Reading the Malaysian PM's statement as a whole, the thing that stands out is that it is carefully crafted to address an obvious question about the Malaysian government's handling of the matter:
Why did Malaysia fail to take advantage of civilian and military primary radar data that were readily available from the beginning?
The implied answer asserted by the PM could be paraphrased as follow:
- Malaysian government did (ultimately) notice that an aircraft had flown across peninsular Malaysia,
- The primary radar track was not connected with MH370 at first due to the absence of transponder data.
- The primary radar track was only connected with MH370 when the relevance of Satcom ping information was appreciated some days later.
Current attempts to pinpoint the location of the aircraft based on extrapolation from Inmarsat ping communications should be seen against this background, and be taken with a large pinch of salt.
It is tempting to overestimate the degree to which Inmarsat data can be relied on, all the more tempting as we have nothing else in the public domain. However, as pointed out by earlier contributors, estimating location from radio signal strength cannot give more than a very rough indication of range. It may be worth the effort and cost to conduct a full-fledged trial to confirm assumptions about signal strength etc. given the actual type of aircraft, antenna, flight attitude, etc.
More sophisticated techniques based on signal transit times etc. are more promising in theory, but the Inmarsat protocols are not primarily designed for this purpose. Transponders on the satellite may not have measured/collected/downlinked timing data except as necessary for link establishment and maintenance. TDMA related data may be most promising to the extent it remains extant.
As another contributor has noted, it is possible that ACARS also sent out pings on VHF which might have been received while MH370 was (again) in the vicinity of land (Malaysia or Indonesia). Space-based SIGINT may also be of assistance, but we will not hear about such efforts.
Obviously, contemporaneous visual satellite imagery would be the the easiest way to spot an aircraft in the middle of the Indian Ocean.