With the rapid development of the Internet of Things, billions of devices should be available for sensing environment variables and reporting events that may be happened. These devices are composed as wireless sensor networks (WSN) for supporting real-world applications. Typically, multiple attributes are being sensed simultaneously for enabling the event detection. In this setting, query processing is a challenge especially when multiple attributes are necessary to be sensed in a certain region interested. To remedy this issue, we propose to construct an indexing tree for facilitating query processing. Specifically, the network region is represented in terms of grid cells. An indexing tree is constructed considering the energy consumption of exchanging the same size of message between neighboring grid cells or sub-regions. Generally, queries in WSN can be classified into different types, and query mechanisms have been proposed for processing these types of queries. The evaluation result shows that our technique is more energy efficient when the more number of queried attributes are considered.