N95- 32224 TDA ProgressReport 42-121 May15,1995 An Efficient Implementation of Forward-Backward Least-Mean-Square Adaptive Line Enhancers H.-G. Yeh Spacecraft TelecommunicationsEquipment Section T. M. Nguyen Communications Systems Research Section An efficient implementation of the forward-backward least-mean-square (FBLMS) adaptive line enhancer is presented in this article. Without changing the characteristics of the FBLMS adaptive line enhancer, the proposed implementation technique reduces multiplications by 25 percent and additions by 12.5 percent in two successive time samples in comparison with those operations of direct implemen- tation in both prediction and weight control. The proposed FBLMS architecture and algorithm can be applied to digital receivers for enhancing signal-to-noise ratio to allow fast carrier acquisition and tracking in both stationary and nonstationary environments. I. Introduction Adaptive line enhancers (ALEs) are useful in many areas, including time-domain spectral estimation for fast carrier acquisition [2-4]. For example, a fast carrier acquisition technique [2], 1 as shown in Fig. 1, will be very useful for a deep-space mission, especially in a nonstationary environment or emergencies. Figure 1 is the block diagram of an ALE in a digital receiver used for both acquisition and tracking. First, the receiver is in the acquisition mode. Second, when the uplink carrier is acquired as indicated by the lock detector, the switch is shifted to the tracking position and the tracking process takes over immediately. With this acquisition scheme, the uplink carrier can be acquired by a transponder in seconds (as opposed to minutes for the Cassini transponder). Although devised to support a space mission, the architecture of the forward-backward least-mean-square (FBLMS) ALE and the associated algorithm proposed in this article are also applicable to other systems, including fixed-ground and mobile communication systems. Note that this proposed ALE scheme in the receiver needs a residual carrier, and does not work directly in suppressed-carrier cases. A conventional ALE system using a least-mean-square (LMS) algorithm is depicted in Fig. 2, where z -1 represents a delay. The analysis of the ALE for enhancing the signal-to-noise ratio (SNR) to allow fast acquisition is given in [2]. The block diagram of a FBLMS adaptive line enhancer is shown in Fig. 3. The performance analysis of the FBLMS adaptive line enhancer is provided in [1]. The FBLMS adaptive line enhancer algorithm enjoys approximately half the misadjustment of that of the LMS algorithm [1]. 1 T. M. Nguyen, H. G. Yeh, and L. V. Lam, "A New Carrier Frequency Acquisition Technique for Future Digital Transpon- ders," to be published in a future issue of The Telecommunications and Data Acquisition Progress Report. 16