Journal article Open Access

Compressed sensing with approximate message passing using in-memory computing

Le Gallo, Manuel; Sebastian, Abu; Cherubini, Giovanni; Giefers, Heiner; Eleftheriou, Evangelos


MARC21 XML Export

<?xml version='1.0' encoding='UTF-8'?>
<record xmlns="http://www.loc.gov/MARC21/slim">
  <leader>00000nam##2200000uu#4500</leader>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">Approximate message passing</subfield>
  </datafield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">Compressed sensing</subfield>
  </datafield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">In-memory computing</subfield>
  </datafield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">Phase-change memory</subfield>
  </datafield>
  <controlfield tag="005">20200120161034.0</controlfield>
  <controlfield tag="001">3249877</controlfield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">IBM Research - Zurich, 8803 Rüschlikon, Switzerland</subfield>
    <subfield code="a">Sebastian, Abu</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">IBM Research - Zurich, 8803 Rüschlikon, Switzerland</subfield>
    <subfield code="a">Cherubini, Giovanni</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="a">Giefers, Heiner</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="a">Eleftheriou, Evangelos</subfield>
  </datafield>
  <datafield tag="856" ind1="4" ind2=" ">
    <subfield code="s">586406</subfield>
    <subfield code="z">md5:77eb782ef6baab0f8b8ae06eb59a73fa</subfield>
    <subfield code="u">https://zenodo.org/record/3249877/files/LeGallo18.pdf</subfield>
  </datafield>
  <datafield tag="542" ind1=" " ind2=" ">
    <subfield code="l">open</subfield>
  </datafield>
  <datafield tag="260" ind1=" " ind2=" ">
    <subfield code="c">2018-08-29</subfield>
  </datafield>
  <datafield tag="909" ind1="C" ind2="O">
    <subfield code="p">openaire</subfield>
    <subfield code="o">oai:zenodo.org:3249877</subfield>
  </datafield>
  <datafield tag="909" ind1="C" ind2="4">
    <subfield code="c">4304-4312</subfield>
    <subfield code="n">10</subfield>
    <subfield code="p">IEEE Transactions on Electron Devices</subfield>
    <subfield code="v">65</subfield>
  </datafield>
  <datafield tag="100" ind1=" " ind2=" ">
    <subfield code="u">IBM Research - Zurich, 8803 Rüschlikon, Switzerland</subfield>
    <subfield code="a">Le Gallo, Manuel</subfield>
  </datafield>
  <datafield tag="245" ind1=" " ind2=" ">
    <subfield code="a">Compressed sensing with approximate message passing using in-memory computing</subfield>
  </datafield>
  <datafield tag="536" ind1=" " ind2=" ">
    <subfield code="c">682675</subfield>
    <subfield code="a">PROJECTED MEMRISTOR: A nanoscale device for cognitive computing</subfield>
  </datafield>
  <datafield tag="536" ind1=" " ind2=" ">
    <subfield code="c">780215</subfield>
    <subfield code="a">Computation-in-memory architecture based on resistive devices</subfield>
  </datafield>
  <datafield tag="540" ind1=" " ind2=" ">
    <subfield code="u">https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode</subfield>
    <subfield code="a">Creative Commons Attribution Non Commercial No Derivatives 4.0 International</subfield>
  </datafield>
  <datafield tag="650" ind1="1" ind2="7">
    <subfield code="a">cc-by</subfield>
    <subfield code="2">opendefinition.org</subfield>
  </datafield>
  <datafield tag="520" ind1=" " ind2=" ">
    <subfield code="a">&lt;p&gt;In-memory computing is a promising non-von Neumann approach where certain computational tasks are performed within resistive memory units by exploiting their physical attributes. In this paper, we propose a new method for fast and robust compressed sensing of sparse signals with approximate message passing recovery using in-memory computing. The measurement matrix for compressed sensing is encoded in the conductance states of resistive memory devices organized in a crossbar array. This way, the matrix-vector multiplications associated with both the compression and recovery tasks can be performed by the same crossbar array without intermediate data movements at potential O(1) time complexity. For a signal of size N, the proposed method achieves a potential O(N)-fold recovery complexity reduction compared with a standard software approach. We show the array-level robustness of the scheme through large-scale experimental demonstrations using more than 256k phase-change memory devices.&lt;/p&gt;</subfield>
  </datafield>
  <datafield tag="024" ind1=" " ind2=" ">
    <subfield code="a">10.1109/TED.2018.2865352</subfield>
    <subfield code="2">doi</subfield>
  </datafield>
  <datafield tag="980" ind1=" " ind2=" ">
    <subfield code="a">publication</subfield>
    <subfield code="b">article</subfield>
  </datafield>
</record>
23
54
views
downloads
Views 23
Downloads 54
Data volume 31.7 MB
Unique views 22
Unique downloads 53

Share

Cite as