Switch-and-Navigate: Controlling Data Ferry
Mobility under Bounded Message Delays
Liang Ma
†
, Ting He
‡
, Ananthram Swami
§
, Kang-won Lee
‡
and Kin K. Leung
†
†
Imperial College, London, UK
‡
IBM T. J. Watson Research Center, Hawthorne, NY, USA
§
Army Research Laboratory, Adelphi, MD, USA
†
{l.ma10, kin.leung}@imperial.ac.uk,
‡
{the, kangwon}@us.ibm.com,
§
ananthram.swami@us.army.mil
Abstract—We consider the problem of controlling mobile
data ferries for message delivery among disconnected, scattered
domains in a highly partitioned network. Existing work on
data ferry control mostly focuses on predetermined ferry routes,
assuming full observations at the ferry and no explicit Quality
of Service (QoS) constraints on the resulting communications.
In this paper, we aim at designing a QoS-enabled ferry control
solution, which handles both partial observations and bounded
message delays. To this end, we extend our previous work on
data ferry control with partial observations into a comprehensive
hierarchical framework called Switch-and-Navigate (SAN), which
consists of a global switch policy for determining the best domain
to visit and a local navigation policy per domain for searching
for nodes within individual domains. Under the assumption of
Markovian node mobility, both the global and the local control
problems are formulated as Partially Observable Markov Deci-
sion Processes (POMDPs) to maximize the discounted effective
throughput over all domains. Due to the fact that the optimal so-
lution to POMDP is PSPACE-hard, we develop heuristic policies
and further approximations for efficient computation. Simulation
results show that the proposed policies can significantly improve
the performance over predetermined alternatives.
I. INTRODUCTION
The demanding requirements of mobile ubiquitous com-
munications have promoted the development of Highly Par-
titioned Mobile Ad Hoc Networks (HP-MANET) in which
the network, self-organized without the aid of any established
infrastructures, is partitioned into several permanently discon-
nected autonomous domains due to physical obstacles, limited
radio transmission range, severe environmental conditions, or
simply security reasons. Applications of such networks can be
found in many challenged environments, such as battlefield
operations and disaster relief in large areas. Existing research
on Delay/Disruption-Tolerant Networking (DTN) (e.g., [1])
has focused on intermittently partitioned networks, assuming
the disconnected links will be reconnected or new routes
can be discovered, which makes the solutions inapplicable to
permanently partitioned networks. In these networks, to bridge
communications between disconnected domains, designated
communication nodes called data ferries have been proposed
to serve as a carrier to deliver messages from one domain to
another. Programmed to move in a predetermined or dynamic
pattern, data ferries are capable of self-navigating within
and between domains to collect and deliver messages upon
contacting regular nodes. Therefore, proper mobility control is
Research was sponsored by the U.S. Army Research Laboratory and the
U.K. Ministry of Defence and was accomplished under Agreement Number
W911NF-06-3-0001. The views and conclusions contained in this document
are those of the authors and should not be interpreted as representing the
official policies, either expressed or implied, of the U.S. Army Research
Laboratory, the U.S. Government, the U.K. Ministry of Defence or the U.K.
Government. The U.S. and U.K. Governments are authorized to reproduce and
distribute reprints for Government purposes notwithstanding any copyright
notation hereon.
Start
Point
D
1
D
2
D
3
D
4
d
14
d
13
d
21
d
43
d
32
contact occurs in a domain
switch from one domain to another
local navigation
obstacle
Fig. 1. Example of Switch-and-Navigate: It follows a Switch-Search-Load-
Carry-and-Switch cycle
needed for data ferries to operate efficiently. A major challenge
in applying data ferries to military networks is that military
units exhibit high mobility, making it difficult to maintain
accurate node information at the data ferry. Moreover, unlike
traditional DTN, messages in military operations usually have
a finite lifetime, which must be considered in data ferry
control. In this paper, we aim to address these challenges by
designing control policies under finite delay constraints, with
the goal of optimizing the effective throughput in the network.
A. Related Work
Most existing work on data ferry control have concentrated
on the stationary or fully observable scenarios, where non-
ferry nodes are either stationary or always report their po-
sitions when moving. Focusing on the case of intra-domain
ferry control, algorithms and prototypes for mobile elements
scheduling were proposed in [2, 3]. Solutions in [4, 5] utilized
a mobile node to move randomly within a stationary sparse
network to deliver data opportunistically. Without requiring
global knowledge, authors of [6, 7] took advantage of non-
randomness of node movements to design proactive control
policies. Their solutions, however, require long-range radio
communications and the ability to change (non-ferry) node
trajectories, which are not feasible in many applications. The
authors of [8, 9] extended the Traveling Salesman Problem
(TSP) to reduce the delivery delay, but ignored the fact that
buffers in mobile nodes have a diverse collection of mes-
sages with non-identical residual lifetimes. A comprehensive
framework for dynamic ferry control was introduced in [10];
however, it assumed unit traveling time between any two
domains and unbounded message delays.
B. Summary of Contributions
In this paper, we address the problem of ferry mobility
control under partial observations and finite delay constraints.
Assuming Markovian node mobility in each domain, we
borrow the control framework of Partially Observable Markov