LO leakage manifests as a spurious continuous wave (CW) tone precisely at the carrier frequency, independent of the modulated signal. This impairment originates from finite DC offsets in the in-phase (I) and quadrature (Q) baseband paths, which are upconverted directly to the RF carrier by the mixer. The resulting vector sum of these offsets creates a static carrier component that degrades Error Vector Magnitude (EVM) and violates spectral emission masks.
Glossary
LO Leakage

What is LO Leakage?
LO leakage, also known as carrier feedthrough, is the unintended radiation of the local oscillator signal at the transmitter output, primarily caused by DC offset voltages at the baseband input of a quadrature modulator.
In a direct conversion transmitter, LO leakage is exacerbated by local oscillator self-mixing, where parasitic coupling allows the LO signal to leak back into the mixer's baseband port, creating an effective DC offset. Mitigation requires a digital I/Q calibration loop that injects a compensating DC offset to null the carrier, often using an observation receiver to measure the residual tone and drive a minimization algorithm.
Key Characteristics of LO Leakage
LO leakage manifests as a spurious tone at the carrier frequency, degrading Error Vector Magnitude (EVM) and violating spectral emission masks. The following cards detail its root causes, measurement, and mitigation.
DC Offset Origin
The primary physical mechanism causing LO leakage is a DC offset voltage superimposed on the baseband I or Q signal at the modulator input. This offset mixes with the LO, producing an unmodulated carrier at the output.
- Self-Mixing: LO signal couples into the baseband input path and mixes with itself, generating a DC component.
- Component Mismatch: Transistor threshold variations in the mixer core create an inherent differential DC offset.
- DAC Imperfections: Non-zero mid-scale output voltage from the digital-to-analog converter introduces a static offset.
Spectral Signature
In the frequency domain, LO leakage appears as a distinct narrowband spike precisely at the carrier frequency (f_c), independent of the modulated signal's bandwidth.
- In-Band Interference: The tone falls directly in the center of the transmitted channel, corrupting the inner constellation points and raising the Error Vector Magnitude (EVM).
- Zero-IF Artifact: This is an inherent challenge of Zero-IF Architecture, as the LO and RF output share the same frequency, making filtering impossible.
- Power Ratio: Leakage is quantified as the ratio of LO power to total transmitted power, typically specified in dBc.
EVM Degradation Mechanism
LO leakage directly displaces the entire transmitted constellation from the origin, creating a systematic offset for every symbol.
- Constellation Shift: All ideal symbol points are translated by a vector equal to the LO leakage amplitude and phase.
- Origin Offset: The center of the constellation is no longer at (0,0), which is a critical metric for modulation accuracy.
- Demodulation Errors: The receiver interprets the offset as a legitimate signal component, increasing the Bit Error Rate (BER) for inner constellation points.
LO-RF Isolation
Finite isolation between the LO port and the RF output port of the mixer provides a direct leakage path that is independent of baseband signals.
- Substrate Coupling: In integrated circuits, the LO signal can travel through the silicon substrate directly to the RF output bond pad.
- Package Crosstalk: Bond wire mutual inductance and package pin capacitance create parasitic paths between the LO input and RF output.
- Board-Level Leakage: Poor PCB layout can allow LO signals to couple onto the RF trace through electromagnetic radiation or shared ground return paths.
Compensation Techniques
LO leakage is corrected digitally by intentionally adding a counteracting DC offset at the baseband I and Q inputs to cancel the carrier feedthrough.
- Static Calibration: A factory procedure measures leakage power using a spectrum analyzer and iteratively adjusts I/Q DC offsets to minimize the carrier tone.
- Envelope Detection Feedback: An on-chip power detector senses the LO leakage level, and a successive approximation algorithm nulls the DC offset.
- Blind Adaptive Cancellation: Algorithms estimate the leakage from the transmitted signal's statistical properties without requiring a dedicated observation receiver.
Temperature and Voltage Drift
LO leakage is not purely static; it varies with operating conditions, requiring periodic recalibration.
- Thermal Drift: Changes in die temperature alter transistor bias points, shifting the inherent DC offset of the mixer.
- Supply Voltage Sensitivity: Fluctuations in the power supply rail modulate the mixer's operating point, directly affecting leakage magnitude.
- Aging Effects: Long-term device parameter shifts, such as hot carrier injection, can permanently change the DC offset characteristics, necessitating field recalibration.
Frequently Asked Questions
Addressing the most common technical queries regarding local oscillator leakage, its root causes, and compensation strategies in direct conversion transmitters.
LO leakage is the unintended radiation of the local oscillator signal at the exact carrier frequency through the transmitter output. It occurs primarily due to DC offset voltages at the baseband input of the quadrature modulator. When a non-zero DC voltage is applied to the I or Q differential inputs, it mixes with the LO signal to produce a continuous wave tone at the carrier frequency. Secondary causes include LO-to-RF isolation limitations within the modulator package, where the LO signal capacitively or inductively couples directly to the RF output port, and PCB trace coupling where the LO routing leaks into the RF path. This spurious tone degrades the Error Vector Magnitude (EVM) and can violate spectral emission masks.
Enabling Efficiency, Speed & Accuracy
Intelligent Analysis, Decision & Execution
We build AI systems for teams that need search across company data, workflow automation across tools, or AI features inside products and internal software.
Talk to Us
Search across company data
Give teams answers from docs, tickets, runbooks, and product data with sources and permissions.
Useful when people spend too long searching or get different answers from different systems.

