**3. Simulation experiment**

Through simulation experiments on a wireless sensor network with multiple sinks, the performances of our scheme have been investigated in detail to verify its effectiveness.

#### **3.1 Conditions of simulation**

In a large scale and dense wireless sensor network with multiple sinks consisting of many static sensor nodes placed in a large scale observation area, only sensor nodes that detected abnormal data set were assumed to transmit the measurement data. The conditions of the si-mulation which were used in the experiments performed are shown in Table1. In the initial state of the simulation experiments, static sensor nodes are randomly arranged in the set ex-perimental area, and multiple sinks are placed on the boundaries containing the corners of this area. The network configuration is shown in Fig.4. In the experiments performed, the value attenuation factor accompanying hop (*dr*) and the "value to self" of each sink node in-troduced in our scheme were set to 0.5 and 100.0, respectively.


Table 1. Conditions of simulation

450 Environmental Monitoring

: *data packet <sup>l</sup>*

*m*

*s*

*k*

*<sup>k</sup>* node *m* routing table

Fig.3 shows an example of the above transmission power control, which means that the tra-nsmission power of each sensor node is switched to low power according to the above con-dition. In this example, node (*m*) is a load concentration node. Node (*m*) has autonomously transmitted the data packet to node (*r*) with the greatest connective value within low power range by low power because its own residual energy has become less than the set threshold [*Te*]. By switching to low power, the energy consumption of node (*m*) is saved, but node (*k*) and node (*l*) may continue to transmit the data packet to node (*m*) because they cannot grasp the updated connective value of node (*m*). In our scheme, therefore, every tenth data packet from the node switched to low power is transmitted by

Through simulation experiments on a wireless sensor network with multiple sinks, the perf-

In a large scale and dense wireless sensor network with multiple sinks consisting of many static sensor nodes placed in a large scale observation area, only sensor nodes that

ormances of our scheme have been investigated in detail to verify its effectiveness.

Next Hop

10.0 20.0 50.0 12.0 25.0 12.0 *node node l node n node q node r node s*

*q*

*n*

*r*

Fig. 3. An example of transmission power control

*Sink1*

high power.

**3. Simulation experiment** 

**3.1 Conditions of simulation** 

Fig. 4. Large scale and dense wireless sensor network consisting of many static sensor nodes

In the experimental results reported, our scheme (Matsumoto et al., 2010) is evaluated through a comparison with existing ones (Dubois-Ferriere et al., 2004; Oyman & Ersoy, 2004; Ohtaki et al., 2006; Utani et al., 2008) where the parameter settings that produced good results in a preliminary investigation were adopted in preference to existing ones.

#### **3.2 Experimental results on simulation model with two sinks**

In this subsection, experimental results on the simulation model with two sinks of our scheme without transmission power control are shown, where the number of sensor nodes was 1000, the range of radio wave and the battery capacity of each sensor node were set to 150m and 0.5J, respectively.

Autonomous Decentralized Control Scheme for Long-Term

*MUAA AAR NS Proposal*

**3.3 Experimental results on simulation model with three sinks** 

sinks.

*0%*

Fig. 6. Transition of delivery ratio

*20%*

*40%*

*60%*

*Delivery ratio (%)*

150m.

*80%*

*100%*

Operation of Large Scale and Dense Wireless Sensor Networks with Multiple Sinks 453

sensor nodes is shown, and the lifetime of the simulation model with two sinks, as in Fig.5, is compared. In Fig.6, the existing schemes in Ohtaki et al., 2006 and Utani et al., 2008, which belong to the category of ant-based routing algorithms, are denoted as *MUAA* and *AAR*, respectively. The existing scheme in Dubois-Ferriere et al., 2004 and Oyman and Ersoy, 2004, which describe representative data gathering for a wireless sensor network with multiple sinks, is denoted as *NS*. From Fig.6, it can be confirmed that our scheme denoted as *Proposal* in Fig.6 achieves a longer-term operation of a wireless sensor network with multiple sinks than the existing ones because it improves and balances the load of each sensor node by the communication load reduction for network control and the autonomous load-balancing data transmission. Through simulation experiments, it was verified that our scheme (Matsumoto et al., 2010) is substantially advantageous for the long-term operation of a large scale and dense wireless sensor network with multiple

