A Comparison of Indoor MIMO Measurements and Ray-Tracing at 24 and 2.55 GHz

Abstract

Colocated $4\times 4$ multiple-input multiple-output measurements at 2.55 and 24 GHz are presented for two university buildings consisting of classrooms and offices. Link gain in hallways and connected laboratories looks similar at the two frequencies when the effect of lower effective receive antenna aperture with increasing frequency is removed. Non-line-of-sight (NLOS) propagation through a wall or around hallway corners exhibits approximately 5–20 dB (11 dB on average) greater loss beyond the 20 dB aperture loss at 24 GHz compared to that at 2.55 GHz. Fixed directional antennas increase path loss (PL) by an average of 13 dB when misaligned. Capacity for normalized signal-to-noise ratio is very similar in the two bands and is close to that for the optimal independent identically distributed case, indicating sufficient multipath for spatial multiplexing at 24 GHz. A ray-tracing study suggests that material loss must increase from 2.55 to 24 GHz to correctly predict the higher PL at 24 GHz in severely obstructed scenarios, indicating a need for future material characterization in high microwave bands. The results suggest that 24 GHz is a viable option to replace medium-range (10–30 m) NLOS wireless services currently operating at 2.4 GHz.