Searchers preparing to resume the underwater hunt for Malaysia Airlines Flight 370 increasingly suspect that some of the electronic signals detected last month didn't come from the jetliner's black-box flight recorders, a senior Australian naval officer said.

The doubts—based on further acoustic analysis of the transmissions by Australian authorities over recent weeks—represent another potential setback in the two-month-old operation. An initial underwater search in the southern Indian Ocean has already failed to find any sign of the missing plane, while a costly air-and-ship search of the ocean's surface turned up only garbage.

Authorities in April clung to hope that electronic transmissions picked up by Australian naval vessel ADV Ocean Shield on four occasions on April 5 and April 8 would provide a breakthrough in the search. But authorities are increasingly considering only the two transmissions on April 5 as relevant to the search, Australian naval Commander James Lybrand, captain of the Ocean Shield search vessel, said in an interview late last week. Further analysis of the streams of signals detected three days later on April 8 has led authorities to doubt that they were from a man-made device, Cmdr. Lybrand said.

Each of the transmissions on April 8 were intermittent and at a frequency of around 27 kHz—much lower than the 37.5 kHz frequency that beacons are designed to emit, and also lower than the 33.3 kHz frequency of other transmissions on April 5. "As far as frequency goes, between 33 kHz and 27 kHz is a pretty large jump," Cmdr. Lybrand said.

The Joint Agency Coordination Center, the Australian agency leading the search, didn't immediately respond to a request for comment on Cmdr. Lybrand's remarks.

Authorities continue to believe that the two April 5 signals—including one held for 2 hours and 20 minutes—are consistent with black-box locator beacons. The signals from that day were detected at a slightly lower frequency than locator beacons are designed to emit, but officials have said this anomaly could have been caused by weakening batteries and the vagaries of deep-sea conditions.

Two other signal streams picked up in the search area had earlier been eliminated as leads, further highlighting the pitfalls of tracking possible noises from locator beacons in the open sea. Signals picked up early in the search by British navy vessel HMS Echo were later shown to have been noises from the ship itself, while a detection from a sonar buoy dropped in the ocean in early April was later determined to have come from a passing commercial freighter.

Click for more from The Wall Street Journal