Automate internal workflows
Use AI to route work, draft outputs, trigger actions, and keep approvals and logs in place.
Useful when repetitive work moves across multiple tools and teams.

Add AI to products and internal tools
Build assistants, guided actions, or decision support into the software your team or customers already use.
Useful when AI needs to be part of the product, not a separate tool.
Related Terms
Understanding LO leakage requires familiarity with the physical impairments that cause it and the metrics used to quantify its impact on transmitter performance.
DC Offset
The primary physical cause of LO leakage. An unwanted constant voltage added to the baseband I or Q signal, typically generated by local oscillator self-mixing or component mismatch in the modulator. This static offset shifts the signal constellation, and when upconverted, manifests as a pure, unmodulated carrier tone at the LO frequency.
- Self-Mixing Mechanism: LO signal couples into the baseband input path and mixes with itself
- DAC Imperfections: Non-zero output from digital-to-analog converters at zero input code
- Correction: Cancelled by applying an intentional opposing DC voltage at the modulator input
Carrier Suppression
A performance metric quantifying the transmitter's ability to minimize LO leakage. Expressed as the power ratio between the intended modulated signal and the residual carrier tone, typically measured in dBc (decibels relative to carrier). High carrier suppression is critical for meeting spectral emission masks.
- Measurement: Ratio of carrier power to total signal power
- Typical Requirement: > 35 dBc for compliant transmission
- Degradation Factors: Temperature drift, aging, and supply voltage variation
I/Q Calibration
The systematic process of measuring and generating correction coefficients to compensate for static gain, phase, and offset errors in a quadrature modulator. LO leakage correction is a subset of this broader calibration, specifically targeting the DC offset component.
- Factory Calibration: Performed once with test equipment, coefficients stored in non-volatile memory
- Startup Calibration: Executed during device boot to account for temperature variations
- Online Tracking: Continuous background estimation using observation receiver feedback
Error Vector Magnitude (EVM)
A comprehensive modulation quality metric measuring the vector difference between the ideal reference constellation point and the actual transmitted signal point. LO leakage directly degrades EVM by shifting the entire constellation away from the origin, creating a systematic offset that increases the error vector for every symbol.
- Impact: LO leakage causes a fixed constellation shift, not random noise
- Measurement: Expressed as a percentage or in dB of the ideal symbol magnitude
- Relationship: EVM degradation is proportional to the square root of the carrier leakage power
Quadrature Modulator
The circuit that combines two orthogonal baseband signals with a local oscillator to generate a modulated RF carrier. This is the physical component where LO leakage originates. Imperfect isolation between the LO port and the RF output port allows the oscillator signal to leak directly to the output without being modulated.
- Leakage Paths: Substrate coupling, bond wire radiation, and package parasitics
- Design Mitigation: Differential circuit topologies and careful layout to minimize LO-to-RF feedthrough
- External Factors: Impedance mismatches at the RF port can reflect LO power back into the output

About the author
Prasad Kumkar
CEO & MD, Inference Systems
Prasad Kumkar is the CEO & MD of Inference Systems and writes about AI systems architecture, LLM infrastructure, model serving, evaluation, and production deployment. Over 5+ years, he has worked across computer vision models, L5 autonomous vehicle systems, and LLM research, with a focus on taking complex AI ideas into real-world engineering systems.
His work and writing cover AI systems, large language models, AI agents, multimodal systems, autonomous systems, inference optimization, RAG, evaluation, and production AI engineering.
Partnered with leading AI, data, and software stack.
How We Work
Custom AI workflows for your Business
One-fit-all AI don't work for modern businesses. At Inferensys, we aim to understand your business & custom requirements; which we use to define most efficient agentic workflows, the data, and the tools for your business.
01
Review the use case
We understand the task, the users, and where AI can actually help.
Read more02
Pick the right approach
We define what needs search, automation, or product integration.
Read more03
Build the first useful version
We implement the part that proves the value first.
Read more04
Improve from there
We add the checks and visibility needed to keep it useful.
Read moreThe first call is a practical review of your use case and the right next step.
Talk to Us