This paper presents the Timing-sync Protocol for Sensor Networks (TPSN), a time synchronization protocol designed for wireless ad-hoc sensor networks. TPSN aims to provide network-wide time synchronization by establishing a hierarchical structure and performing pairwise synchronization along the edges of this structure. The algorithm works in two steps: first, a hierarchical structure is established in the network, and then pairwise synchronization is performed to establish a global timescale throughout the network. Eventually, all nodes in the network synchronize their clocks to a reference node. TPSN is compared to Reference Broadcast Synchronization (RBS), a timing synchronization algorithm for sensor networks based on receiver-receiver synchronization. The paper shows that TPSN provides roughly a 2x better performance than RBS. TPSN can synchronize a pair of motes to an average accuracy of less than 20 µs and a worst-case accuracy of around 50 µs. The synchronization accuracy does not degrade significantly with the increase in the number of nodes being deployed, making TPSN completely scalable. The paper also discusses the error analysis of TPSN, showing that TPSN provides a simple, scalable, and efficient solution to the problem of timing synchronization in sensor networks. TPSN is completely flexible and can be easily tuned to meet the desired levels of accuracy as well as algorithmic overhead. TPSN also comes with an auxiliary benefit of improving the accuracy of other basic services in sensor networks such as localization, target tracking, and aggregation. The paper describes the implementation of TPSN on Berkeley motes, showing that it can synchronize a pair of neighboring motes to an average accuracy of less than 20 µs. The paper also discusses the results of synchronization error for TPSN and RBS, showing that TPSN provides a 2x better performance than RBS. The synchronization error is measured by observing the phase shift between the two waveforms corresponding to the two motes. The results show that TPSN can synchronize a pair of motes to an average accuracy of less than 20 µs and a worst-case accuracy of around 50 µs. The paper also discusses the results of multihop results, showing that the error does not blow up with hop distance. In fact, for this scenario, the error almost becomes a constant beyond 3-hop distance. The synchronization error between any pair of motes will probabilistically take different values from the normal distribution obtained in the earlier section. The randomness in the sign as well as the magnitude of the synchronization error and drift prevents the error from blowing up. If all the motes have been drifting in the same direction and the error was a deterministic quantity, the error would have blown up with the number of hops. However, the probability of that event is very low. In fact, we have observed instances where the error value was large. To present a clear picture, we detail the obtained statistics in Table 2. WeThis paper presents the Timing-sync Protocol for Sensor Networks (TPSN), a time synchronization protocol designed for wireless ad-hoc sensor networks. TPSN aims to provide network-wide time synchronization by establishing a hierarchical structure and performing pairwise synchronization along the edges of this structure. The algorithm works in two steps: first, a hierarchical structure is established in the network, and then pairwise synchronization is performed to establish a global timescale throughout the network. Eventually, all nodes in the network synchronize their clocks to a reference node. TPSN is compared to Reference Broadcast Synchronization (RBS), a timing synchronization algorithm for sensor networks based on receiver-receiver synchronization. The paper shows that TPSN provides roughly a 2x better performance than RBS. TPSN can synchronize a pair of motes to an average accuracy of less than 20 µs and a worst-case accuracy of around 50 µs. The synchronization accuracy does not degrade significantly with the increase in the number of nodes being deployed, making TPSN completely scalable. The paper also discusses the error analysis of TPSN, showing that TPSN provides a simple, scalable, and efficient solution to the problem of timing synchronization in sensor networks. TPSN is completely flexible and can be easily tuned to meet the desired levels of accuracy as well as algorithmic overhead. TPSN also comes with an auxiliary benefit of improving the accuracy of other basic services in sensor networks such as localization, target tracking, and aggregation. The paper describes the implementation of TPSN on Berkeley motes, showing that it can synchronize a pair of neighboring motes to an average accuracy of less than 20 µs. The paper also discusses the results of synchronization error for TPSN and RBS, showing that TPSN provides a 2x better performance than RBS. The synchronization error is measured by observing the phase shift between the two waveforms corresponding to the two motes. The results show that TPSN can synchronize a pair of motes to an average accuracy of less than 20 µs and a worst-case accuracy of around 50 µs. The paper also discusses the results of multihop results, showing that the error does not blow up with hop distance. In fact, for this scenario, the error almost becomes a constant beyond 3-hop distance. The synchronization error between any pair of motes will probabilistically take different values from the normal distribution obtained in the earlier section. The randomness in the sign as well as the magnitude of the synchronization error and drift prevents the error from blowing up. If all the motes have been drifting in the same direction and the error was a deterministic quantity, the error would have blown up with the number of hops. However, the probability of that event is very low. In fact, we have observed instances where the error value was large. To present a clear picture, we detail the obtained statistics in Table 2. We