Özet:
It is well known that the efficiency of a channel estimator employing the strategy of comb-type pilot placement can be controlled by adjusting the positions of pilot tones. In this paper, by considering this situation, in order to maximize the estimation precision of the least squares (LS) algorithm utilized as a channel estimator in orthogonal frequency division multiplexing – interleave division multiple access (OFDM-IDMA) scheme, the genetic algorithm (GA) possessing a wide range of uses due to its powerful problem solving capability was utilized in the optimization of pilot positions. Besides, the computational load of mean square error (MSE) which is used as the objective function of GA was avoided by employing its upper bound during the optimization process. The upper bound of MSE was achieved by utilizing the Gershgorin disc theorem. In the simulations, the suggested pilot arrangement strategy based on the GA was compared to the conventional techniques like equispaced and random pilot placements in point of two criteria known as bit error rate (BER) and MSE. Simulation results put forth that GA-based pilot design strategy establishes a very clear superiority over the other considered methods by providing significant MSE and BER performances.