but they offer lower material costs and are suitable for low‑cost, low‑power‑density radar systems. Silicon‑based chips excel in integration, low power consumption, and mass‑production cost control. In civilian applications (e.g., 4D imaging radar, automotive radar), silicon‑based T/R chips have already dropped to the hundred‑yuan (RMB) price level in China. Different processes serve different scenarios—they are not simply replacements for one another.
