18392161. Matrix-Vector Multiplications Based on Charge-Summing Memory Cell Strings simplified abstract (IMEC VZW)
Contents
Matrix-Vector Multiplications Based on Charge-Summing Memory Cell Strings
Organization Name
Inventor(s)
Jonas Doevenspeck of Herent (BE)
Maarten Rosmeulen of Gent (BE)
Stefan Cosemans of Leuven (BE)
Matrix-Vector Multiplications Based on Charge-Summing Memory Cell Strings - A simplified explanation of the abstract
This abstract first appeared for US patent application 18392161 titled 'Matrix-Vector Multiplications Based on Charge-Summing Memory Cell Strings
The abstract describes an example embodiment of a patent application related to matrix-vector multiplications using charge-summing memory cell strings. The patent application involves an in-memory compute device that performs analog multiply-and-accumulate operations on data inputs and weight inputs.
- String of serially connected memory cells over a semiconductor channel structure
- Source junction connectible to one end of the memory cell string via a string select switch
- Readout circuit with a sense node connectible to one end of the memory cell string via a charge transfer switch
- Control circuitry that applies pass mode signals, data input signals, and stop signals sequentially to each memory cell along the string
- Control circuitry enables the string select switch and the charge transfer switch
Potential Applications: - Artificial intelligence - Machine learning - Signal processing - Data analytics - Edge computing
Problems Solved: - Efficient matrix-vector multiplications - Analog multiply-and-accumulate operations - Reduced power consumption - Improved performance in computing tasks
Benefits: - Faster processing speeds - Lower energy consumption - Enhanced accuracy in calculations - Compact and scalable design - Cost-effective implementation
Commercial Applications: Title: "In-Memory Compute Device for Analog Multiply-and-Accumulate Operations" This technology could be utilized in various industries such as: - Semiconductor manufacturing - Cloud computing - IoT devices - Autonomous vehicles - Robotics
Questions about the technology: 1. How does this in-memory compute device improve efficiency in matrix-vector multiplications? 2. What are the key advantages of using charge-summing memory cell strings for analog operations?
Original Abstract Submitted
Example embodiments relate to matrix-vector multiplications based on charge-summing memory cell strings. An example in-memory compute device for performing analog multiply-and-accumulate operations on a set of data inputs and a set of weight inputs includes a string of serially connected memory cells formed over a semiconductor channel structure, a source junction controllably connectible to one end of the string of memory cells via a string select switch, a readout circuit including a sense node controllably connectible to one end of the string of memory cells via a charge transfer switch, and control circuitry. The control circuitry is configured to apply pass mode signals, data input signals, and stop signals sequentially according to each memory cell's position along a string. The control circuitry is also configured to enable the string select switch and the charge transfer switch.
