Adaptive Body Biasing (ABB): A Technique for Minimizing Power in VLSI Adaptive Body Biasing (ABB) is a technique used in low-power VLSI design to reduce the...
Adaptive Body Biasing (ABB) is a technique used in low-power VLSI design to reduce the static power consumption of circuits by adjusting their body size and geometry dynamically. This is achieved through a feedback loop that monitors the operating temperature and adjusts the device dimensions accordingly.
How ABB works:
Body size optimization: Initially, the body size of a transistor is set to a fixed value, typically equal to the minimum feasible size.
Temperature measurement: Once the transistor is operating, its temperature is measured. This measurement is used to calculate the actual body size required to achieve a desired power consumption.
Dynamic scaling: Based on the measured temperature, the body size is dynamically scaled up or down. This can be achieved by:
Increasing the body size for smaller temperatures.
Decreasing the body size for larger temperatures.
Benefits of ABB:
Significant power reduction: ABB can achieve power reductions of up to 85%, compared to traditional static power optimization techniques like sizing for minimum power.
Improved power-performance trade-off: By optimizing the body size dynamically, the designer can achieve the desired power consumption while maintaining desired performance levels.
Reduced gate count: By optimizing the body size, ABB can be implemented with fewer transistors, resulting in smaller chip size and lower fabrication costs.
Example:
To illustrate the principle of ABB, consider a transistor with a minimum body size (e.g., 50 nm) and a maximum body size (e.g., 100 nm). If the transistor is operated at a temperature of 100 °C, the calculated optimal body size would be 70 nm. This means that the body size is dynamically scaled up to 70 nm to achieve the desired power consumption