Alert button

A Homogeneous Processing Fabric for Matrix-Vector Multiplication and Associative Search Using Ferroelectric Time-Domain Compute-in-Memory

Sep 24, 2022
Figure 1 for A Homogeneous Processing Fabric for Matrix-Vector Multiplication and Associative Search Using Ferroelectric Time-Domain Compute-in-Memory
Figure 2 for A Homogeneous Processing Fabric for Matrix-Vector Multiplication and Associative Search Using Ferroelectric Time-Domain Compute-in-Memory
Figure 3 for A Homogeneous Processing Fabric for Matrix-Vector Multiplication and Associative Search Using Ferroelectric Time-Domain Compute-in-Memory
Figure 4 for A Homogeneous Processing Fabric for Matrix-Vector Multiplication and Associative Search Using Ferroelectric Time-Domain Compute-in-Memory

Share this with someone who'll enjoy it:

In this work, we propose a ferroelectric FET(FeFET) time-domain compute-in-memory (TD-CiM) array as a homogeneous processing fabric for binary multiplication-accumulation (MAC) and content addressable memory (CAM). We demonstrate that: i) the XOR(XNOR)/AND logic function can be realized using a single cell composed of 2FeFETs connected in series; ii) a two-phase computation in an inverter chain with each stage featuring the XOR/AND cell to control the associated capacitor loading and the computation results of binary MAC and CAM are reflected in the chain output signal delay, illustrating full digital compatibility; iii) comprehensive theoretical and experimental validation of the proposed 2FeFET cell and inverter delay chains and their robustness against FeFET variation; iv) the homogeneous processing fabric is applied in hyperdimensional computing to show dynamic and fine-grain resource allocation to accommodate different tasks requiring varying demands over the binary MAC and CAM resources.

* 8 pages, 8 figures  

Share this with someone who'll enjoy it: