Media-based MIMO: Outperforming known limits in wireless

The idea of Media-based Modulation (MBM), introduced in [1] [2], is based on embedding information in the variations of the transmission media (channel states). MBM offers several advantages vs. legacy systems, including “additivity of information over multiple receive antennas”, and “inherent diversity over a static fading channel”. MBM is particularly suitable for transmitting high data rates using a single transmit and multiple receive antennas. However, complexity issues limit the amount of data that can be embedded in channel states using a single transmit unit. To address this shortcoming, the current article introduces the idea of Layered Multiple Input-Multiple Output Media-Based Modulation (LMIMO-MBM). LMIMO-MBM enables forming a high-rate constellation as superposition of constituent vectors due to separate transmit units. Relying on such a layered structure, LMIMO-MBM can significantly reduce both hardware and algorithmic complexities, as well as the training overhead. Simulation results show excellent performance in terms of Symbol Error Rate (SER) vs. Signal-to-Noise Ratio (SNR). For example, a 4 × 16 LMIMO-MBM is capable of transmitting 32 bits of information per (complex) channel-use, with SER 10
^-5
at E /N
₀
≃ -3.5dB (or SER 10
^-4
at E/N
₀
= -4.5dB). This performance is achieved using a single transmission (no extension in time/frequency), and without adding any redundancy for Forward-Error-Correction (FEC). Application of FEC can further improve the performance. For example, applying Reed-Solomon codes enables transmitting 30 bits of information per (complex) channel-use with a Frame Error Rate (FER) 10
^-5
at E/N
₀
≃ -6dB. Under a set of mild conditions, by applying FEC with error correction capability t, the slope of the error rate vs. SNR (with hard decision decoding) will asymptotically increase by a factor of t +1.