Search results

The purpose of this paper is to explore and evaluate the performance comparison of carbon nanotubes (CNT) and nickel silicide (NiSi) nanowires interconnects as prospective alternatives to copper wire interconnects.

The increasing resistivity of the copper wire with scaling and rising demands on current density drives the need for identifying new wiring solutions. This paper explores the various alternatives to copper. The metallic bundle CNTs and NiSi nanowires are promising candidates that can potentially address the challenges faced by copper. This paper analyzes various electrical models of carbon nanotube and recently introduced novel interconnect solution using NiSi nanowires.

The theoretical studies proves CNTs and NiSi nanowires to be better alternatives against copper on the ground of performance parameters, such as effective current density, delay and power consumption. NiSi nanowire provides highest propagation speed for short wire length, and copper is the best for intermediate wire length, while bundle CNTs is faster for long wire length. NiSi nanowire has lowest power consumption than copper and CNTs.

This paper investigates, assess and compares the performance of carbon nanotubes (CNT) and NiSi nanowires interconnects as prospective alternatives to copper wire interconnects in future VLSI chips.

The purpose of this paper is to analyze and compare the characteristics of hybrid conventional complementary metal oxide semiconductor/magnetic tunnel junction (CMOS/MTJ) logic gates based on spin transfer torque (STT) and differential spin Hall effect (DSHE) magnetic random access memory (MRAM).

Spintronics technology can be used as an alternative to CMOS technology as it is having comparatively low power dissipation, non-volatility, high density and high endurance. MTJ is the basic spin based device that stores data in form of electron spin instead of charge. Two mechanisms, namely, STT and SHE, are used to switch the magnetization of MTJ.

It is observed that the power consumption in DSHE based logic gates is 95.6% less than the STT based gates. DSHE-based write circuit consumes only 5.28 fJ energy per bit.

This paper describes how the DSHE-MRAM is more effective for implementing logic circuits in comparison to STT-MRAM.

The purpose of this paper is to explore the functioning of very‐large‐scale integration (VLSI) interconnects and modeling of interconnects and evaluate different approaches of testing interconnects.

In the past, on‐chip interconnect wires were not considered in circuit analysis except in high precision analysis. Wiring‐up of on‐chip devices takes place through various conductors produced during fabrication process. The shrinking size of metal‐oxide semiconductor field effect transistor devices is largely responsible for growth of VLSI circuits. With deep sub‐micron (DSM) technology, the interconnect geometry is scaled down for high wiring density. The complex geometry of interconnects and high operational frequency introduce wire parasitics and inter‐wire parasitics. These parasitics causes delay, power dissipation, and crosstalk that may affect the signal integrity in VLSI system. Accurate analysis, sophisticated design, and effective test methods are the requirement to ensure the proper functionality and reliability of VLSI circuits. The testing of interconnect is becoming important and a challenge in the current technology.

The effects of interconnect on signal integrity, power dissipation, and delay emerges significantly in DSM technology. For proper performance of the circuit, testing of interconnect is important and emerging challenge in the nanotechnology era. Although some work has been done for testing of interconnect, however, it is still an open area to test the parasitics effects of VLSI/ultra‐large‐scale integration interconnects. Efforts are required to analyze and to develop test methods for crosstalk, delay and power dissipation in current technology with solutions to minimize this effect.

This paper reviews the functioning of VLSI interconnects from micrometer to nanometer technology. The development of various interconnect models from simple short circuit to latest resistance inductance capacitance transmission line model are discussed. Furthermore, various methodologies such as built‐in self test and other techniques for testing interconnect for crosstalk and delay are discussed.

This paper aims to reduce the worst-case crosstalk effects for resistance, inductance and capacitance (RLC) interconnects using the bus encoding technique. In current nanoscale technology, power dissipation, propagation delay and crosstalk performance of interconnects determine the overall performance of a chip. Signal integrity issues due to crosstalk in the form of voltage glitches, overshoots, undershoots, undesirable noise, propagation speed ups and downs, etc. are some of the major deterrents for high-performance RLC modelled (VLSI) interconnects. This research paper primarily proposes two novel encoding methods (I and II) for RLC modelled interconnects to reduce the effect of crosstalk, simultaneous switching noise (SSN) and power consumption.

The proposed methods are based on the bus encoding method that is effective and well-suited for the reduction of the crosstalk noise. This method encodes or transforms incoming data in a manner that encoded data contain minimum or no crosstalk effects. The proposed encoding method uses the bus invert (BI) method. The proposed encoding methods are able to avoid the worst-case crosstalks while consuming lesser power during transmission in VLSI interconnects.

