Superimposed signaling inspired channel estimation in full-duplex systems

Area of Science:

• Electrical Engineering and Computer Science
• Wireless Communication Systems
• Signal Processing

Background:

• Residual self-interference (SI) cancellation is a critical challenge in enabling full-duplex (FD) communication systems.
• Existing methods often struggle with accurate SI and channel estimation, limiting FD system performance.
• Digital baseband processing is a key stage where SI mitigation is essential.

Purpose of the Study:

• To propose a novel technique for estimating self-interference (SI) and communication channels in full-duplex (FD) systems.
• To develop a bandwidth-efficient channel estimation method using asymmetric shifted modulation constellations.
• To provide a framework for determining optimal parameters for accurate channel estimation.

Main Methods:

• Introduced asymmetric shifted modulation constellations by adding a constant real number to conventional constellation points.
• Developed an Expectation-Maximization (EM) estimator tailored for the proposed asymmetric constellations.
• Derived a closed-form lower bound for the Mean Square Error (MSE) of channel estimation.

Main Results:

• The proposed technique enables bandwidth-efficient and unambiguous channel estimation.
• The Expectation-Maximization (EM) estimator achieves lower MSE and bit error rates compared to data-aided methods.
• The method demonstrates robustness against increasing SI signal power.

Conclusions:

• The asymmetric shifted modulation constellation technique offers a superior approach to SI and channel estimation in FD systems.
• The derived MSE bound facilitates the calculation of minimum shift energy for accurate estimation.
• This method significantly enhances the performance and reliability of full-duplex communication.