Noise-Shaping Sample Rate Conversion Without Multipliers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional sample rate conversion systems require high computational complexity and power consumption due to multiplication operations, which limits their efficiency in achieving low latency and high performance, especially in audio processing applications.
Innovation Solution
The implementation of a noise shaping technique using a delta-sigma modulator for sample rate conversion, which suppresses quantization noise and eliminates the need for multipliers, allowing for asynchronous and synchronous conversions with reduced complexity and power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If conventional polyphase filters with multipliers and adders are used for sample rate conversion, then the desired sample frequencies are achieved, but computational complexity and power consumption increase significantly
Solution Approach 1:
The patent changes the fundamental parameter of the conversion method from multiplication-based polyphase filtering to noise shaping modulation. By using a delta-sigma modulator that operates on the principle of quantization noise shaping, the system achieves sample rate conversion through a different physical mechanism that requires no multipliers, thereby reducing computational complexity while maintaining conversion speed
Solution Approach 2:
The patent replaces the mechanical/computational system of multipliers and adders with an electronic modulation system. The noise shaping circuit uses feedback loops and quantization mechanisms to achieve the same sample rate conversion function without requiring complex arithmetic operations, substituting a simpler electronic process for a complex computational one
2Speed
If conventional polyphase filters with multipliers and adders are used for sample rate conversion, then the desired sample frequencies are achieved, but power consumption increases significantly
Solution Approach 1:
The patent fundamentally changes the operational parameters from high-power multiplication operations to low-power modulation operations. The noise shaping approach uses simple comparators and feedback mechanisms that consume minimal power compared to the multipliers and adders required by conventional polyphase filters, while maintaining the same sample rate conversion functionality
Solution Approach 2:
The patent extracts and eliminates the power-consuming multiplier operations from the sample rate conversion process. By using noise shaping modulation, the system achieves the same conversion function without requiring the complex arithmetic units that are the primary source of power consumption in conventional approaches
3Device complexity
If noise shaping technique with delta-sigma modulator is used, then computational complexity and power consumption are reduced, but quantization noise must be suppressed
Solution Approach 1:
The patent converts the harmful quantization noise into a beneficial tool for sample rate conversion. By deliberately shaping the quantization noise spectrum through feedback mechanisms, the system pushes noise energy to frequency bands that can be easily filtered, thereby transforming what would be a harmful artifact into an acceptable byproduct of the conversion process
Solution Approach 2:
The patent employs feedback loops within the noise shaping circuit to continuously monitor and correct quantization noise. The modulator uses feedback to shape the noise spectrum, directing quantization errors to higher frequency bands where they can be removed by simple low-pass filtering, thereby maintaining signal quality while reducing complexity
4Use of energy by stationary object
If noise shaping technique with delta-sigma modulator is used, then power consumption is reduced by eliminating multipliers, but distortion and aliased mirror images must be rejected
Solution Approach 1:
The patent performs preliminary noise shaping and spectrum management before the final decimation stage. By pre-shaping the quantization noise and managing spectral content in advance, the system prevents aliasing artifacts from being generated in the first place, rather than having to deal with them after they occur
Data Source
AI summary
Systems and methods for low power sample rate conversion are based on a noise shaping technique. A sample rate conversion circuit includes a clock synchronization circuit configured to receive an input sample sequence at a first sample rate and generate a valid sample sequence that is sampled at a second sample rate different from the first sample rate. The valid sample sequence may include valid samples from a registered sequence sampled at an oversampled rate greater than the first sample rate with invalid samples in the registered sequence being excluded from the valid sample sequence. The sample rate conversion circuit also includes a noise shaping circuit coupled to the clock synchronization circuit and configured to encode the valid sample sequence into a noise-shaped output sequence at the second sample rate by suppressing quantization noise from the valid sample sequence.