Decision-Directed Adaptive Algorithms for Mobile Radio Channel Estimation: A Comparative Study SEEDAHMED S. MAHMOUD 1 , ZAHIR M. HUSSAIN 1 , and PETER O’SHEA 2 1. School of Electrical and Computer Engineering, RMIT University, Melbourne Emails: S2113794@student.rmit.edu.au , zmhussain@ieee.org 2. School of Electrical and Electronic Systems Engineering Queensland University of Technology, Brisbane AUSTRALIA Abstract:- In this paper two important adaptive algorithms for channel impulse response estimation are studied. The selected algorithms are the decision-directed recursive least square (RLS) and the decision-directed least mean square (LMS) algorithms, weighted and unweighted. These algorithms are tested to estimate the impulse response of the newly proposed geometrical-based hyperbolic distributed scatterer (GHDS) channel model. It is shown that the performance of both the ideal decision weighted LMS (IDWLMS) and the ideal decision weighted RLS (IDWRLS) algorithms is better than that of the unweighted versions of these algorithms. Keyword:- Adaptive algorithm, channel estimation; decision-directed; least mean square(LMS); recursive least square (RLS). 1 Introduction Over the past few decades, radio communication systems have undergone extensive developments. The demands that a radio system must fulfil are greater by the day. Channel estimation algorithms allow the receiver to approximate the impulse response of the channel and explain the behaviour of the channel. This knowledge of the channel's behaviour is well utilized in modern radio communications for improvement of the system performance. Adaptive channel equalizers utilize channel estimates to overcome the effects of inter symbol interference (ISI). Diversity techniques (e.g. in the IS-95 Rake receiver) utilize channel estimation to implement a matched filter such that the receiver is optimally matched to the received signal instead of the transmitted one [1]. Adaptive beamformers utilize some of the channel-estimated parameters to adjust their weights. Maximum likelihood detectors utilize channel estimates to minimize the error probability [2]. One of the most important benefits of channel estimation is that it allows the implementation of coherent demodulation. Coherent demodulation requires the knowledge the phase of the signal. This can be accomplished by using channel estimation techniques. Once a model is established, its parameters should be continuously updated (estimated) in order to minimize the error as the channel conditions change. If the receiver has a prior knowledge of the information sent over the channel, it can utilize this knowledge to obtain an accurate estimate of the impulse response of the channel. This method is simply called training sequence based channel estimation. It has the advantage that can be used in any radio communications system quite easily. However, it has some drawbacks. One of the obvious drawbacks is that it is wasteful of bandwidth. Precious bits in a frame that might have been otherwise used to transport information are stuffed with training sequences for channel estimation. Blind methods, on the other hand, require no training sequences. They utilize certain underlying mathematical information about the kind of data being transmitted. These methods might be bandwidth efficient but still have their own drawbacks. They are notoriously slow to converge (more than 1000 symbols may be required for an FIR channel with 10 coefficients). Decision directed estimation is a subclass of blind estimation techniques; it uses the detected symbols to reconstruct the transmitted signal, and then uses this signal in place of the original signal. In [3], performance comparison was made between the decision-directed algorithm and the constant modulus algorithm (CMA), as well as the spectral self-coherence restoral (SCORE) algorithm. Results