Radio on Fiber-Distributed Antenna System (RoF- DAS) is known to improve coverage and performance of wireless communications. Currently, various multiplexing schemes in RoF-DAS for transport Multiple Input Multiple Output (MIMO) signals in a fiber link are studied. The RoF-DAS over Wave- length Division Multiplexing - Passive Optical Network (WDM- PON) with optical Time Division Multiplexing (TDM) is actively researched. This system uses optical switcher to multiplex a set of MIMO radio signals in a fiber and transfer to the Remote Antenna Unit (RAU). Since a pair of switches multiplex and demultiplex MIMO radio signals, two switches are required to synchronize completely. Therefore, there is a problem that high performance switches are required to perform the clock extraction. This paper proposes asynchronous optical TDM. In this proposal, synchronization mismatch is compensated by estimating the amount of drifting and cancel it out at the RAU. The bit error ratio (BER) performance is evaluated by using computer simulation.

1. Transportation of MIMO Radio Signals over
RoF-Distributed Antenna System and its
Performance Analysis in the Presence of
Incomplete Synchronization in Optical TDM
Nara Institute of Science and Technology , Japan
Network System Lab
Tatsuya Kidani, Takeshi Higashino, Minoru Okada
1

2. Outline
• Background
• Concept of RoF-DAS using optical TDM
• Configuration of RoF-DAS using optical
TDM
• Channel Model with synchronization
mismatch
• Simulation
• Summary & Future Work
2

3. Background

Requirements of Next Generation Network
Source2
Source1
High speed and
large capacity
wireless access
Adaptive
control of base
station
Accommodations
of various
communication
standards
(Heterogeneous)
Source1:Cisco VNI Mobile, 2011
Source2:ZTE ZTE technology N0.1, 2011

RoF-DAS using optical TDM (Time Division
Multiplexing)
[Radio on Fiber (RoF),Distributed Antenna System (DAS)]
3

4. RoF-DAS
RoF-DAS utilizes many antennas (RAU)
geographically located at different location.
CCS : Centralized Control Station, RAU : Remote Antenna Unit
This system can use multiple RAUs to
coordinate as MIMO system
RAU
O/E
Conversion
E/O
Modulation/
Demodulation
,scheduling
（time，frequency），
Signal processing
(MIMO, etc.)
CCS
4
Mobile Terminal
RoF：Radio on Fiber
RoF-link
MIMO

5. Concept of RoF-DAS using
optical TDM
• Radio on Fiber (RoF)
• Distributed Antenna
System (DAS)
• Optical TDM
• Merit
– High Coverage
– Low Spatial Correlation
– Reduction of
transmission power
CCS : Centralized Control Station, RAU : Remote Antenna Unit
5
MT : Mobile Terminal

6. MIMO technique
x1
x2
Transmitter
h11
h12
h21
hM 1 h22
y1
y2
Receiver
H
xN
hM 2
hMN
yM
H air
 h1,1 h1, 2

 h2,1 h2, 2




h
 M ,1 h M , 2
 h1, N 


 
 hM , N 1 

 hM , N 

y  H air x  n y
MIMO technique can increase
channel capacity proportional to
the number of antennas.
:Receive Vector
𝐇 𝑎𝑖𝑟 :Channel Matrix
x
:Transmit Vector
n
:Noise Vector
6

7. Configuration of RoF-DAS using
optical TDM
a
b
c
d
e
a.Optical pulse source
emits periodic pulse
train
b.RF signal modulates
the optical pulse
intensity
c.Employ optical delay
lines and the optical
signals are combined
d.O/E conversion is
performed and signal is
demultiplexed
e.RF original signal are
LN-MZM : lithium niobate - Mach-Zehnder modulator
re-generated from TDM
PD : Photo Diode
demultiplexed signal
with band pass filters
7
(BPFs)

8. Configuration of RoF-DAS using
optical TDM
• Requirement of
SW
– Accuracy
• If Synchronization
mismatch happens,
(Problem)
– Each signals is
mixed (High
Correlation)
We estimate the amount of
synchronization mismatch &
compensate it.
Prevent performance degradation
8

9. Channel Model with
Synchronization Mismatch
• Ideal MIMO
𝒚 = H 𝒂𝒊𝒓 𝒙 + 𝒏
• Channel Model with synchronization mismatch,
𝒙′ =
=
𝑥1 1 − 𝜏
𝑥2 𝜏
𝑥2 1 − 𝜏
𝑥3 𝜏
𝑥3 1 − 𝜏
𝑥4 𝜏
𝑥1 𝜏
𝑥4 1 − 𝜏
1− 𝜏
𝜏
0
0
0
1− 𝜏
𝜏
0
0
0
1− 𝜏
𝜏
𝜏
0
0
1− 𝜏
𝒚′ = 𝑯 𝑻𝒙 + 𝒏
𝜏:Amount of mismatch
(0 ≤ 𝜏 ≤ 1)
𝑥1
𝑥2
𝑥3 = 𝑻𝒙
𝑥4
Delay Matrix
9

10. Compensation of synchronization
mismatch (Proposed)
• ZF with compensation
We implemented the delay matrix estimator at
the RAU.
In this proposal, we utilities pilot sequence 𝒑,
𝐓est = 𝐓𝐩𝐩T + 𝐧RoF 𝐩T
𝐧RoF : Noise in RoF
Compensated transmission signal

𝒙′′ = 𝑻−1 𝒙
𝑒𝑠𝑡
𝒙 𝑐𝑜𝑚 = 𝑻𝑻−1 𝒙
𝑒𝑠𝑡
10

11. Block Diagram
CCS (Centralized Control Station) RAUs (Remote Antenna Units)
Signal
Processor
E/O
TDM
Multiplexing
Feedback 𝑻−1
RoF
O/E
TDM
Demultiplexing
BPF
Delay
Matrix
Estimator
11

12. Simulation
Modulation
Number of RAUs
Number of RAU’s Antennas
Number of MTs
Number of MT’s Antennas
Clock Delay
MIMO Detection
Pilot Sequence
Noise
Fading Channel (Air)
QPSK, 16QAM, 64QAM
4
1
1
4
0.1~0.9
Zero-Forcing
M-sequence
AWGN
i.i.d, Flat Rayleigh Fading
12

13. Channel Capacity
• Shannon-Hartley theorem
𝛾0
𝐶 = log 2 det(𝑰 + (𝑯 𝑎𝑖𝑟 𝑻) 𝐻 (𝑯 𝑎𝑖𝑟 𝑻))
𝑛𝑡
( 𝑯 𝑒𝑠𝑡 = 𝑯 𝑎𝑖𝑟 𝑻)
𝛾0 : ratio between
transmission power and
noise power
𝑛 𝑡 : number of antenna at CCS
𝑰 : identity matrix
13

14. Simulation (SNR=35dB &
16QAM)
14

15. Simulation (𝜏 = 0.4)
8.5dB
15

16. Summary & Future Work
• Summary
– To avoid critical degradation due to the
incomplete synchronization, a new
compensation scheme is proposed.
– This scheme can estimate amount of
synchronization mismatch and also gives an
improvement in BER performance.
• Future Work
– Application of Optical OFDM (Orthogonal
Frequency Division Modulation).
– The compensation of nonlinear distortion due
to RoF characteristic.
16