DPSK Symbol Detection Using Multi-Previous Phase Likelihoods
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Differential phase shift keying (DPSK) communication devices face challenges in accurately determining symbols due to phase errors, leading to error propagation, which can affect the reliability of soft-output values and reduce the accuracy of error detection in wireless communication systems.
Innovation Solution
A method is introduced to calculate phase differences between consecutive signals, using a first phase difference between a previous signal and a signal preceding it, and a second phase difference between the target signal and the previous signal, to determine the likelihood of each symbol and accurately determine the expected symbol or phase difference, thereby reducing error propagation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional DPSK demodulation is used to determine symbols based on phase difference between consecutive signals, then the system can operate with simple structure, but error propagation occurs when phase errors are present, reducing reliability
Solution Approach 1:
The patent segments the phase difference calculation into multiple independent components: calculating phase differences between the target signal and each of the M previous signals separately, then combining these segmented results through likelihood calculation. This segmentation allows error isolation and prevents error propagation while maintaining structural organization.
Solution Approach 2:
The patent transitions from conventional 2-signal phase difference measurement to M-signal multi-dimensional phase difference analysis. By incorporating M previous signals instead of just one, the system adds temporal dimensionality to the measurement, creating a more robust phase difference determination that accounts for multiple reference points and reduces the impact of single-point phase errors.
2Measurement precision
If phase difference calculation uses only the immediate previous signal, then the calculation is simple and fast, but accuracy decreases when phase errors occur in the previous signal
Solution Approach 1:
The patent performs preliminary phase difference calculations between the target signal and all M previous signals before final symbol determination. These preliminary calculations are stored as likelihood values, allowing the system to quickly reference pre-computed phase relationships during demodulation, thus improving accuracy without proportionally increasing real-time calculation burden.
Solution Approach 2:
The patent creates multiple copies of phase difference measurements by calculating the phase difference between the target signal and each of the M previous signals separately. These copied measurements are then combined through likelihood calculation to produce a more accurate result, effectively using redundancy to improve measurement precision while managing computational complexity.
3Reliability
If error-detecting code such as CRC is used to detect errors, then error detection capability is improved, but the soft-output reliability decreases when error propagation occurs
Solution Approach 1:
The patent implements beforehand cushioning by calculating likelihood values for multiple previous signals before making the final symbol determination. This creates a buffer of pre-computed phase difference information that cushions against the impact of phase errors in any single previous signal, preserving soft-output reliability even when errors are present in the communication channel.
Data Source
AI summary
An method of determining a symbol according to a phase difference between input signals input in order of time may include calculating a first phase difference between a phase of a first previous signal received prior to a target signal and a phase of a second previous signal received prior to the first previous signal; calculating a second phase difference between a phase of the target signal and the phase of the second previous signal; calculating target likelihoods based on the first phase difference and the second phase difference; and determining an expected phase difference between the target signal and the first previous signal or an expected symbol for the target signal, based on the target likelihoods.