OFDM Sampling Interval Alignment for Low-Complexity CS Demodulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional compressive sensing algorithms for OFDM signals in wireless communication are complex, particularly in reconstructing information symbols, due to their high implementation complexity, which is proportional to the number of signal items (N) in the NP-hard minimum l0 norm optimization problem.
Innovation Solution
The method involves controlling the sampling interval of the OFDM baseband signal to be integral multiples of the length of a periodic square signal's minimum timeslot, allowing for efficient demodulation using compressive sensing by aligning non-zero frequency-domain sparse vector items with subcarriers, thereby reducing algorithm complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional compressive sensing algorithms are used for OFDM signal demodulation, then signal reconstruction is achieved, but the algorithm complexity becomes very high (proportional to N³ for minimum l0 norm method)
Solution Approach 1:
The patent changes the sampling interval parameter to be an integer multiple of the cyclic prefix length. This parameter modification enables the measurement matrix to have a special structure where non-zero elements align with subcarrier positions, transforming the complex N³ optimization problem into a simpler N-proportional problem while maintaining reconstruction accuracy
Solution Approach 2:
The patent segments the frequency domain into discrete subcarriers and aligns the compressed sensing measurement structure with these segments. By ensuring the sampling interval is an integer multiple of the cyclic prefix, the non-zero elements of the transformed signal are segmented to exactly match subcarrier positions, enabling efficient recovery
2Reliability
If the minimum l0 norm optimization problem is solved exhaustively, then the original signal can be reconstructed, but the computational burden becomes impossible to handle
Solution Approach 1:
The patent performs preliminary design of the sampling interval before the actual compression and reconstruction process. By pre-setting the sampling interval to be an integer multiple of the cyclic prefix length, the system prepares the measurement matrix structure in advance to enable efficient reconstruction without exhaustive search during operation
Solution Approach 2:
The patent replaces the expensive exhaustive optimization approach with a cheap, efficient alternative. By exploiting the special structure created by the integer-multiple sampling interval, the system uses simple correlation or matching operations instead of complex optimization algorithms, achieving the same reconstruction goal with minimal computational cost
Data Source
AI summary
An OFDM signal modulation method based on compressed sensing is provided, the method includes: acquiring a demodulation parameter of a receiver, where the demodulation parameter includes Tnyq of a periodic square signal of the receiver in sampling, Tnyq is a length of a minimum timeslot during which the periodic square signal maintains a high level or a low level; performing an OFDM modulation on an inputted information symbol to generate an OFDM baseband signal, and controlling a sampling interval TsampleOFDM=Ts/k of the OFDM baseband signal to be integral multiple of Tnyq, where Ts is a duration of valid data in the OFDM baseband signal, and k is the number of sub-carriers in the OFDM baseband signal; and up-converting the OFDM baseband signal and transmitting the processed signal to the receiver. Also provided are a demodulation method, device and system.