> *0 1000 2000 3000 4000 5000 6000 7000 8000 The total transmission number of data packets*

In this subsection, through experimental results on the simulation model with three sinks, the effectiveness of the transmission power control introduced in our scheme is evaluated. In the following experimental results, the battery capacity of each sensor node was set to 0.2J, and the range of radio wave of high power transmission in each sensor node was set to 200 m and it of low power transmission in each sensor node was set to

As the first experiment on the simulation model with three sinks, it was assumed that the evaluation node marked in Fig.4 detected an abnormal value and transmitted the data packet with this abnormal value periodically, as in the above subsection 3.2. The routes used by

Fig. 5. Routes used by applying our scheme to the simulation model with two sinks

As the first experiment on the simulation model with two sinks, it was assumed that the evaluation node marked in Fig.4 detected an abnormal value and transmitted the data packet with this abnormal value periodically. The routes used by applying our scheme are shown in Fig.5. Of the 3000 data packets transmitted from the evaluation node, the routes used by the first 500 data packets are illustrated in Fig.5(a), those used by the 1000 data packets are in Fig.5(b), those used by the 2000 data packets are in Fig.5(c), and those used by a total of 3000 data packets are in Fig.5(d). From Fig.5, it can be confirmed that our scheme enables the autonomous load-balancing transmission of data packets to two sinks using multiple routes.

Next, it was assumed that data packets were periodically transmitted from a total of 20 sens-or nodes placed in the set simulation area. In Fig.6, the transition of the delivery ratio of the total number of data packets transmitted from a total of 20 randomly selected

*evaluation node*

*evaluation node*

Fig. 5. Routes used by applying our scheme to the simulation model with two sinks

(c) 1 to 2000 data packets (d) 1 to 3000 data packets

As the first experiment on the simulation model with two sinks, it was assumed that the evaluation node marked in Fig.4 detected an abnormal value and transmitted the data packet with this abnormal value periodically. The routes used by applying our scheme are shown in Fig.5. Of the 3000 data packets transmitted from the evaluation node, the routes used by the first 500 data packets are illustrated in Fig.5(a), those used by the 1000 data packets are in Fig.5(b), those used by the 2000 data packets are in Fig.5(c), and those used by a total of 3000 data packets are in Fig.5(d). From Fig.5, it can be confirmed that our scheme enables the autonomous load-balancing transmission of data packets to two sinks using multiple ro-

Next, it was assumed that data packets were periodically transmitted from a total of 20 sens-or nodes placed in the set simulation area. In Fig.6, the transition of the delivery ratio of the total number of data packets transmitted from a total of 20 randomly selected

(a) 1 to 500 data packets (b) 1 to 1000 data packets

*evaluation node*

*evaluation node*

utes.

sensor nodes is shown, and the lifetime of the simulation model with two sinks, as in Fig.5, is compared. In Fig.6, the existing schemes in Ohtaki et al., 2006 and Utani et al., 2008, which belong to the category of ant-based routing algorithms, are denoted as *MUAA* and *AAR*, respectively. The existing scheme in Dubois-Ferriere et al., 2004 and Oyman and Ersoy, 2004, which describe representative data gathering for a wireless sensor network with multiple sinks, is denoted as *NS*. From Fig.6, it can be confirmed that our scheme denoted as *Proposal* in Fig.6 achieves a longer-term operation of a wireless sensor network with multiple sinks than the existing ones because it improves and balances the load of each sensor node by the communication load reduction for network control and the autonomous load-balancing data transmission. Through simulation experiments, it was verified that our scheme (Matsumoto et al., 2010) is substantially advantageous for the long-term operation of a large scale and dense wireless sensor network with multiple sinks.

Fig. 6. Transition of delivery ratio
