The increase in effectiveness of Japanese communications security practices in 1934 and 1935 succeeded in denying some communications intelligence information to the U.S., but the Navy continued to refine its techniques in the COMINT-COMSEC “battles” of the mid- and late 1930s.
Japanese naval operations associated with the Fleet Maneuvers of 1936 were once again minor compared to the very large exercises of 1930 and 1933. The Japanese were involved in occupation duty in China and their economy could not support extensive fleet exercises. Consequently, there was little opportunity for the U.S. Navy to intercept intra-fleet communications; all the information about the exercise came from the radio traffic from the shore stations and the flagships.
Despite the situation, the COMINT organizations were able through traffic analysis to locate the main body of the Japanese fleet, to predict accurately proposed fleet movements and to provide CINCAF with 15 to 30 days warning of the Japanese fleet mobilization.
Some of the problems associated with exploitation of the exercise are best illustrated by the following report from the U.S. Navy intercept station in the Philippines:
A sufficient number of calls were identified to enable this activity to keep an almost constant check on Fleet maneuvering activity. However, during the period of this maneuver the value of secret calls was continually brought to (our) attention. Following a Fleet that is using secret calls is an entirely different one from following a Fleet using service calls of common knowledge. It was not found possible for the average operator to identify secret calls of force afloat from key notes or operator characteristics, and it is the firm conviction of this activity that much that has been said in the past relative to the ease with which secret calls can be identified is in error. Of course if the transmitting operator slipped and mixed in service calls, identifications were instantaneous but such errors were rare. Such identifications as were made of individual vessels were obtained after many hours of research and checking. It was noted that one or more units of a force would shift calls and the remaining units therein would not shift. It was proven beyond any doubt that Orange uses secret calls only when engaging in tactical exercises. When the Fleet cruises, service calls are used… No identifications what so ever were made of the secret addresses.
The U.S. Navy COMINT analysts had not solved the entire Japanese call sign or address system in 1936, but they were able to identify some of the callsigns and associate them with a particular ship at times. The main source of information for the U.S. Navy was Japanese operator compromise. Other techniques used by the Navy were association of a secret and a service call on the same frequency, unusual procedures associated with a particular ship, reuse of a serial number suffix, collective calls answered in error, and the volume of traffic originated and handled by the flagships.
Despite the mixed success of these rudimentary traffic-analytic techniques, the Navy learned the value of keeping detailed records of traffic. Its final report of the 1936 maneuvers included a description of each communications circuit under the following parameters: frequency usage, procedures, schedules, use of code/cipher, relays, hearability, use of power, operator expertise, method of U.S. reception and quality, callsign identification, chronological description of the communications system and the activity of the correspondents. The Navy drew very complete net diagrams and published and disseminated them to all the COMINT units in the Asiatic Command.
The final report observed that a “maneuvering code separator” between message groups indicated a different key in enciphering. The report also displayed the kinds of messages sent, a study of message volumes, a study of the message serial number system, and a compilation of weekly traffic volumes by station and by frequency. The Navy continued the development of TA techniques right up to the start of World War II. However, more significant and certainly more crucial advances would occur during the war itself.
American radio traffic analysis materialized during the Navy’s exploitation of the Japanese fleet exercises of the 1920s and 1930s. Its COMINT organization was at war with the Japanese naval communicators during those years of challenge in the Pacific. The absence of readily available decryptions of Japanese enciphered traffic was the impetus for traffic-analytic developments.
The U.S. Navy acknowledged the value of timely combat intelligence and discerned that radio traffic analysis was the main source of timely intelligence about operational Japanese fleet activities. The groundwork was laid in the period between the wars for the vital contribution of traffic analysis to the eventual victory in the Pacific.