We measured the client connection count for each AS using PrivCount. We used the IPv4 and IPv6 datasets (dated 26th November 2017) from CAIDA  to map client IP addresses to its autonomous system (AS). We ran the measurement in two phases, each consisting of four consecutive one-day measurements, and sum the measurements over all days. In the first phase, beginning on 2018-06-22, we included all ASes (59,597 in total), and in the second phase, beginning on 2018-06-27, we included only those 1,653 ASes that did not have zero in their confidence interval in at least one of the days in the first phase. Thus the first phase measurement acted as a filter to reduce the ASes under consideration to those with likely positive true counts.
Contains the list of all ASNs in the IPv4 and IPv6 datasets (dated 26th November 2017) from CAIDA .
Contains the first phase results - the sum of per-AS client connection counts for those 1,653 ASes.
Contains the second phase results - the sum of per-AS client connection counts for those 1,653 ASes.
Average the connection counts for the 1,653 ASes across all eight days. The new standard deviation can be computed as:
For more details about the experiments, refer:
Understanding Tor Usage with Privacy-Preserving Measurement ACM Internet Measurement Conference (IMC), October 2018 Akshaya Mani*, T Wilson Brown*, Rob Jansen, Aaron Johnson, and Micah Sherr (* co-first authors)
 Rob Jansen and Aaron Johnson, "Safely Measuring Tor", ACM CCS 2016.
 Routeviews Prefix-to-AS mappings (pfx2as) for IPv4 and IPv6. http://data.caida.org/datasets/routing/routeviews-prefix2as/.