Crosstalk Noise in FPGAs

1
Crosstalk Noise in FPGAs

• Yajun Ran
• Malgorzata Marek-Sadowska
• Department of Electrical and Computer Engineering
• University of California, Santa Barbara, CA

2
Outline

• Crosstalk noise trends in FPGAs
• Conventional crosstalk-avoidance schemes
• Twist switch box
• Experimental results
• Conclusions

3
Crosstalk noise

• Capacitive coupling
• Functional failure
• Delay noise

A
A
V
V
4
FPGA structure

• Fc
• Fs
• Segmentation
• CLB size

5
Interconnects in FPGAs

• Huge capacitance
• Buffered wires
• Larger logic delay

6
Crosstalk still a problem

• Expermental settings
• Segment lengthper stage
• Minimum space
• Minimum width
• Berkeley PTM model

7
Crosstalk reduction methods

• Coupling capacitance
• wire spacing
• Neighboring switching
• shielding
• Weak victim/strong aggressor
• buffer sizing

8
Noise cancellation

• Crosstalk can cause slowdown and speedup

A. B. Kahng etal. Interconnect Tuning
Stategiesfor High-Performance ICs, DATE99
9
Experiments
10
Noise distribution

• Crosstalk requires
• timing correlation
• logic correlation
• Worst-case accumulation alonga path is unlikely

11
Noise distribution
12
Noise distribution
Xi
n
Y ?
Xi
1
pdf
normal distribution
noise
13
Noise distribution
n
2n

• normal distribution approximation
• same mean ?n ?2n
• different deviation ?n 2?2n

14
Previous switch boxes
5
5
4
4
3
3
2
2
1
1
subset universal Wilton Common
neighbor wires are still neighbors if they follow
the same direction. BAD for crosstalk noise!
15
Twist switch box
1
2
3
4
5

• Neighbor relations are broken
• Worst-case accumulation more difficult

5
5
4
4
3
3
2
2
1
1
1
2
3
4
5
16
Twist switch box - construction

• Always exists if N gt 3
• Can be derived from previous switch box

1
2
3
4
5
5
5
4
4
3
3
2
2
1
1
1
2
3
4
5
17
Twist switch box - construction
f(t)
Wilton
Twist
W-E
t
St
S-N
t
St
W-N
N-t2
S(N-t)2
N-E
t1
St1
S Separation factor S gt 1 S
lt N-1
W-S
t-1
St-1
N-t
S-E
S(N-t)
18
Experiments settings

• VPR - place and route
• 20 largest MCNC benchmarks
• 2 segmentation schemes
• .13um technology
• aggressor counts

CLB
CLB
CLB
CLB
19
Experiments coupling length
Coupling Length
longest one coupling 34 blocks!
20
Experiments delay noise
8 improvement

21
Conclusions and future work

• Crosstalk noise will be a serious problemeven in
  FPGAs
• Trade-off among different crosstalkavoidance
  strategies
• Twist switch box naturally suppressescrosstalk
  noise
• More accurate noise analysis methods