Digital Self-Interference Cancellation: Algorithmic Performance Analysis and Software Defined Radio Based Realization

Nowadays, radio spectrum scarcity has become a pivotal problem in the field of telecommunication because of the unprecedented surge in the number of spectrum users. Full-Duplex (FD) radio is an emerging technology that allows wireless users to transmit and receive data simultaneously with the same frequency, which eventually improves the spectral efficiency and throughput of the system. However, the main limiting factor for achieving FD is the self-interference problem which occurs due to presence of the strong transmit signal as compared to the desired received signal at the receiving node. This paper presents the performance analysis of three adaptive algorithms, i.e., Least Mean Square, Normalized Least Mean Square, and Recursive Least Square for digital self-interference cancellation. In addition, the paper presents the first steps of a design and Software Defined Radio based implementation. The simulation results are evaluated for figures of merits such as amount of self-interference cancellation and mean square error. The so-called pseudo real-time implementation is carried out for two cases: simulations only with realistic parameters and with an RF cable. A maximum amount of 31 dB of digital self-interference cancellation is achieved by simulation, and up to 20 dB with a RF cable.