![](https://csdnimg.cn/release/download_crawler_static/16176453/bg1.jpg)
Round-Trip Carrier Synchronization Exploiting the
Broadcast Nature for Distributed Beamforming
Jiawei Hao, Li Guo, Ranjie Hu, Jiaru Lin
Key Lab of Universal Wireless Communications, Ministry of Education
Beijing University of Posts and Telecommunications, Beijing 100876, P.R. China
Email: {hungryhao, guoli, huranjie, jrlin}@bupt.edu.cn
Abstract—Distributed beamforming is an energy-efficient wire-
less communication technique that allows cooperative nodes to
transmit a common message to a receiver. This paper investigates
carrier synchronization for distributed beamforming in large-
scale wireless networks. In order to pursue less timeslots for
synchronization, we present a time-slotted round-trip carrier
synchronization protocol exploiting the broadcast nature of the
wireless links with master-slave architecture among transmitters.
Numerical simulations are presented and the results indicate that
this protocol can achieve longer valid communication time for
beamforming.
I. INTRODUCTION
In wireless communication networks, distributed beamform-
ing is a cooperative technique which describes the case that
some separate nodes with common information work togeth-
er to form a beam in a specific direction. It can provide
significant benefits in signal-to-noise ratio gains, extend the
communication range and provide increased energy efficiency.
In cognitive radio networks, secondary nodes use distributed
beamforming to minimize the interference to the primary users
while ensuring the average received power lies above certain
threshold.
Since each node in distributed beamforming has an in-
dependent local oscillator, phase and frequency offsets exist
among them. Nodes require some methods to synchronize
their carrier signals so that the bandpass transmissions com-
bine constructively with reasonable phase alignment after
propagation to the intended destination [1]. Lots of carrier
synchronization methods have been proposed for distribut-
ed beamforming: master-slave architecture, one-bit feedback
scheme, time-slotted round-trip carrier synchronization, etc. In
[2], the authors presented a master-slave scheme to realize the
frequency synchronization for distributed beamforming. The
one-bit feedback scheme was proposed in [3]. An iterative
algorithm to adjust phase at each transmitter, which makes
nearly perfect phase alignment at the receiver under static
channels, was presented. This algorithm uses positive one-bit
feedback information to align carrier phases and does not need
the channel state information. But its convergence rate is not
so fast. Similar algorithms were proposed in [4], [5], [6] to
improve the performance of the one-bit feedback scheme.
This work was supported by National Natural Science Foundation of China
(No. 61271178 and No. 61171100), National Key Scientific and Technological
Project of China (No. 2012ZX03004005002, No. 2012ZX03003007 and No.
2013ZX03003004).
The round-trip carrier synchronization protocol was pro-
posed in [7] with two cooperative transmitters. The synchro-
nization was established by using continuously transmitted
beacons at different frequencies. A new time-slotted round-
trip carrier synchronization protocol using a single frequency
for all beacons was proposed in [8], which makes the signals
arrive at the same phase after beamforming and considers both
single-path and multi-path channels. And in [9], the authors al-
so considered the transmitters and/or destination mobility. [10]
presented another time-slotted round-trip protocol in large-
scale wireless networks which uses a master-slave architecture.
A similar protocol called “two-way” synchronization in [11]
makes the transmitters achieve the actual synchronization
before beamforming.
In this paper, we propose a different approach by exploit-
ing the broadcast nature of the wireless links in large-scale
wireless networks based on the protocol in [10]. A master-
slave architecture among transmitters is used and a similar
time-slotted round-trip scheme is proposed to achieve carrier
synchronization at the receiver for distributed beamforming.
The details of the protocol are described with single-path
channels and the process is in the time-division duplexed oper-
ation. The contribution of our synchronization protocol is that
less timeslots are required to achieve carrier synchronization
while longer valid communication time for beamforming can
be realized.
II. S
YSTEM MODEL
We consider the system model in Fig. 1. M source nodes
(N
1
,N
2
, ..., N
M
), each of which possesses a single isotropic
antenna and is located arbitrarily, send a common message to
one destination node (N
0
) by using distributed beamforming.
The channel from node N
i
to node N
j
is modeled as a linear
time-invariant (LTI) system with impulse response h
i,j
(t).We
assume that the noise in each channel is additive, Gaussian
and the impulse response is reciprocal in both directions, i.e.,
h
i,j
(t)=h
j,i
(t).
Suppose that the nodes in the system do not possess a
common time reference such that each node keeps its local
time with independent local oscillator. Half-duplex constraint
is required to be satisfied among all nodes in the process.
This means that nodes cannot transmit and receive signals
simultaneously.
978-1-4799-8078-9/15/$31.00 ©2015 IEEE
22nd International Conference on Telecommunications (ICT 2015)
368