This paper aimed at a comparative study of the performance of the Least Mean Square (LMS) and Recursive
Least Square (RLS) algorithms of the adaptive beam forming antenna on the CDMA based network. The smart
antenna test-bed used include a single input multiple output system which consists of one transmitter and six
receivers. During experimentation, the interference and noise reduction capabilities of the adaptive antenna
were investigated using the LMS and RLS algorithms. Also, the two adaptive algorithms were simulated and
evaluated for a 6 uniform linear array elements with inter-element spacing of 0.5λ on the CDMA based network
using MATLAB version 7.5. In the simulation, the angle of arrival of the desired signal and the interfering
signal were at 30 ̊ and 60 ̊ respectively. The desired signal was assumed to be a complex sinusoidal phase
modulated and the interfering signal assumed as random signal. Results obtained showed that the recursive
least square algorithm started to converge from iteration number 20 while least mean square algorithm started
to converge from iteration number 50. Result also showed that the recursive least error was 0.001 at 20
iterations, while least square error was 0.025 at 20 iterations. Result obtained also showed that RLS algorithm
has lower bit error rate of order 10-3 compared to LMS algorithm.