MIMO in the real world: where the promised gain goes

Started by Yuri_Phased_eng · 2026-11-18 03:32 UTC · 12 views · 2 replies
Y Member
Collected from a few failed attempts and one that worked.

Common pitfalls:
1) Spatial correlation at the receiver is the usual culprit — closely spaced antennas at a high elevation angle are correlated.
2) The propagation environment has to provide enough angular spread; a corridor is not a rich scattering environment.
3) Mutual coupling and pattern overlap reduce the effective diversity order.
4) Measure the correlation matrix in situ rather than trusting the theoretical capacity.
A Member
The in-situ correlation matrix point is the one most people skip. In practice we got more mileage from the eigenvalue spread than from the correlation coefficient itself: a pair that looks decorrelated at 0.3 can still have a nearly degenerate second eigenvalue once mutual coupling is included, and that is what kills the second stream. The other thing that bit us was calibration drift between receive chains. Half a dB of amplitude and about 5 degrees of phase error was enough to pull the measured gain well below 1 dB in a 2x2.
P Member
We've hit this exact pitfall. Ended up switching vendors - sometimes it's a supply issue, not a design issue.
广告 AdRenesas 瑞萨微控制器芯片产品照片Renesas Electronics微控制器、模拟与射频/时钟方案 · MCUs, analog, RF and timing访问官网 →免费展示 · 资料来自公开信息 · 如需撤下请联系 ads@rfforum.com
广告 Ad广告位招商 / Advertise hereads@rfforum.com · 刊例 Rate cardT2 · 728×90
Log in or register to reply.