Parallel Sigma-Delta Modulator Architecture for Shorter Critical Paths

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current sigma-delta modulators face challenges in processing high-resolution data at high sampling rates due to long critical path delays and complexity in digital circuitry, which makes it difficult to implement efficient digital logic, especially at clock frequencies of a gigahertz.

Innovation Solution

A parallel digital-to-digital sigma-delta modulator architecture with multiple stages, each having an input and output bus, a processing circuit, and a feedback distribution bus to distribute error words, allowing for parallel processing and reduced clock rates, thereby minimizing critical path delays and simplifying digital circuit design.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If full resolution data is processed at high sampling rates in a single SDM stage, then data conversion speed is improved, but critical path delays become prohibitively long

Engineering Contradiction:
Improvedata conversion speedVSAvoidcritical path delays
Core Design Contradiction:
SpeedVSLoss of time

Solution Approach 1:

The patent divides the single high-speed processing stage into multiple parallel SDM stages operating at lower clock rates. Each stage processes a portion of the data in parallel, breaking the long critical path into shorter segments that can be completed within acceptable time frames while maintaining overall high data conversion throughput.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from a single-stage serial processing architecture to a multi-stage parallel architecture, adding the dimension of parallelism. This allows multiple data paths to operate simultaneously, effectively increasing the processing capacity without proportionally increasing the critical path delay of any single path.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Speed

If fast adder architectures are used to reduce delays, then processing speed is improved, but digital circuit complexity becomes prohibitive

Engineering Contradiction:
Improveprocessing speedVSAvoiddigital circuit complexity
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

By segmenting the processing into multiple stages with simpler adder architectures, the patent avoids the need for extremely complex fast adders in a single stage. Each stage uses moderate-speed adders that are easier to design and implement, while the overall system achieves high speed through parallel operation of multiple stages.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent combines multiple parallel processing paths with moderate complexity into a unified multi-stage architecture. This merging allows the system to achieve the processing power of complex fast adders while using simpler, more manufacturable circuit elements in each individual stage.

Inventive Principle:
Principle #5Merging (Combining)

3Loss of time

If parallel processing is implemented, then critical path delays are reduced, but feedback distribution complexity increases

Engineering Contradiction:
Improvecritical path delaysVSAvoidfeedback distribution complexity
Core Design Contradiction:
Loss of timeVSDevice complexity

Solution Approach 1:

The feedback distribution bus is designed as a universal multi-functional structure that serves all parallel SDM stages. This single feedback path handles error word distribution across multiple stages, reducing the need for separate feedback paths for each stage and thereby managing complexity while maintaining parallel processing benefits.

Inventive Principle:
Principle #6Universality (Multi-functionality)

4Manufacturing precision

If regular tiled layout is used for analog circuitry, then manufacturing precision is improved, but digital logic implementation becomes infeasible at gigahertz frequencies

Engineering Contradiction:
Improvelayout precisionVSAvoiddigital logic implementation
Core Design Contradiction:
Manufacturing precisionVSEase of manufacture

Solution Approach 1:

The patent segments the digital logic into multiple manageable stages that can be implemented using regular tiled layout techniques. By breaking down the complex digital logic into smaller, modular stages, the design becomes amenable to standard manufacturing processes while still achieving gigahertz-frequency operation through parallel processing.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS8847803B2Parallel multibit digital-to-digital sigma-delta modulation
Publication Date: 2014.09.30 MEDIATEK INC
  • US8847803B2 patent drawing
  • US8847803B2 patent drawing
  • US8847803B2 patent drawing

AI summary

Digital input words are received in parallel by a parallel digital-to-digital sigma-delta modulator. Error words corresponding to quantization error are added in parallel to the input words to form encoded data words. The encoded data words are quantized into parallel output words and the error words resulting from such quantization are distributed across parallel modulator stages to effect a predetermined quantization error spectral distribution. The quantized output words are output in parallel.