Distributed MIMO Measurements for Integrated Communication and Sensing in an Industrial Environment

TL;DR

A measurement system and results of sub-6 GHz distributed multiple-input-multiple-output (MIMO) measurements performed in an industrial environment showed that new channel models are needed that are spatially consistent and deal with the nonstationary channel properties among the antennas.

Abstract

Many concepts for future generations of wireless communication systems use coherent processing of signals from many distributed antennas. The aim is to improve communication reliability, capacity, and energy efficiency and provide possibilities for new applications through integrated communication and sensing. The large bandwidths available in the higher bands have inspired much work regarding sensing in the mmWave and sub-THz bands; however, the sub-6 GHz cellular bands will still be the main provider of wide cellular coverage due to the more favorable propagation conditions. In this paper, we present a measurement system and results of sub-6 GHz distributed MIMO measurements performed in an industrial environment. From the measurements, we evaluated the diversity for both large-scale and small-scale fading and characterized the link reliability. We also analyzed the possibility of multistatic sensing and positioning of users in the environment, with the initial results showing a mean-square error below 20 cm on the estimated position. Further, the results clearly showed that new channel models are needed that are spatially consistent and deal with the nonstationary channel properties among the antennas.

Figures (20)

• Figure 1: An illustration of a three-node, multilink setup. The dashed line between the two Rubidium clocks (Rb-clock 1 and Rb-clock 2) illustrates that if the two Rubidium clocks are well synchronized---over several hours---then they can be disconnected for some time without losing the synchronization of the radios. To the RF ports of the usrp, one can either connect single antennas or switched arrays.
• Figure 2: During one tdma slot, only one antenna is transmitting while all others are receiving. In the next TDMA slot, the next antenna is transmitting while all other are listening. By saving all channels, even the reciprocal ones, one can use the information for over-the-air calibration.
• Figure 3: The TDMA-based signal structure. Each antenna is assigned a dedicated TDMA slot. During each transmission, the antenna transmits $R$ repetitions of the sounding signal, with some being used as guards and the rest for averaging to increase the signal-to-noise ratio.
• Figure 4: Depiction of three of the $H_\mathrm{a}$ antennas distributed in space. In TDMA slot 1.0, the antenna 1.0 is transmitting while all the others are listening. In then next tdma slot, antenna $2$ is transmitting. Since antennas are distributed in space, it is clear from the figure that an agc is needed; antenna $H_\mathrm{a}$ might need all the available gain when antenna $1$ is transmitting, while that same gain setting might saturate the adc when antenna $2$ is transmitting.
• Figure 5: (a) A photograph showing a view of the environment. The hall is approximately 30 x 11 with a ceiling height between 810 depending on location. (b) A photograph showing the placement of four antennas (circled in red). In total, there were twelve distributed antennas; six on each side of the hall. They were situated 4m above the floor, with a separation of 4m.