Programmable Gain Array Feedback for WLAN DC-Offset Cancellation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In zero IF receivers for WLAN applications, the large capacitive coupling and cascaded gain lead to dc-offset issues that can saturate the programmable gain array (PGA), requiring a large chip area and complex circuitry to manage the long dc-offset transients, especially in IEEE 802.11a and 802.11g applications where gain settling time is critical.
Innovation Solution
The implementation of a feedback circuit within the PGA that includes a transconductance amplifier, a transimpedance amplifier, and a voltage amplifier forms a main amplifier stage, with a feedback loop that senses imbalances and integrates a correction signal to be negatively fed back, effectively shifting the high-pass pole to lower or higher frequencies and eliminating dc-offsets through an inside-opamp dc-offset canceller (DOC) technique.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a large time constant is used to prevent deep signal damage, then signal integrity is improved, but chip area increases and dc-offset transient time increases
Solution Approach 1:
The amplifier is divided into multiple stages (first amplifier stage, second amplifier stage, third amplifier stage) with different gain values and bandwidth characteristics. The first stage provides high gain with large bandwidth, the second stage provides medium gain with medium bandwidth, and the third stage provides low gain with small bandwidth. This segmentation allows the system to achieve the required overall gain while maintaining signal integrity without requiring a single large time constant that would increase chip area.
Solution Approach 2:
The patent implements dynamic gain control by selectively enabling or disabling amplifier stages based on signal conditions. The gain control circuit can activate different combinations of amplifier stages to provide the required gain while optimizing bandwidth and settling time. This dynamic approach allows the system to adapt to different signal levels and maintain performance without requiring maximum gain settings that would increase chip area and transient time.
2Power
If large cascaded gain is used to achieve dynamic range, then signal amplification is improved, but dc-offset saturation occurs
Solution Approach 1:
The total gain of 0 to 50 dB is distributed across three amplifier stages with different gain ranges. The first amplifier stage provides 0 to 20 dB gain, the second stage provides 0 to 20 dB gain, and the third stage provides 0 to 10 dB gain. This segmentation prevents any single stage from having excessive gain that would amplify dc-offset to saturation levels, while the combined gain of all stages achieves the required dynamic range.
Solution Approach 2:
The patent implements a feedback circuit that senses the output of the amplifier stages and adjusts the gain control signals to prevent dc-offset saturation. The feedback mechanism monitors the output signal level and dynamically adjusts the gain of individual stages to maintain linear operation, preventing the accumulation of dc-offset that would lead to saturation in high-gain configurations.
3Power
If high gain is used in baseband, then dynamic range is improved, but gain settling time increases
Solution Approach 1:
The baseband amplification is divided into three stages with progressively lower gain values. The first stage provides the highest gain with the fastest settling time, the second stage provides medium gain with medium settling time, and the third stage provides the lowest gain with the slowest settling time. The overall settling time is dominated by the first stage which has the fastest response, thereby reducing the total gain settling time while maintaining the required dynamic range through the combined gain of all stages.
Solution Approach 2:
The gain control circuit dynamically selects which amplifier stages are active based on the required gain level and time constraints. For applications requiring fast settling, the system can use only the first amplifier stage with its fast response. For applications requiring maximum dynamic range and where settling time is less critical, additional stages can be enabled. This dynamic configuration allows optimization of the trade-off between gain and settling time based on real-time requirements.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This solution reduces chip area requirements, shortens receiver settling time during dc-offset transients, and extends bandwidth by shifting high-pass poles, thereby improving the dynamic range and reducing noise and non-linearity in the PGA.
Implementation Method 1
A feedback circuit is connected between an output of the voltage amplifier and the low-impedance node between the transconductance amplifier and the transimpedance amplifier. The transconductance amplifier, the transimpedance amplifier, and the voltage amplifier form a main amplifier stage. The feedback circuit senses an imbalance in an output of the main amplifier stage, whereby a correction signal is integrated and negatively fed back to the low-impedance node
Data Source
AI summary
An amplifier circuit includes a transconductance amplifier at an input side of the amplifier circuit, a transimpedance amplifier connected to an output of the transconductance amplifier, and a voltage amplifier connected to an output of the transimpedance amplifier. The transconductance amplifier and the transimpedance amplifier form a low-impedance node at an interface thereof. A feedback circuit is connected between an output of the voltage amplifier and the low-impedance node between the transconductance amplifier and the transimpedance amplifier. The transconductance amplifier, the transimpedance amplifier, and the voltage amplifier form a main amplifier stage. The feedback circuit senses an imbalance in an output of the main amplifier stage, whereby a correction signal is integrated and negatively fed back to the low-impedance node between the transconductance amplifier and the transimpedance amplifier.