It is observed that the proposed encoders reduced/eliminated the worst-case crosstalk by reducing SSN. The encoding method I also reduces Type 0 crosstalk by 100 per cent, while Type 1 crosstalk is reduced by 36.4 per cent and Type 2 is reduced by 16.8 per cent. The average simultaneous switching is reduced by 51.1 per cent. Similarly, encoding method II reduces switching activity by 10.3 per cent, whereas the coupling activity is reduced by 35.4 per cent. Furthermore, encoding method II also reduced Type 0, Type 1 and Type 2 crosstalk by 100, 36.9 and 27.1 per cent, respectively. Hence, the proposed encoding methods reduced the worst-case crosstalk completely.

In VLSI technology, the reduction in feature size and the increase in operating frequency are quite rapid. This leads to higher propagation delay, crosstalk and power dissipation through the interconnects. Most of the previously proposed encoders/decoders have turned out to be unsuitable for RLC modelled interconnects. Hence, the proposed encoder would be extremely useful for crosstalk reduction in newer operating conditions.

The encoding method I identifies the harsh crosstalks, that is Type 0 and Type 1, in the inverted and non-inverted forms of incoming data with respect to the previous data. The data having minimum crosstalk in the inverted and non-inverted forms are only sent through the transmission line. The encoding method I also removes the worst-case crosstalk and simultaneously reduces other mild crosstalks. The removal of worst-case crosstalk improves the overall performance of the interconnect. The encoding method II identifies Type 2 crosstalk along with Type 0 and Type 1 similar to encoding method I. Furthermore, the encoding method II exhibits an improvement over method I in terms of reduction in crosstalk and power dissipation.

This paper proposes a novel encoding method to reduce worst-case crosstalk effects that reduces SSN. The proposed encoding methods achieve their purpose of crosstalk reduction for several technology nodes.

To review and explore optical fiber and carbon nanotube (CNT) as prospective alternatives to copper in VLSI interconnections.

As the technology moves to deep submicron level, the interconnect width also scales down. Increasing resistivity of copper with scaling and rising demands on current density drives the need for identifying new wiring solutions. This paper explores various alternatives to copper. Metallic CNTs, optical interconnects are promising candidates that can potentially address the challenges faced by copper.

Although, the theoretical aspects proves CNTs and optical interconnect to be better alternative against copper on the ground of performance parameters such as power dissipation, switching delay, crosstalk. But copper would last for coming decades on integration basis.

This paper reviews the state‐of‐the‐art in CNT interconnect and optical interconnect research; and discusses both the advantages and challenges of these emerging technologies.

The purpose of this research paper is to analyze the combined effects of driver size and coupling parasitics on crosstalk noise and delay for static and dynamically switching victim line. Furthermore, this paper shows the effect of inductance on delay and qualitatively optimizes its value to obtain minimum delay.

The interwire parasitics are the primary sources of crosstalk or coupled noise that may lead to critical delays/logic malfunctions. This paper is based on simulating a pair of distributed resistance inductance capacitance (RLC) interconnects coupled capacitively and inductively for measurements of crosstalk noise/delay. The combined effects of driver sizing and interwire parasitics on peak overshoot noise/delay are observed through simulation program with integrated circuit emphasis (SPICE) simulations for different switching patterns. Furthermore, the analysis of inductive effect on propagation delay as a function of coupling capacitance is carried out and the optimization of delay is worked out qualitatively. The simulations are carried out at 0.13 μm, 1.5 V technology node.

This paper observes the contradictory effects of coupling parasitics on wire propagation delay; however, the effect on peak noise is of a different kind. Further, this paper shows that the driver size exhibits opposite kind of behavior on propagation delay than peak over shoot noise. It is observed that the delay is affected in presence of inductance; thus, the optimization of delay is carried out.

The effects of driver sizing and interwire parasitics are analyzed through simulations. The optimum value of coupling capacitance for delay is found qualitatively. These findings are important for designing very large scale integration (VLSI) interconnects.

This paper aims to present the edge scattering dominant circuit modeling. The effect of crosstalk on gate oxide reliability (GOR), along with the mitigation using shielding technique is further studied.

An equivalent distributed Resistance Inductance Capacitance circuit of capacitively coupled interconnects of multilayer graphene nanoribbon (MLGNR) has been considered for T Simulation Program with Integrated Circuit Emphasis (TSPICE) simulations under functional and dynamic switching conditions. Complementary metal oxide semiconductor driver transistors are modeled by high performance predictive technology model that drive the distributed segment with a capacitive load of 0.001 fF, V_DD and clock frequency as 0.7 V and 0.2 GHz, respectively, at 14 nm technology node.

The results reveal that the crosstalk induced delay and noise area are dominated by the overall mean free path (MFP) (i.e. including the effect of edge roughness induced scattering), in contrary to, acoustic and optical scattering limited MFP with the temperature, width and length variations. Further, GOR, estimated in terms of average failure rate (AFR), shows that the shielding technique is an effective method to minimize the relative GOR failure rate by, 0.93e-7 and 0.7e-7, in comparison to the non-shielded case with variations in interconnect’s length and width, respectively.

