Wireless Sensor Networks (WSN) is a new research on wireless technology. How to route data is an important issue for WSN [26] [27] [28]. Some known researches on routing for Ad-Hoc networks are infeasible for wireless sensor network due to the hardware limitation of the sensor nodes. In this thesis, we will introduce a new routing scheme for wireless sensor network whose purpose is to reduce energy consumption and distribute the energy cost evenly.
In the following paragraphs, we will introduce the idea of wireless senor network in Section 1.1, and the routing problem in wireless sensor network will be presented in Section 1.2. Furthermore, we will specify our contribution in Section 1.3, and introduce the organization of this thesis in Section 1.4.
1.1 Wireless sensor networks
With the increasing advances in hardware and wireless network technologies, the small wireless devices will be able to provide access to information anytime and anywhere [6]. The wireless sensor network is a kind of application that is formed with a set of small sensor devices that are deployed in an ad hoc fashion and cooperate on sensing a physical phenomenon. Although the sensor nodes are not reliable, hundreds or thousands of these nodes deployed in the network can provide a high quality and fault tolerant sensing network. It is important in many applications for military or commercial. For example, for a security system, acoustic, seismic, and video sensors can be used to form an ad hoc network to detect intrusions. Sensor nodes can also be used to monitor machines for fault detection and diagnosis.
The differences between wireless sensor network and other wireless networks, such as Mobile Ad-hoc Networks (MANET) are:
(1) Critical of energy consumption:
The small volume of sensor node causes the critical battery capacity. The energy consumption becomes the most critical issue of the design of sensor node. The energy consumption is often less than 1μW.
(2) Low communication bandwidth:
The bandwidth of wireless sensor network is about 10 – 1000kb/s. It is relative low to the traditional wireless networks.
(3) Limited memory space and computing power:
Due to the small volume and low cost of each sensor node, the memory space and computing power are critically limited. The memory space ranges from several kilo bytes to hundreds kilo bytes. The computing power ranges from 4MHz to 100MHz.
(4) The large scale of deployment:
A wireless sensor networks consists of hundreds even thousands of small wireless sensor nodes. Those sensor nodes are deployed in a large wide area for sensing a physical phenomenon.
(5) High node failure rate:
The terrible deployed environment may make sensor nodes easy to be broken, and there may be some obstacles blocked the communication signal and made sensor nodes temporary unavailable.
Wireless sensor networks can contain hundreds or thousands of sensing nodes. It is desirable to make these nodes as cheap and energy-efficient as possible and rely on their large numbers to obtain high quality results. Network protocols must be
designed to minimize the energy consumption on sensor nodes. In addition, since the limited wireless channel bandwidth must be shared among all the sensors in the network, routing protocols for these networks should be able to perform local collaboration to reduce bandwidth requirements.
There are lots of applications for wireless sensor networks. In military applications, the wireless sensor network can be used for command, control, and communication. In health applications, they can be deployed on patients to monitor and assist the disabled patients. In commercial applications, they can be used for managing inventory, and product quality monitoring. There are many other applications such as disaster area monitoring, traffic monitoring…etc.
1.2 The transmission in wireless sensor networks
In most applications for wireless sensor network, sensor nodes are deployed widely within an area and the mission is to sense certain data (as shown in figure 1-1).
How to send the sensed data back to the base station? We need a good solution that can reduce the energy cost on sensor nodes to keep sensor nodes alive as long as possible.
For example, one of the methods to route data is that each node which has data to send just forwards the data to its neighboring nodes. It is easy to implement but each node on the data path spends energy on receiving and transferring data. Moreover, every node has to keep awake in order to listen if there is any data forwarded to it.
This takes lots of energy and makes the nodes die quickly. Actually, not all nodes need to keep awake to maintain the network connectivity, we can organize a set of nodes and each of the nodes is responsible to a region and those regions cover all the nodes
in the deployed area to collect all the data sensed by sensor nodes. This is what we called “clustering”.
Figure 1-1: A sensor network with hop-by-hop routing
Generally speaking, cluster-based routing protocol has following advantages:
(1) Energy saving for cluster member
Each cluster has a cluster head which will keep awake to collect data from the other nodes in this cluster. Cluster member (nodes except cluster head) can turn off the radio module until it senses any data. Therefore, if there is nothing sensed, the cluster members spend extremely low energy during the round.
(2) Easy to maintain routing table
For cluster members, each of them only need to remember which node is the cluster head. This takes less memory space then other routing approach, e.g.
DSR, AODV. This is critical for sensor nodes due to its small RAM size.
Sensed some data
Sensed some data
Sensor node Base station
(3) Data aggregation at cluster head
Sensor nodes may sense much redundant data in the same region. Cluster head can perform data aggregation function to reduce the data. Cluster head can spend less energy since the data has been reduced.
Wireless sensor nodes are expected to work as long as possible so how to prolong the network lifetime is a very important issue in wireless sensor network.
Cluster-based routing protocol is a good method which can save energy easily because of the features mentioned above. There are some cluster based solutions [1]
[2] [3] and we will introduce them in section 2.
1.3 Contributions
We presented a new energy efficient routing scheme for wireless sensor network.
Unlike the traditional cluster based routing schemes, we use the base station to organize the cluster distribution instead of node self election. By our scheme, sensor nodes can save more energy than some other solutions so that the network lifetime can be increased. Moreover, the energy cost is distributed evenly to each node. This makes nodes can live longer to keep the deployed area being covered.
We did some simulations for our proposed scheme and the results showed that the properties we claimed are true. In terms of network lifetime, our scheme has 60%-90% improvement compare to LEACH and HEED. By our scheme, the lifetime of wireless sensor network can be maximized. Power saving is an important issue in wireless sensor network and we did a well improvement in this field.
1.4 Synopsis
The rest of this thesis is organized as follows. The related work of the routing protocol on wireless sensor network will be presented in section 2. We will propose a centralized energy efficient routing scheme in section 3. In section 4, we will introduce the simulation we have done and show the result. Some comparisons are made with the other routing approaches for wireless sensor network. Finally we will make a conclusion in section 5.