A major challenge for future 5G wireless communication which is expected to be deployed in 2020 is to achieve an appealing tradeoff between spectral efficiency and energy efficiency. This requires a new approach in the design of conventional multiple input multiple output MIMO technology. Spatial modulation SM, a promising new MIMO technology which activates only one transmit antenna at any time instant and conveys active antenna index as an added source of information to improve spectral and energy efficiency. Generalized spatial modulation GSM, a recent advancement in SM, where more than one antenna is activated per time slot overcomes the coercing on SM, where number of transmit antennas has to be power of two. GSM with multiple activated antennas called multiple active-spatial modulation MA-SM is introduced as an extension of GSM to further increase spectral efficiency by transmitting different constellation symbols through activated antennas. In this paper, a new and efficient signal detection scheme called ordered block minimum mean square error OBMMSE is used for MA-SM, which can achieve exceptional tradeoff between error performance and computational complexity than the conventional optimal maximum likelihood ML detection. We also consider performance of the detection scheme under correlated frequency flat fading channel. Through the simulation results, it is shown that proposed scheme can achieve significant reduction in complexity with slight performance degradation when compared with that of ML detection. © 2018 IEEE.