Xiwen JIANG
Supervised by Florian Kaltenberger and Luc Deneire Labex UCN@Sophia
http://ucnlab.eu
Exploiting TDD Channel
Reciprocity in Massive MIMO
Future Services and Requirements
Wireless Network Evolution
2G 3G 4G 5G
ucnlab.eu 3
SISO SU-MIMO MU-MIMO Massive MIMO
Single Input Single Output
Multiple Input Multiple Output
Multi-user
MIMO Large number of antennas at the Base Station
Massive MIMO in 5G
Massive MIMO
Capacity
Cost
Latency Scheduling
Robustness
FDD vs. TDD in Massive MIMO
ucnlab.eu 5
Downlink Band Uplink Band Frequency
Time
Downlink Band
Uplink Band
Time Frequency
Downlink Uplink
FDD: Heavy feedback for getting channel information at base station in Massive MIMO
TDD: channel information at base station can be obtained from channel reciprocity
Hardware impairment -- Calibration
BS TX Downlink UE RX
Base Station (BS) Channel in The Air User Equipment (UE)
UE TX Uplink
BS RX
Not Reciprocal Reciprocal Not Reciprocal
Calibration: Compensate Hardware Impairment
System Model
22/04/20
15 - 7
Site A (BS) N Antennas
Site B (UE)
Single Antenna
Experiment Setup
Site B Site A
• 1,9GHz
• 5MHz bandwidth
• LTE-like OFDM frames
Frequency and time synchronized boards
ExpressMIMO2 card ExpressMIMO2 card
OpenAir4G software modem
Experiment Results 1
22/04/20
15 - - p 9
-1.5 -1 -0.5 0 0.5 1 1.5
-1.5 -1 -0.5 0 0.5 1 1.5
Real
Imaginary
f11
f22
300 fij(ij)
1
f44
f33
Measured 4x1 MISO System Calibration Matrix
Experiment Results 2
4x1 Beamforming Gain based on TDD calibration
1.9050 1.906 1.907 1.908 1.909 1.91 1
2 3 4 5 6 7
Frequency (GHz)
Beamforming Gain (dB)
Ideal
No Calibration Full
Diagonal Curves for ideal, full
and diagonal modes
almost overlap each other.
Experiment Results 3
22/04/20
15 - 11
Variation of Calibration parameter during the time
0 200 400 600 800 1000
-10 -8 -6 -4 -2 0 2 4 6 8 10
Time (Minutes)
Deviation (%)
Magnitude Angle
