🔢 mathematics

An Application of Complex Fuzzy Soft Matrices in Signal Processing

This paper introduces complex fuzzy soft matrices and demonstrates their application in signal processing by combining them with the Fourier transform via a cross product to develop an algorithm that effectively identifies reference signals from detected data, showing superior performance compared to other methods.

Original authors: Olayemi R. Oladokun, Taiwo O. Sangodapo

Published 2026-03-19

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Olayemi R. Oladokun, Taiwo O. Sangodapo

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: Finding a Needle in a Haystack (But the Haystack is Foggy)

Imagine you are a radio operator trying to find a specific song playing on the radio. But there's a catch: the radio station is full of static, the signal is fuzzy, and there are hundreds of other songs playing at the same time. In the real world, signals aren't always "on" or "off," "clear" or "garbage." They are often a mix of both.

This paper proposes a new, super-smart way to sort through that noise to find the exact song (or reference signal) you are looking for. It does this by combining two powerful tools: Complex Fuzzy Soft Matrices and the Fourier Transform.

Here is how it works, step-by-step:

1. The Problem: The World Isn't Black and White

Traditional math is like a light switch: it's either ON or OFF.

Is this house expensive? Yes or No.
Is this signal clear? Yes or No.

But real life is more like a dimmer switch. A house might be "sort of expensive." A signal might be "mostly clear but a little fuzzy."

The authors use Fuzzy Logic to handle this "maybe" territory. They take it a step further with Complex Fuzzy Sets. Think of this as adding a "phase" or a "rhythm" to the fuzziness. It's not just how much of the signal is there (amplitude), but when it happens (phase). This is crucial for radio waves and sound.

2. The Tool: The "Fuzzy Spreadsheet" (Matrices)

To manage all this messy data, the authors use something called a Complex Fuzzy Soft Matrix.

The Analogy:
Imagine you are a real estate agent trying to find the perfect house for a client.

The Clients (Parameters): You have a list of wants: "Cheap," "Green Surroundings," "Modern."
The Houses (Universe): You have a list of houses: House A, House B, House C.

In a normal spreadsheet, you'd put a "1" if a house fits the criteria and a "0" if it doesn't.
In this paper's Fuzzy Matrix, you put a number between 0 and 1.

House A is 0.8 "Modern."
House B is 0.2 "Modern."

Now, imagine this spreadsheet is made of complex numbers (numbers with a real part and an imaginary part). This allows the spreadsheet to hold not just the strength of the signal, but also its timing (phase). It's a 3D spreadsheet that captures the full "vibe" of the signal, not just its volume.

3. The Process: The "Cross-Product" Dance

The paper describes an algorithm to find the "Reference Signal" (the target) among many "Interest Signals" (the noise).

The Analogy: The Dance-Off
Imagine you have a Reference Signal (let's call him "The Star Dancer"). You have a crowd of other dancers (the signals detected by the receiver). You want to find out which one of the crowd is actually the Star Dancer in disguise.

Sampling: You take a snapshot of the Star Dancer and every other dancer at the same time intervals (like taking 4 photos per second).
The Matrix: You organize these snapshots into two giant grids (Matrices). One grid is the Star, the other is the Crowd.
The Cross-Product (The Dance): You make the Star Dancer "dance" with every member of the crowd. In math terms, this is a Cross Product.
- If the crowd member moves exactly like the Star, the dance is perfect, and the score is high.
- If they move differently, the dance is clumsy, and the score is low.
The Scoreboard: You calculate a "similarity score" for every dancer. The paper uses a "Max-Min" rule: it looks for the dancer who consistently matches the Star the best, ignoring the worst moments.

4. The Secret Weapon: The Fourier Transform

The paper compares two ways of doing this: using standard matrix math vs. using the Fourier Transform.

The Analogy: The Prism

Standard Math: Looking at a signal is like looking at a white light bulb. You see the total brightness, but you can't tell what colors are inside.
Fourier Transform: This is like putting a prism in front of the light bulb. It splits the white light into a rainbow (Red, Orange, Yellow, etc.).

In signal processing, the Fourier Transform breaks a complex wave down into its individual frequencies (the "colors" of the signal).

The Result:
The authors found that using the Fourier Transform was like using a high-definition prism. It gave a much clearer, higher "optimal value."

Without the Prism: You might think two signals are similar because they are both "loud."
With the Prism: You realize one signal is "loud red" and the other is "loud blue." They aren't the same!

Because the Fourier method separates the signal into its true components, it identified the correct reference signal much more accurately than the standard method.

Summary: What Did They Achieve?

The authors built a new mathematical "filter" (Complex Fuzzy Soft Matrices) that can handle messy, uncertain, and rhythmic signals. They tested it by trying to find a specific signal hidden in a bunch of noise.

They discovered that by combining this fuzzy filter with the Fourier Transform (the prism that breaks signals into colors), they could identify the correct signal with much higher confidence.

In plain English: They invented a smarter way to tune a radio in a stormy weather, proving that if you look at the signal through a "prism" (Fourier Transform) rather than just squinting at the static, you can find the music you're looking for much faster and more accurately.

1. Problem Statement

Traditional signal processing and modeling tools often rely on crisp, deterministic, and precise logic (binary "yes/no" states). However, real-world signal environments are inherently uncertain, imprecise, and noisy. Classical models struggle to represent the "more-or-less" nature of signal attributes (such as amplitude, phase, and noise levels) or the ambiguity in identifying a specific reference signal among a large set of detected signals.

