dc.citation.conferencePlace |
KO |
- |
dc.citation.conferencePlace |
제주도 |
- |
dc.citation.title |
2023년 대한전자공학회 하계종합학술대회 |
- |
dc.contributor.author |
Park, Keonhee |
- |
dc.contributor.author |
Lee, Kyuho Jason |
- |
dc.date.accessioned |
2024-01-31T18:38:05Z |
- |
dc.date.available |
2024-01-31T18:38:05Z |
- |
dc.date.created |
2023-07-24 |
- |
dc.date.issued |
2023-06-28 |
- |
dc.description.abstract |
Spiking Neural Network (SNN) draws attention for low energy consumption due to transmitting only binary information. Another way to implement low power hardware is Compute-in-Memory (CIM). It removes external memory access and minimizes data transaction energy. However, power-hungry Analog-to-Digital Converter (ADC) is essential in analog CIM. In time-domain computation, ADC is replaced with time-to-digital converter. Nevertheless, time-domain CIM supporting multi-bit computation has low energy efficiency by digital logic. To resolve this issue, this paper proposes time-domain SNN CIM with no digital circuit and it improves 83% throughput by pipelined architecture. It achieves 701.7 TOPS/W of energy efficiency. |
- |
dc.identifier.bibliographicCitation |
2023년 대한전자공학회 하계종합학술대회 |
- |
dc.identifier.uri |
https://scholarworks.unist.ac.kr/handle/201301/74675 |
- |
dc.publisher |
대한전자공학회 |
- |
dc.title |
Compute-in-Memory와 Time-Domain 연산을 활용한 Spiking Neural Network 가속기 |
- |
dc.type |
Conference Paper |
- |
dc.date.conferenceDate |
2023-06-28 |
- |