Analysis of a novel time domain pilot scheme in convolutional multiplexed multicarrier systems

Abstract

This paper analyzes the performance of a novel time domain pilot scheme using least square channel estimation in convolutional multiplexed multicarrier (CMM) systems. To reduce decoding complexity and achieve higher spectral efficiency, data symbols are encoded through serial concatenated structure of convolutional multiplexing technique, while simultaneously pilot sequences are generated and multiplexed with orthogonal frequency division multiplexing (OFDM) symbols in time domain. The least square channel estimation is implemented in time domain and data symbols are detected with iterative decoding. The pilot sequences are generated based on complementary codes, whose periodic auto-correlation function (ACF) and cross-correlation function (CCF) are perfect, and there are no multiple access interference (MAI) and inter-symbol interference (ISI) when the maximum time delay is less than length of zero correlation zone (ZCZ) of the pilot sequences. The time domain pilot scheme is turned out to have high estimation accuracy with low calculation complexity. Mean square estimation (MSE) results are verified to be independent of unknown data symbols and insensitive to the power fraction for pilots. Combined with iterative detecting, bit error rate (BER) and frame error rate (FER) performance of proposed scheme can match with ideal channel estimation scenario whose channel impulse response (CIR) is known by the receiver. Besides, data transmission scheme with interference cancellation (IC) can retrieve half of the spectral efficiency loss than scheme without IC and only sacrifices tolerable estimation accuracy.