Prioritization of Patient Vital Signs Transmission Using Wireless Body Area Networks - E-Healthcare Systems and Wireless Communications - page 262

sifications. Specifically, we aim at maximizing the
e2e reliability of these three traffic classes while
meeting the delay requirements. Here reliability
is defined as the ratio of the number of received
data packets (containing vital sign information) at
the sink (or base station) and the number of total
packets sent from a source node.
Our solution focuses on a wireless communi-
cation network with light and moderate conges-
tion , as categorized in Table 2, where γ R , γ Y , and
γ G are the current reliability of “red”, “yellow”,
and “green” patients, respectively. When the
network is in no congestion status, in fact, the
use of standard protocols is enough to guarantee
the services to all patients. On the other hand,
when the network is in heavy congestion status,
additional mechanisms are needed to guarantee
the service to patients under critical conditions.
For example, transmission of vital signs from
patients in non-critical conditions can be held on
until the congestion becomes less severe (e.g.,
adopting source rate control). Moreover, adap-
tive sampling techniques consisting in reducing
the sampling rates of the sensors deployed on
patients under non-critical conditions can be
applied in order to reduce the traffic. Once the
overall traffic is reduced, our solution, which is
tailored for light or moderate congestion, can be
applied to guarantee e2e Quality of Service (QoS).
Therefore, we focus on providing a solution for
light and moderate congestion; in these states,
using our communication solution, the services
to the patients under critical conditions can be
guaranteed.
To sum up, the objectives of our work are:

• Provide partial cognitive radio (Mitola &

Maguire, 1999) capability to sensors in or-
der to avoid EMI.

In order to achieve these objectives, a cross-
layer communication solution is proposed to offer
a prioritization service and to maximize reliability
while meeting the e2e delay requirement based
on the patient's condition and data content. Our
solution adopts a modular design : the modules
include Medium Access Control (MAC) , Routing ,
and Scheduling . Each module is individually de-
signed to meet the domain-specific requirements;
then, the three modules are jointly optimized to
obtain the best performance possible. The quality
of multiple channels is considered in the MAC and
routing modules, which leads to the interference-
aware design of Multi-channel Quality-based
MAC (MQ-MAC) and Channel Quality Based
Routing (CQBR). Moreover, a two-level data
packet scheduling scheme is proposed to maximize
the reliability for all three classes of traffic while
guaranteeing their e2e delay requirements. These
modules are also designed to be of low complex-
ity so that resource-limited sensors can run them.
Note that we aim at providing a solution to
situations where the traffic is near to the network
capacity. Our solution is based on the well-known
Crossbow's wireless sensors IMote2/TelosB,
which use the IEEE 802.15.4/ZigBee standard
(IEEE Computer Society, 2006). To improve the
network performance, our solution can be easily
migrated to other high-speed wireless platforms
such as 802.11b/g/n.

• Provide a communication solution for in-

hospital networks with light and moderate
congestion .

Table 2. Network Congestion Types

Congestion Type Network Condition

• Maximize the reliability for all three class-

es of traffic while guaranteeing their e2e
delay requirements.

No Congestion γ R = γ Y = γ G =1

Light Congestion γ R = γ Y =1, γ G <1

Moderate Congestion γ R =1, γ Y <1, γ G <1

Heavy Congestion γ R <1, γ Y <1, γ G <1