Considering realistic circuit modeling for MLGNR interconnects by incorporating the edge roughness induced scattering mechanism, the outcomes exhibit more penalty in terms of crosstalk induced noise area and delay. The shielding technique is found to be an effective mitigating technique for minimizing AFR in coupled MLGNR interconnects.

To analyze the effect of voltage scaling on crosstalk.

Voltage scaling has been often used for reducing power dissipation of CMOS driven interconnects. An undesired effect observed due to voltage scaling is increase in propagation delay. Thus, a trade off lies between power dissipation and propagation delay with voltage scaling. However, voltage scaling can result in overall reduction of power delay product. Therefore, their lies an optimized supply voltage where‐in power dissipation and propagation delay can be optimized. Many of the previous researches have discussed about power dissipation and propagation delay only with voltage scaling. This paper for first time shows the effect on crosstalk in voltage scaled interconnects. In this paper, we primarily study the noise for an input signal having transition time of 50 ps. The simulations are run for interconnect length of 2 and 4 mm. These parameters are varied for four different cases of stimulations to aggressor and victim lines viz. V_A (input at aggressor node A) and V_B (input at victim node B) switching in same direction; V_A is switching and V_B at static low; V_A and V_B are switching in opposite direction; V_A is switching and V_B at static high.

It is quite encouraging to observe that irrespective of interconnect length and technology node used, an optimized voltage scaling reduces normalized crosstalk level.

Voltage scaling can be effectively used for crosstalk reduction by the new era VLSI interconnect designers. This paper shows simulation results for crosstalk reduction in different nano‐sized CMOS driven RLC‐modeled interconnects.

The purpose of this paper is to analyze the effect of driver size and number of shells on propagation delay and power for multi‐walled carbon nanotubes (MWCNT) interconnects at 22 nm technology node.

An equivalent circuit model of MWCNT is used for estimation and analysis of propagation delay and power. The delay and power through MWCNT and Cu interconnects are compared for various driver sizes and number of MWCNT shells.

The SPICE simulation results show that the MWCNT interconnect has lower propagation delay than Cu interconnects. The delay ratio of MWCNT to Cu decreases with increase in length for different driver size and number of MWCNT shells. However, the delay ratio increases with reduction in number of MWCNT shells. The ratio of average power consumption (MWCNT/Cu) also decreases with the variation in driver size and numbers of shells with respect to the length of interconnect. The theoretical study proves CNTs to be better alternatives against copper on the ground of performance parameters.

Several challenges remain to be overcome in the areas of fabrication and process integration for CNTs. Lowering of metal nanotube contact resistance would be vital, especially for local interconnect and via applications. Moreover, rigorous characterization and modeling of electromagnetic interactions in CNT bundles; 3‐D (metal) to 1‐D (CNT) contact resistance; impact of defects on electrical and thermal properties; and high‐frequency effects are being seen as additional challenges.

This paper investigates, assesses and compares the performance of carbon nanotubes (CNT) based interconnects as prospective alternatives to copper wire interconnects in future VLSI chips. Multi walled CNTs assure for long/global interconnect applications.

This paper aims to address the various issues of board‐level (off‐chip) interconnects testing. A new algorithm based on the boundary scan architecture is developed to test off‐chip interconnect faults. The proposed algorithm can easily diagnose which two interconnects are shorted.

The problems in board‐level interconnects testing are not simple. A new algorithm is developed to rectify some of the problems in existing algorithms. The proposed algorithm to test board‐level interconnect faults is implemented using Verilog on Modelsim software. The output response of each shorting between different wires of different nodes is different, which is the basis of fault detection by the proposed algorithm. The test vectors are generated by the test pattern generator and these test vectors are different for different nodes. This work implements built in self test using boundary scan technique.

The dominant‐1 (wired‐OR, denoted as WOR), dominant‐0 (wired‐AND, denoted as WAND) and stuck‐at faults are tested using the proposed algorithm. The proposed algorithm is also compared with the several algorithms in the literature, i.e. modified counting, walking one's algorithm and others. This paper's results are found to be better than the existing algorithms.

The limitation of the proposed algorithm is that, at any time, the faults on any seven nodes can be tested to avoid aliasing. So, the groups are formed out of total nodes, in a multiple of seven to carry out the testing of faults.

The proposed algorithm is free from the problems of syndromes and utilizes a smaller number of test vectors.

Various existing algorithms namely modified counting, walking one's algorithm and others are discussed. A new algorithm is developed which can easily detect board‐level dominant‐1 (WOR), dominant‐0 (WAND) and stuck‐at faults. The proposed algorithm is completely free from aliasing and confounding syndromes.