Hands Free Communication Improvement in Airplane by a New Dual RNQ Adaptive Algorithm

In hands-free communications in airplane, speech signals are strongly corrupted by different types of noise that are produced by vibrations, frictions, air flow, etc. In this paper, we propose a new dual recursive non quadratic (DRNQ) adaptive filtering algorithm combined with the forward blind source separation for speech enhancement in an airplane context. The proposed DRNQ algorithm aims to enhance the communication in the cockpit of the airplane and allow making the communication from and to the airplane as intelligible as possible. The proposed DRNQ algorithm has been intensively tested with different types of noises recorded in real situation. The obtained results in terms of several objective criteria like the segmental signal to noise ratio, the system mismatch and the cepstral distance have proven the efficiency of the proposed DRNQ algorithm in such application in compassion with the classical dual forward normalized least mean square (DFNLMS) algorithm.