Localization systems for wireless sensor networks are complex systems that have many applications in real life. In this dissertation, an efficient localization system tailored for wireless sensor networks is proposed. The system is comprised of four subparts that work together to perform target localizations. The first subpart is a node localization system (TALS) that is capable of generating estimates of the nodes' location based on ranging measurements and anchor nodes. TALS utilizes trigonometric properties and identities to compute the nodes' locations with low computational overhead and high accuracy. Moreover, TALS takes advantage of redundancy to employ data fusion techniques to enhance the quality of the estimate.

The second subpart is a medium access control (MAC) that is capable of handling high data traffic when an event occurs in the sensor field. EB-MAC and EVA-MAC are two medium access controls that are tailored for event based systems that characterizes the target localization application. EB-MAC and EVA-MAC share same principles but different implementations where the latter is the enhanced version of the former. The basic idea is to shift contention generated from the data phase by introducing a control gap thus achieving a collision free schedule for data transmission. The third subpart is the acoustic target localization which triangulates potential targets based on acoustic sources. The 7 point trilateration technique estimates the target's location based on time difference of arrival. Finally, the data fusion subpart takes advantage of redundancy to enhance the quality of the estimate. The FuzzyART data fusion technique assigns probabilistic weights according to a decision tree based on spatial correlation and consensus vote.

Extensive simulations in software were performed for each subpart where comparisons with popular techniques were performed. Enhancements in accuracy, speed and computational overhead were achieved. Furthermore, a hardware implementation of the acoustic target localization system was performed on MicaZ motes and PIC microcontroller demo boards acting as a co-processor. The system presented robustness to noisy measurements, fault tolerance and high accuracy.