Summary
International Symposium on Antennas and Propagation
2009
Session Number:3E4
Session:
Number:3E4-4
Adaptive Interference Suppression in Communication Systems using Direct Data Domain Least Squares (D3LS) Approach
Santana Burintramart, Akkarat Boonpoonga, Tapan K. Sarkar,
pp.1123-1126
Publication Date:2009/10/21
Online ISSN:2188-5079
Summary:
Adaptive array has been introduced to wireless communications to enhance frequency reuse by adaptively suppress undesired signals and receive only the desired one. Many methods have been proposed in the literature including least mean squares (LMS) [1], recursive least squares (RLS) [2], sample matrix inversion (SMI) [3], and their other variation forms [4]. The major deficiencies of these methods are their convergence rate and the required training sequence. Even though another type of adaptive algorithms, such as constant modulus algorithm (CMA) [5], has been introduced to solve the training issue, it is also sensitive to its initial condition [6]. In contrasts to the classical methods, the direct data domain least squares (D3LS) algorithm has been developed and proposed for interference cancellation, mainly for radar scenarios where the direction of the signal of interest (SOI) is known a priori [7], [8]. The D3LS method has an advantage over the classical methods in the sense that it does not require training data and it is a single snapshot processing. That means there is no convergence issue for the D3LS. However, in communication scenarios, the direction of the SOI is usually not known, and the D3LS cannot be applied. In this paper, the concept of the D3LS is presented to communication systems where the direction of the SOI is not known a priori. Instead, the method utilizes the knowledge of the training sequence in its weight calculation. It will be shown that the number of training sequence required is equal to the number of antennas in the array. Simulation results show that the proposed method can effectively suppress undesired signals and recover the desired one.