OFDM Signal Identification via Kurtosis Periodicity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for blind identification of OFDM signals are inefficient when the cyclic prefix is small or absent, and require time and frequency synchronization, making them unsuitable for real-time implementation in opportunistic telecommunication systems.
Innovation Solution
A method that samples the received signal, divides it into blocks, and applies Fourier transforms to determine kurtosis periodicity in the frequency domain, allowing for the identification of OFDM signals without relying on cyclic prefixes or synchronization, and determines the presence and parameters of the signal based on spectral peaks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If known methods exploit statistical properties induced by cyclic prefix for OFDM signal identification, then identification can be performed, but the method becomes inefficient when cyclic prefix length is small compared to useful duration
Solution Approach 1:
The patent changes the identification parameter from cyclic prefix-based statistical properties to temporal periodicity of the autocorrelation function. This allows the method to work effectively regardless of cyclic prefix length, as temporal periodicity remains a fundamental property of OFDM signals even when cyclic prefix is small or absent.
Solution Approach 2:
Instead of relying on the cyclic prefix structure to create identifiable statistical properties, the patent inverts the approach by exploiting the temporal periodicity inherent in the OFDM symbol structure itself. This reversal makes the identification method independent of cyclic prefix characteristics.
2Measurement precision
If methods require time and frequency synchronization for OFDM signal identification, then identification accuracy can be improved, but real-time implementation becomes difficult
Solution Approach 1:
The patent enables the identification method to be self-sufficient by using temporal periodicity detection that does not require external synchronization information. The autocorrelation function's periodicity naturally reveals OFDM signal characteristics without needing time or frequency synchronization, making the system self-service capable.
Solution Approach 2:
The method performs preliminary autocorrelation analysis on the received signal to detect temporal periodicity before any synchronization operations would be needed. This preliminary action provides identification capability that works independently of subsequent synchronization requirements.
3Adaptability or versatility
If methods are designed to be independent of cyclic prefix presence, then adaptability improves, but robustness to noise and multipath may deteriorate
Solution Approach 1:
The patent extracts the temporal periodicity property from the autocorrelation function, which is a fundamental characteristic of OFDM signals. By focusing on this extracted property rather than cyclic prefix-specific features, the method achieves both independence from cyclic prefix and robustness to noise and multipath conditions.
Solution Approach 2:
The method exploits the periodic nature of the autocorrelation function of OFDM signals. This periodic action in the temporal domain provides a reliable identification mechanism that is inherently robust to noise and multipath effects, while being independent of cyclic prefix characteristics.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This method provides robust and noise-resistant identification of OFDM signals, independent of cyclic prefix length and synchronization, enabling effective adaptation in opportunistic telecommunication systems.
Implementation Method 1
each block (B l) then being subjected to a Fourier transform to obtain a plurality of symbols (â l (f)) at a plurality of frequencies
Implementation Method 2
for each frequency of said plurality, the kurtosis (κ(f)) of a set of symbols (â l (f),l = 1,..., L) thus obtained at this frequency, then it is determined whether said kurtosis has a periodicity
Data Source
Figure 1~2
Figure 3
Figure 4A~4B
AI summary
The present invention relates to a method for identifying an OFDM signal in a received signal, according to which the received signal is sampled and cut into blocks of P samples, each block (I) being subjected to an inverse Fourier transform in order to obtain a plurality of symbols (II) with a plurality of frequencies. For every frequency of said plurality of frequencies, the kurtosis (III) of a set of symbols (IV) obtained at said frequency is calculated, after which it is determined whether the kurtosis has a periodicity in the frequency domain using kurtosis values calculated for the frequencies of said plurality of frequencies, and the presence of an OFDM signal is identified in the received signal if the kurtosis has such a periodicity.