The paper is Subhajit Ganguly, "Relation Between Harappan And Brahmi Scripts" (copyright 2012 uploaded in 2017).
It is posted on the vixra site (an ungated pre-print archive mostly for non-professional researchers) which is notorious for crackpot postings, particularly in some of its forums such as those related to fundamental physics, but is not always so far off the mark in some of its other forums. Nothing on the vixra site, including this paper, should be taken as proven, but every once and a while one of its authors produce some interesting ideas like this one that deserve a little further attention. There are several other papers posted at the site by the same author on the same theme.
The abstract states:
Around 45 odd signs out of the total number of Harappan signs found make up almost 100 percent of the inscriptions, in some form or other, as said earlier. Out of these 45 signs, around 40 are readily distinguishable. These form an almost exclusive and unique set. The primary signs are seen to have many variants, as in Brahmi. Many of these provide us with quite a vivid picture of their evolution, depending upon the factors of time, place and usefulness. Even minor adjustments in such signs, depending upon these factors, are noteworthy. Many of the signs in this list are the same as or are very similar to the corresponding Brahmi signs. These are similarities that simply cannot arise from mere chance. It is also to be noted that the most frequently used signs in the Brahmi look so similar to the most frequent Harappan symbols. The Harappan script transformed naturally into the Brahmi, depending upon the factors channelizing evolution of scripts.
The Harappan script is last attested ca. 1900 BCE when that civilization collapsed and remains undeciphered with disputed theories regarding its origins and the origins of the language it describes. There are credible claims, however, that it continued to be used on a limited basis until 1500 BCE or even as late as 1100 BCE. There is serious dispute over whether the Harappan script codes a true language, or merely a proto-language made up mostly of trademarks, analogous to the Vinca script of the Neolithic Vinca culture in the Balkans. Per Wikipedia
A couple of credible academic claims have been made, however, that the script actually codes the Harappan language.An opposing hypothesis that has been offered by Michael Witzel and Steve Farmer, is that these [Harappan] symbols are nonlinguistic signs which symbolise families, clans, gods, and religious concepts—similar to components of coats of arms or totem poles. In a 2004 article, Farmer, Sproat, and Witzel presented a number of arguments stating that the Indus script is nonlinguistic, principal among them being the extreme brevity of the inscriptions, the existence of too many rare signs (increasing over the 700-year period of the Mature Harappan civilization), and the lack of the random-looking sign repetition typical of language.
The earliest attested version of the Brahmi script is roughly the 3rd century BCE and its origins are disputed among professionals, and it wouldn't take many undiscovered inscriptions to bridge the gap in an area where there is probably a significant amount remaining to be discovered in the archaeological record. The leading view is that the Brahmi script is ultimately derived from the Phoenician script that is the source of the Roman and Greek alphabets, for example, but this is not a consensus view and the argument for an indigenous origin of the script also has credible supporters within academic linguistics.