The paper addresses the challenge of identifying a specific reference signal from a multitude of signals detected by a digital receiver in an uncertain environment. The goal is to develop a mathematical framework that can handle the fuzziness and complexity of signal data more effectively than traditional methods.

2. Methodology

The authors propose a hybrid methodology that integrates Complex Fuzzy Soft Set (CFSS) theory with Fourier Transform techniques. The approach involves the following steps:

A. Theoretical Framework: Complex Fuzzy Soft Matrices (CFSM)

Complex Fuzzy Sets (CFS): Unlike standard fuzzy sets where membership is a real number in $[0, 1]$ , CFS extends the range to the unit circle in the complex plane. Elements are represented as $r e^{i\omega}$ , where $r$ is the amplitude (degree of membership) and $\omega$ is the phase term. This allows for the simultaneous representation of magnitude and phase information.
Soft Sets: Used to parameterize the data, allowing for flexible modeling of attributes (e.g., "near to downtown," "cheap").
Matrix Representation: The authors define Complex Fuzzy Soft Matrices to store and manipulate these sets computationally. Operations such as Union, Intersection, and Complement are defined using max/min operators on the absolute values of the complex elements.

B. Algorithm for Signal Identification

The paper presents two distinct algorithmic approaches for identifying a reference signal $R$ among detected signals $S_i$ :

Approach 1: Direct Matrix Cross-Product (Section 5)

Sampling: Received signals $S_1, \dots, S_m$ are sampled $N$ times.
Matrix Construction: Two matrices, $A$ and $B$ , are constructed. Matrix $A$ contains the absolute values of the sampled signals, and Matrix $B$ contains the sampled reference signal (or a second set of signals for comparison).
Multiplication: A standard matrix product $A \times B$ is computed.
Decision Making: A "Max-Min" decision matrix is derived by finding the minimum value in each column of the product matrix.
Selection: The signal corresponding to the maximum value in the resulting decision vector is identified as the reference signal.

Approach 2: Fourier Transform Integration (Section 6)
This approach is highlighted as the superior method in the paper.

Discrete Fourier Transform (DFT): Signals are transformed into the frequency domain. The amplitude term $X[k]$ is restricted to $[0, 1]$ to fit the fuzzy set definition, while the phase term represents the frequency component.
Inverse DFT (IDFT): Signals are reconstructed using complex fuzzy set logic, treating the amplitude as the fuzzy membership and the phase as the complex argument.
Cross-Product Calculation: The algorithm calculates the cross-product between the samples of the unknown signal $x_i(n)$ $x_{i} (n)$ and the reference signal $r(n)$ $r (n)$ .
- The operation involves taking the minimum of the amplitude terms and the minimum of the phase terms for corresponding components.
- The absolute value of the resulting complex cross-product terms is calculated.
Optimization: For each signal sample, the maximum absolute value of the cross-product is determined.
Comparison: The maximum values for all candidate signals are compared. The signal yielding the highest optimal value is identified as the reference signal.

3. Key Contributions

Novel Application: The paper pioneers the application of Complex Fuzzy Soft Matrices specifically to signal processing and system identification, bridging the gap between soft computing theory and engineering signal analysis.
Hybrid Algorithm: It introduces a specific algorithm that combines the parameterization of soft sets with the frequency-domain analysis of Fourier transforms to handle signal uncertainty.
Complex Representation: By utilizing the complex plane (amplitude + phase), the model captures more signal characteristics than traditional real-valued fuzzy matrices, which only consider magnitude.
Comparative Analysis: The authors provide a comparative demonstration showing that the Fourier Transform-based approach yields higher optimal values and more accurate identification than the direct matrix cross-product method.

4. Results

The authors validated their methodology through numerical examples:

Example 1 (Matrix Approach): Using a set of four signals sampled four times, the algorithm successfully identified a reference signal based on the max-min decision matrix.
Example 2 (Fourier Approach): Two electromagnetic signals ( $x_1$ $x_{1}$ and $x_2$ $x_{2}$ ) were sampled twice and compared against a reference signal $r$ $r$ .
- Signal $x_1$ yielded a maximum cross-product value of 0.18.
- Signal $x_2$ yielded a maximum cross-product value of 0.23.
- Conclusion: Since $0.23 > 0.18$ , the algorithm correctly identified $x_2$ as the signal with the highest resemblance to the reference signal.
Performance: The results confirmed that the Fourier Transform method provided a "higher optimal value," leading to a more definitive identification of the reference signal $R$ .

5. Significance

Handling Uncertainty: The proposed model offers a robust mathematical tool for signal processing in noisy or ambiguous environments where binary logic fails. It effectively models the "degree of resemblance" rather than a simple match/mismatch.
Phase Information: By incorporating the phase term (via complex numbers), the method accounts for timing and frequency shifts that are critical in signal processing but often ignored in standard fuzzy logic.
Decision Support: The algorithm provides a clear, quantitative metric (the optimal value) for decision-making in digital receivers, aiding in the selection of the most relevant signal from a large dataset.
Future Potential: The paper suggests that this framework can be extended to continuous Fourier transforms for continuous noise signals, offering a scalable solution for advanced communication systems and medical signal analysis.

In summary, the paper demonstrates that Complex Fuzzy Soft Matrices, when coupled with Fourier Transforms, offer a superior framework for signal identification, providing higher accuracy and better handling of uncertainty compared to traditional crisp or standard fuzzy methods.