Every existing deep learning (DL)-based detector for orthogonal frequency division multiplexing with index modulation (OFDM-IM) is optimized for one instance for modulation parameters (MPs) such as the modulation orders and the number of activae subcarriers but does not operate well for other sets of MPs. In this paper, we propose a novel index-bit-detection neural network with modulation parameter-independent structure (IBDNN-MPI) for OFDM-IM, which aims to operate with various MPs. IBDNN-MPI adopts a concatenation of Inception module and fully-connected layers to detect index bits effectively. It is shown via simulations that IBDNN-MPI implementations show better bit error rate (BER) performance and less complexity than existing DL-based detectors optimized for each set of MPs. They have BER performances close to the maximum likelihood and the log-likelihood ratio detectors. Moreover, they can be applicable to dual-mode OFDM-IM at slightly increased computational complexity and cover the scenarios with imperfect channel state information.INDEX TERMS Deep learning, Inception module, orthogonal frequency division multiplexing with index modulation (OFDM-IM), log-likelihood ratio (LLR) detector, maximum likelihood (ML) detector, parameter-independent detector. I. INTRODUCTION 1 O RTHOGONAL frequency division multiplexing (OFDM) 2 is a well-known multi-carrier system in high-speed 3 wireless communication and has an issue that the sub-channel 4 orthogonality can be lost and inter channel interference (ICI) 5 can occur due to the characteristic of rapidly varying wire-6 less channels [1]. Recently, OFDM with index modulation 7 (OFDM-IM) [2] and dual mode OFDM-IM (DM-OFDM-IM) 8 [6]-[9] are proposed. In OFDM-IM, only a part of subcarriers 9 are activated and modulated to carry message bits which are 10 called carrier bits. According to the indices of the activated 11 subcarriers, additional message bits are virtually transmitted, 12 which are called index bits. The OFDM-IM can have higher 13 energy efficiency and reliability than the conventional OFDM 14 by mitigating the ICI. In DM-OFDM-IM [6], all subcarriers 15 are activated and modulated with two clearly distinguishable 16 constellations while only a part of subcarriers are activated in 17 OFDM-IM. The index bits in DM-OFDM-IM are transmitted 18 depending on the pattern which constellation is selected 19 This article has been accepted for publication in IEEE Access.