VLSI Design of Approximate Message Passing for Signal Restoration and Compressive Sensing

• Published in: IEEE journal on emerging and selected topics in circuits and systems Vol. 2; no. 3; pp. 579 - 590
• Main Authors: Maechler, P., Studer, C., Bellasi, D. E., Maleki, A., Burg, A., Felber, N., Kaeslin, H., Baraniuk, R. G.
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.09.2012; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Approximate message passing (AMP); Approximation algorithms; Complexity theory; compressive sensing (CS); Computer architecture; ell_{1} -norm minimization; fast discrete cosine transform (DCT); field-programmable gate array (FPGA); Matching pursuit algorithms; Semiconductors; Signal restoration; Sparse matrices; sparse signal recovery; Very large scale integration; very-large scale integration (VLSI)
• ISSN: 2156-3357; 2156-3365; 2156-3365
• DOI: 10.1109/JETCAS.2012.2214636

Sparse signal recovery finds use in a variety of practical applications, such as signal and image restoration and the recovery of signals acquired by compressive sensing. In this paper, we present two generic very-large-scale integration (VLSI) architectures that implement the approximate message passing (AMP) algorithm for sparse signal recovery. The first architecture, referred to as AMP-M, employs parallel multiply-accumulate units and is suitable for recovery problems based on unstructured (e.g., random) matrices. The second architecture, referred to as AMP-T, takes advantage of fast linear transforms, which arise in many real-world applications. To demonstrate the effectiveness of both architectures, we present corresponding VLSI and field-programmable gate array implementation results for an audio restoration application. We show that AMP-T is superior to AMP-M with respect to silicon area, throughput, and power consumption, whereas AMP-M offers more flexibility.