*4.3.4. Discussion*

88 Wireless Sensor Networks – Technology and Protocols

*4.3.2. Energy significant networks* 

data generation rate is from 0.0001 to 100.

**Figure 11.** Tradeoffs of energy significant

*4.3.3. Delay significant networks* 

generation rate is larger than 30.

the best when event generation rate is larger than 10.

From Fig. 10 we find that when the event generation rate is between 0.0001 and 4.0, nonaggregation is the best comparing with full and partial aggregation. When the data generation rate is between 4 and 30, the partial aggregation is the best, and the full

Here we discuss the case when energy is significant. The parameters are defined to be as below: α=1, β=2 and γ=1. According to proposed TOI we can get the best TOI values when

Fig. 11 shows the result. We find from the figure that in the region of data generation rate between 0.0001 and 4.0, the non-aggregation is the best TOI. When the data generation rate is about 4-10, the figure shows that the partial aggregation is the best; the full aggregation is

In delay significant networks, α, β and γ is defined as 1, 1, 2, as shown in Fig. 12. From the figure we find that when event generation rate is between 0.0001 and 4.0, the nonaggregation is investigated to be the best TOI; and when the event generation rate is from 4 to 30, the partial aggregation is the best; the full aggregation is the best TOI when event

aggregation is the best when data generation rate is larger than 30.

Let us summarize the data aggregation with best TOI according to different event generation rate in table 2. From the table we find that when event generation rate is small (0.0001-4.0, 6.0) the non-aggregation is the best TOI. Moreover, from the figures we find that in accuracy significant networks, the event data range of best TOI at non-aggregation is longer (0.0001-6.0) than any others. This is because, in non-aggregation, the data accuracy is 100%; and the other two have low data accuracy; when event generation rate is larger than 6, non-aggregation has very long delay because of the congestions around the sink node.

When event generation rate is moderate (4, 6-30), the partial aggregation is the best TOI except the case energy significance networks. In energy significance networks, the number of transmission in partial aggregation is much more than full aggregation, so the energy has great impact on partial aggregation with the significant of β=2. When data generation rate is large, due to the large number of transmission, the energy consumption is very high in nonaggregation and partial aggregation; therefore, the best TOI is the full aggregation in the networks with large event generation rate.


**Table 2.** The best data aggregation
