Media-Based MIMO: A New Frontier in Wireless Communications

The idea of Media-based Modulation (MBM), introduced in [1], [2], is based on embedding information in the variations of the transmission media (channel state). This is in contrast to legacy wireless systems where data is embedded in a Radio Frequency (RF) source prior to the transmit antenna. 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 (Single Input-Multiple Output Media-Based Modulation, or SIMOMBM). However, complexity issues limit the amount of data that can be embedded in the channel state 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, vs. SIMO-MBM. Exploiting the layered constellation structure, a fast iterative algorithm is proposed for signal detection, and a practical (small size and low complexity) RF configuration is presented for embedding information in the channel state. Simulation results show excellent performance in terms of Symbol Error Rate (SER) vs. Signal-to-Noise Ratio (SNR). For example, a 4 × 16 LMIMOMBM is capable of transmitting 32 bits of information per (complex) channel-use, with SER ' 10−5 at Eb/N0 ' −3.5dB (or SER ' 10−4 at Eb/N0 = −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). This means, in addition to its excellent SER vs. energy/rate performance, MBM relaxes the need for complex FEC structures used in legacy wireless systems, and thereby minimizes the transmission delay. 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 Eb/N0 ' −6dB. It is shown that, 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. Overall, LMIMO-MBM provides a promising alternative to MIMO and Massive MIMO for the realization of 5G wireless networks.