scispace - formally typeset
Proceedings ArticleDOI

A 6.5-mW 5-Gbps On-Chip Differential Transmission Line Interconnect with a Low-Latency Asymmetric Tx in a 180nm CMOS Technology

Reads0
Chats0
TLDR
In this paper, an on-chip differential transmission-line (DTL) interconnect is proposed to reduce delay and power consumption in long global interconnects, which can transmit signals at near light-of-speed with small power dissipation of Tx.
Abstract
This paper proposes an on-chip differential- transmission-line (DTL) interconnect to reduce delay and power consumption in long global interconnects The DTL interconnect can transmit signals at near light-of-speed with small power dissipation of Tx The proposed DTL interconnect consists of Tx, DTL and Rx, and an asymmetric Tx is employed to reduce offset delay in Tx In the measurement result, 5 Gbps signal transmission can be achieved through 3 mm-length interconnect, and delay and total power consumption are 140 ps and 65 mW, respectively A 180 nm standard CMOS process was utilized Figure of merit (FoM) for on-chip interconnects is proposed to evaluate delay and power consumption The proposed DTL interconnect is compared with the conventional RC, DTL and optical interconnects, and it achieves the highest FoM at more than 5 mm length

read more

Citations
More filters
Journal ArticleDOI

A Bidirectional- and Multi-Drop-Transmission-Line Interconnect for Multipoint-to-Multipoint On-Chip Communications

TL;DR: A novel transceiver that consists of a single differential-amplifier and serves as both a transmitter and a receiver for transmitting signals to multipoint for on-chip high-speed networks that have big impact in chip performances is proposed.
Proceedings Article

A Bidirectional-and Multi-Drop-Transmission-Line Interconnect for Multipoint-to-Multipoint On-Chip Communications

TL;DR: In this paper, a bidirectional and multi-drop-transmission-line interconnect for on-chip high-speed networks that have big impact in chip performances is investigated.
Journal ArticleDOI

A Q-Learning Based Self-Adaptive I/O Communication for 2.5D Integrated Many-Core Microprocessor and Memory

TL;DR: Simulation results show that the proposed adaptive 2.5D I/Os (in 65 nm CMOS) can achieve an average of 12.5 mW I/O power, 4 GHz bandwidth and 3.125 pJ/bit energy efficiency for one channel under 10-6 BER.
Proceedings ArticleDOI

Reinforcement learning based self-adaptive voltage-swing adjustment of 2.5D I/Os for many-core microprocessor and memory communication

TL;DR: An online reinforcement Q-learning algorithm is developed to perform a self-adaptive voltage-swing control of 2.5D through-silicon interposer (TSI) I/O circuits for energy-efficient communication between many-core microprocessor and memory.
Proceedings ArticleDOI

An energy-efficient 2.5D through-silicon interposer I/O with self-adaptive adjustment of output-voltage swing

TL;DR: Experimental results show that the adaptive 2.5D TSI I/Os designed in 65nm CMOS can achieve an average of 13mW I/O power, 4GHz bandwidth and 3.25pJ=bit energy efficiency for one channel under 10-6 BER, which has 21.42% reduction of power and 14.47% energy efficiency improvement.
References
More filters
Journal ArticleDOI

Optical interconnections for VLSI systems

TL;DR: The combination of decreasing feature sizes and increasing chip sizes is leading to a communication crisis in the area of VLSI circuits and systems, and the possibility of applying optical and electrooptical technologies to such interconnection problems is investigated.
Journal ArticleDOI

Intra-chip wireless interconnect for clock distribution implemented with integrated antennas, receivers, and transmitters

TL;DR: This is the first known demonstration of an on-chip clock transmitter with an integrated antenna and the second demonstration of a clock receiver with anIntegrated antenna, where the receiver's frequency and interconnection distance have approximately been doubled over previous results.
Journal ArticleDOI

Multilevel metal capacitance models for CAD design synthesis systems

TL;DR: An empirical model for multilevel interconnect capacitance that allows designers to compute capacitances of arbitrary complex metal geometries by a novel strategy of constructing complex geometry from simple primitive cells is presented.
Journal ArticleDOI

Near speed-of-light signaling over on-chip electrical interconnects

TL;DR: In this article, the propagation limits of electrical signals for systems built with conventional silicon processing are explored and a design which takes advantage of the inductance-dominated high-frequency regime of on-chip interconnect is shown capable of transmitting data at velocities near the speed of light.
Proceedings ArticleDOI

A 51mW 1.6GHz on-chip network for low-power heterogeneous SoC platform

TL;DR: A 1.6GHz on-chip network integrating two processors, memories, and an FPGA provides 11.2GB/s bandwidth in 0.18/spl mu/m 6M CMOS technology supporting globally asynchronous, locally synchronous mode and programmable clocking.
Related Papers (5)