This paper describes the elimination of feedback signals to implement a stable on-channel repeater regardless of input signals in the frequency bandwidth desired by the user. Toward this objective, the spread spectrum technique is applied to an adaptive filter, in which the least mean square algorithm removes the frequency characteristics of input signals. OCRs have been commercialized using adaptive filters based on the LMS algorithm in existing mobile communication services, including in GSM, CDMA, WCDMA, and LTE. However, in current mobile communication networks, various services with different signal characteristics can become mixed in the user’s frequency band, and bands with no signals may also exist. If the integrated bandwidth and the occupied bandwidth do not coincide, a bandpass filter cannot completely suppress other bands. Because the LMS algorithm is affected by the frequency characteristics of input signals as well as the feedback path of the object to be removed, feedback signals cannot be canceled for the entire band. Therefore, in this paper, the SS technique is applied to the existing LMS method to remove the frequency characteristics of input signals, thereby eliminating feedback signals in a more stable manner. The performance of the proposed scheme is verified through simulation and experiments.
Read full abstract