In this paper we estimate the end-to-end total bandwidth (BW) and available bandwidth (ABW) of a path between a pair of nodes in an IEEE 802.11b ad hoc network, both as functions of packet length, using dispersion traces between probing packet pairs of different lengths. The pairs of packets that suffer the minimum delay are used to estimate the maximum achievable transmission rate (BW), while the variability of the dispersions is used to estimate the fraction of that BW that is available for data transmission (ABW). To model the dependence of the variability of the dispersion traces on the ABW, we use a neuro-fuzzy estimator. The system is shown to be timely, accurate and efficient.