MSK Demodulation Using Matched Filters and Viterbi Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current wireless communication technologies face challenges in achieving high-speed data transfer rates, particularly with IEEE 802.15.4 compatible devices, as existing demodulation methods are not sufficient to support data rates beyond 250 kb/s without advanced digital signal processing techniques, especially in analog radio constraints.
Innovation Solution
A method for digitally demodulating a minimum-shift keying (MSK) signal using a three-symbol sliding window to rate all possible combinations of modulating symbols through a bank of matched filters, constructing a 4-state trellis, and driving a Viterbi decoder to recover data, which improves upon simple non-coherent demodulation techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If advanced digital signal processing techniques are used to achieve high data transfer rates (2 Mb/s), then data transfer rate is improved, but device complexity increases
Solution Approach 1:
The received signal is processed in segments of three consecutive symbols at a time using a sliding window approach. Each segment is independently evaluated against all eight possible symbol combinations through matched filters, breaking down the complex continuous demodulation problem into discrete manageable segments.
Solution Approach 2:
All eight possible combinations of three modulating symbols are pre-calculated and stored as reference waveforms. During demodulation, these pre-computed references are directly compared with the received signal segment, eliminating the need for real-time complex calculations and enabling fast decision making.
2Reliability
If coherent demodulation with matched filters and Viterbi decoding is implemented, then demodulation performance and robustness are improved, but power consumption increases
Solution Approach 1:
The algorithm processes three symbols at a time rather than requiring continuous processing of the entire signal stream. This partial action approach with a sliding window allows the system to achieve near-optimal performance (6 dB gain) while reducing computational burden and power consumption compared to full-sequence coherent demodulation.
Solution Approach 2:
The system uses simple 1-bit ADCs instead of complex multi-bit ADCs, and employs a simplified matched filter bank with pre-computed references. These low-cost components are replaced or reconfigured for each processing segment, achieving high performance with minimal power consumption suitable for battery-operated devices.
3Device complexity
If simple non-coherent demodulation is used to reduce complexity and power consumption, then device complexity and power consumption are reduced, but demodulation performance deteriorates
Solution Approach 1:
The patent introduces an intermediate processing stage between simple non-coherent detection and full coherent demodulation. A sliding window with matched filters serves as an intermediary that processes three-symbol segments with pre-computed references, achieving 6 dB performance gain over non-coherent methods while maintaining relatively simple hardware suitable for low-power applications.
Data Source
AI summary
A two stage process is applied for recovering the modulating content from the received I-Q waveforms of a MSK modulated signal. In the first stage, at each incoming symbol the I-Q waveform segments of the input belonging to the three most recently received symbols are used in hypothesis testing. A matched filter bank produces ratings for each of the possible three symbol modulating patterns in proportion to the likelihood that the combination in question may have produced the current but by now impaired input segment. While the three symbol window slides symbol-by-symbol over the input the successive hypothesis tests are not independent as each symbol is involved in three consecutive tests. The dependence thus created lays the foundation and provides the branch metrics for applying the Viterbi algorithm for the determination of the modulating symbols in the second stage.