This paper presents pathChirp, a new active probing tool for estimating available bandwidth on a communication network path. Based on the concept of "self-induced congestion," pathChirp features an exponential flight pattern of probes called a chirp. Packet chirps offer several advantages over current probing schemes based on packet pairs or packet trains. By rapidly increasing the probing rate within each chirp, pathChirp obtains a rich set of information to dynamically estimate available bandwidth. Since it uses only packet interarrival times for estimation, pathChirp does not require synchronous or highly stable clocks at the sender and receiver. The paper tests pathChirp with simulations and Internet experiments, finding that it provides good estimates of available bandwidth while using only a fraction of the number of probe bytes used by current state-of-the-art techniques. PathChirp estimates available bandwidth by launching a number of packet chirps from the sender to the receiver and then conducting a statistical analysis at the receiver. It uses the shape of the queuing delay signature to make an estimate of the per-packet available bandwidth. It then takes a weighted average of the estimates corresponding to each chirp to obtain estimates of the per-chirp available bandwidth. Finally, it makes estimates of the available bandwidth by averaging the estimates obtained in the time interval. PathChirp segments each signature into regions belonging to excursions and regions not belonging to excursions. Based on the principle of self-induced congestion, it assumes that increasing queuing delays signify less available bandwidth than the instantaneous chirp rate at that moment while decreasing delays signify the opposite. It uses the relative queuing delay within a chirp to detect excursions and avoids using hard queuing delay thresholds. PathChirp's performance is analyzed in simulations and experiments. It is compared with existing tools such as TOPP and pathload. The results show that pathChirp outperforms these tools in terms of estimation accuracy and efficiency. It is also shown to be robust in multi-hop scenarios. The paper concludes that pathChirp is a useful tool for estimating available bandwidth, especially for applications requiring rapid estimates with a light probing load. The pathChirp tool is available as open-source freeware at spin.rice.edu/Software/pathChirp.This paper presents pathChirp, a new active probing tool for estimating available bandwidth on a communication network path. Based on the concept of "self-induced congestion," pathChirp features an exponential flight pattern of probes called a chirp. Packet chirps offer several advantages over current probing schemes based on packet pairs or packet trains. By rapidly increasing the probing rate within each chirp, pathChirp obtains a rich set of information to dynamically estimate available bandwidth. Since it uses only packet interarrival times for estimation, pathChirp does not require synchronous or highly stable clocks at the sender and receiver. The paper tests pathChirp with simulations and Internet experiments, finding that it provides good estimates of available bandwidth while using only a fraction of the number of probe bytes used by current state-of-the-art techniques. PathChirp estimates available bandwidth by launching a number of packet chirps from the sender to the receiver and then conducting a statistical analysis at the receiver. It uses the shape of the queuing delay signature to make an estimate of the per-packet available bandwidth. It then takes a weighted average of the estimates corresponding to each chirp to obtain estimates of the per-chirp available bandwidth. Finally, it makes estimates of the available bandwidth by averaging the estimates obtained in the time interval. PathChirp segments each signature into regions belonging to excursions and regions not belonging to excursions. Based on the principle of self-induced congestion, it assumes that increasing queuing delays signify less available bandwidth than the instantaneous chirp rate at that moment while decreasing delays signify the opposite. It uses the relative queuing delay within a chirp to detect excursions and avoids using hard queuing delay thresholds. PathChirp's performance is analyzed in simulations and experiments. It is compared with existing tools such as TOPP and pathload. The results show that pathChirp outperforms these tools in terms of estimation accuracy and efficiency. It is also shown to be robust in multi-hop scenarios. The paper concludes that pathChirp is a useful tool for estimating available bandwidth, especially for applications requiring rapid estimates with a light probing load. The pathChirp tool is available as open-source freeware at spin.rice.edu/Software/pathChirp.