Parallel Demodulator Architecture for High Symbol Rate Timing Recovery

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current satellite communication systems face challenges in efficiently demodulating high symbol rate data streams due to limitations in clock speed requirements, which lead to increased design complexity and cost, particularly in maintaining accurate symbol timing recovery at high data rates.

Innovation Solution

A dynamic and flexible parallel architecture for symbol timing recovery that processes data streams at or below the modem's clock rate, preserving temporal dependencies between samples and reducing the need for high-frequency processing, allowing for efficient and robust broadband services in shared bandwidth networks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional serial demodulation architecture is used to process high symbol rate data streams, then accurate symbol timing recovery can be achieved, but the processing clock frequency must be at least twice the symbol rate, leading to increased design complexity and manufacturing cost

Engineering Contradiction:
Improvesymbol timing recovery accuracyVSAvoidprocessing clock frequency requirement
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the high symbol rate data stream processing into multiple parallel channels, each operating at a lower clock frequency. By segmenting the processing tasks across parallel paths with different timing offsets, the system achieves accurate symbol timing recovery without requiring a single high-frequency clock, thus resolving the contradiction between timing accuracy and device complexity

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from a single-dimensional serial processing approach to a multi-dimensional parallel processing architecture. By adding the dimension of parallel channels with staggered timing, the system can recover symbol timing accurately while operating at lower individual clock frequencies, effectively resolving the frequency requirement contradiction

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

2Device complexity

If parallel architecture is used to reduce processing clock frequency, then design complexity and cost are reduced, but maintaining accurate symbol timing recovery at high symbol rates becomes challenging

Engineering Contradiction:
Improveprocessing clock frequency requirementVSAvoidsymbol timing recovery accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent incorporates feedback mechanisms where timing error detection results from multiple parallel channels are combined and used to adjust the timing offsets of subsequent processing stages. This feedback loop ensures that accurate symbol timing recovery is maintained despite the lower individual clock frequencies in each parallel channel

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent merges the timing information from multiple parallel channels through a combination logic that integrates results from channels with different timing offsets. By combining these distributed measurements, the system achieves accurate symbol timing recovery that would be difficult to obtain in a single channel operating at lower frequency

Inventive Principle:
Principle #5Merging (Combining)

3Productivity

If high processing clock frequency is used to handle high symbol rate data streams, then throughput is maintained, but timing constraints on physical design become tighter and manufacturing cost increases

Engineering Contradiction:
Improvedata processing throughputVSAvoidtiming constraint complexity
Core Design Contradiction:
ProductivityVSEase of manufacture

Solution Approach 1:

The patent segments the high-throughput processing requirement into multiple parallel channels operating at relaxed timing constraints. Each channel processes a portion of the data stream at a manageable clock frequency, collectively achieving the required overall throughput while avoiding the manufacturing complexity associated with high-frequency design

Inventive Principle:
Principle #1Segmentation

4Measurement precision

If sampling rate is increased to at least twice the symbol rate for accurate demodulation, then symbol timing recovery is improved, but all front-end signal processing must be performed at this high rate, increasing system complexity

Engineering Contradiction:
Improvesymbol timing recovery accuracyVSAvoidsignal processing clock rate
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent implements dynamic timing offsets in the parallel channels, where each channel operates at a lower fixed clock rate but with adjustable phase relationships. This dynamic configuration allows the system to achieve the effective sampling rate requirement for accurate symbol timing while keeping individual processing stages at lower, less complex clock rates

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP2667529B1Method and apparatus for parallel demodulation of high symbol rate data streams in a communications system
Publication Date: 2018.06.20 HUGHES NETWORK SYST
  • EP2667529B1 patent drawingFigure 1A
  • EP2667529B1 patent drawingFigure 1B
  • EP2667529B1 patent drawingFigure 1C

AI summary

A dynamic and flexible architecture and methods for demodulation of high data-rate streams with high symbol-rates, such as in satellite communications systems or computer network communications systems, is provided. A data stream of a data transmission is received, the data stream corresponding to a plurality of data symbols. A plurality of data samples corresponding to each of the data symbols is generated. Further, one or more representative data samples, corresponding each of the data symbols, are generated based at least in part on timing control signals and the generated data samples for the respective data symbol. The generated data samples corresponding to each of the data symbols other than the representative data samples are dropped. The timing control signals are then adjusted based at least in part on the generated representative data samples.