Search

We present a method to efficiently apply turbo coding to OFDM in-phase/quadrature index modulation (OFDM-IQ-IM) for both cases of presence and absence of puncturing. We show that using the posterior probabilities of turbo-coded bits in the Bahl-Cocke-Jelinek-Raviv (BCJR) decoding algorithm yields the same decoding performance as using the likelihood probabilities of turbo-coded bits. We investigate the difference in error rate between systematic bits and parity check bits of systematic turbo coding and the difference in error rate between index selecting bits (ISBs) and symbol selecting bits (SSBs) of OFDM-IQ-IM. Based on the results, we devise a new bit mapping scheme that can efficiently split turbo-coded bits into ISBs and SSBs in OFDM-IQ-IM. Simulations show that the proposed bit mapping scheme improves the bit error rate (BER) of the turbo-coded 4QAM-based OFDM-IQ-IM approach and that the turbo-coded 4QAM-based OFDM-IQ-IM approach applying the proposed bit mapping scheme outperforms the turbo-coded 4QAM-based OFDM approach in the high signal-to-noise ratio (SNR) region.

Based on grouping the pulses that compose the sub-carrier of the time-multiplexed binary offset carrier (TMBOC), we present an unambiguous correlation function that has no side-peaks causing the ambiguity in the TMBOC signal tracking, and also, that has a higher degree of the sharpness of the main-peak (DSM) over those of the conventional unambiguous correlation functions. Splitting the pulses composing the TMBOC sub-carrier into 12 groups, and, subsequently, forming the 12 group correlations corresponding to the 12 groups, we investigate the contribution of each of the 12 group correlations to the side peaks and main peak of the TMBOC autocorrelation function. Then, we propose a combining method of the 12 group correlations based on the investigation to remove the side peaks while improving the DSM. Numerical results demonstrate that the proposed unambiguous correlation function not only has no side-peaks, but also has a higher DSM, and, consequently, has a better tracking error performance than those of the conventional unambiguous correlation functions.