In-situ die-to-die impedance estimation for high-speed serial links
View Patent ↗A receiver includes an analog-to-digital converter (ADC) module that receives a test signal via a transmission channel and provides a time domain representation of the test signal as received by the receiver, and a processor that determines a time domain representation of an impedance of the transmission channel based on the time domain representation of the test signal.
1. A receiver, comprising:
a multiplexor including an input coupled to a transmission channel;
an equalizer coupled to a first output of the multiplexor, the multiplexor configured to provide a data signal to the equalizer in a first mode;
an analog-to-digital converter (ADC) module coupled to a second output of the multiplexor, the multiplexor further configured to provide a test signal to the ADC module in a second mode, the ADC module to provide a time domain representation of the test signal as received by the receiver; and
a processor configured to determine a time domain representation of an impedance of the transmission channel based on the time domain representation of the test signal.
2. The receiver of claim 1 wherein in determining the time domain representation of the impedance of the transmission channel, the processor further estimates a time domain pulse response of the transmission channel (S 21 ) based upon the time domain representation of the test signal.
3. The receiver of claim 2 wherein in estimating the time domain pulse response, the processor further estimates a finite difference function based upon the time domain representation of the test signal, such that
x
′
(
t
)
=
x
(
n
+
1
)
-
x
(
n
)
Δ
t
where x(n) is the time domain representation of the test signal at a discrete time n, and
where n=0, 1, 2, . . . N−1, N.
4. The receiver of claim 2 wherein in determining the time domain representation of the impedance of the transmission channel, the processor further performs a Fast Fourier Transform (FFT) on the time domain pulse response of the transmission channel, such that
X
(
k
)
=
∑
n
=
0
N
-
1
x
(
n
)
·
e
-
i
(
2
π
N
)
kn
where X(k) is a frequency domain representation of the pulse response of the transmission channel (S 21 ), and k=0, 1, 2, . . . N−1, N.
5. The receiver of claim 4 wherein in determining the time domain representation of the impedance of the transmission channel, the processor further estimates a frequency domain representation of the impedance of the transmission channel (S 11 ).
6. The receiver of claim 5 wherein, in estimating the frequency domain representation of the impedance of the transmission channel (S 11 ), the processor operates such that
1= S 21 2 +S 11 2 .
7. The receiver of claim 5 wherein in determining the time domain representation of the impedance of the transmission channel, the processor further performs an Inverse FFT on the frequency domain representation of the impedance of the transmission channel (S 11 ), such that
y
(
n
)
=
1
N
∑
n
=
0
N
-
1
y
(
k
)
·
e
i
(
2
π
N
)
kn
.
8. A method comprising:
receiving, at a multiplexor of a receiver, a data signal via a transmission channel;
providing, by the multiplexor in a first mode, the data signal to an equalizer of the receiver;
receiving, at the multiplexor, a test signal via the transmission channel;
providing, by the multiplexor in a second mode, the test signal to an analog-to-digital converter (ADC) module of a receiver;
providing, by the ADC module, a time domain representation of the test signal as received by the receiver; and
determining, by a processor of the receiver, a time domain representation of an impedance of the transmission channel based on the time domain representation of the test signal.
9. The method of claim 8 wherein in determining the time domain representation of the impedance of the transmission channel, the method further comprises:
estimating a time domain pulse response of the transmission channel (S 21 ) based upon the time domain representation of the test signal.
10. The method of claim 9 wherein in estimating the time domain pulse response, the method further comprises:
estimating a finite difference function based upon the time domain representation of the test signal, such that
x
′
(
t
)
=
x
(
n
+
1
)
-
x
(
n
)
Δ
t
where x(n) is the time domain representation of the test signal at a discrete time n,
and where n=0, 1, 2, . . . N−1, N.
11. The method of claim 9 wherein in determining the time domain representation of the impedance of the transmission channel, the method further comprises:
performing a Fast Fourier Transform (FFT) on the time domain pulse response of the transmission channel, such that
X
(
k
)
=
∑
n
=
0
N
-
1
x
(
n
)
·
e
-
i
(
2
π
N
)
kn
where X(k) is a frequency domain representation of the pulse response of the
transmission channel (S 21 ), and k=0, 1, 2, . . . N−1, N.
12. The method of claim 11 wherein in determining the time domain representation of the impedance of the transmission channel, the method further comprises:
estimating a frequency domain representation of the impedance of the transmission channel (S 11 ).
13. The method of claim 12 wherein, in estimating the frequency domain representation of the impedance of the transmission channel (S 11 ), the method operates such that
1= S 21 2 +S 11 2 .
14. The method of claim 12 wherein in determining the time domain representation of the impedance of the transmission channel, the method further comprises:
performing an Inverse FFT on the frequency domain representation of the impedance of the transmission channel (S 11 ), such that
y
(
n
)
=
1
N
∑
n
=
0
N
-
1
y
(
k
)
·
e
i
(
2
π
N
)
kn
.
15. A non-transitory computer-readable medium including code for performing a method, the method comprising:
receiving, at a multiplexor of a receiver, a data signal via a transmission channel;
providing, by the multiplexor in a first mode, the data signal to an equalizer of the receiver;
receiving, at the multiplexor, a test signal via the transmission channel;
providing, by the multiplexor in a second mode, the test signal to an analog-to-digital converter (ADC) module of a receiver;
providing, by the ADC module, a time domain representation of the test signal as received by the receiver; and
determining, by a processor of the receiver, a time domain representation of an impedance of the transmission channel based on the time domain representation of the test signal.
16. The computer-readable medium of claim 15 wherein in determining the time domain representation of the impedance of the transmission channel, the method further comprises:
estimating a time domain pulse response of the transmission channel (S 21 ) based upon the time domain representation of the test signal.
17. The computer-readable medium of claim 16 wherein in estimating the time domain pulse response, the method further comprises:
estimating a finite difference function based upon the time domain representation of the test signal, such that
x
′
(
t
)
=
x
(
n
+
1
)
-
x
(
n
)
Δ
t
where x(n) is the time domain representation of the test signal at a discrete time n,
and where n=0, 1, 2, . . . N−1, N.
18. The computer-readable medium of claim 16 wherein in determining the time domain representation of the impedance of the transmission channel, the method further comprises:
performing a Fast Fourier Transform (FFT) on the time domain pulse response of the transmission channel, such that
X
(
k
)
=
∑
n
=
0
N
-
1
x
(
n
)
·
e
-
i
(
2
π
N
)
kn
where X(k) is a frequency domain representation of the pulse response of the transmission channel (S 21 ), and k=0, 1, 2, . . . N−1, N.
19. The computer-readable medium of claim 18 wherein in determining the time domain representation of the impedance of the transmission channel, the method further comprises:
estimating a frequency domain representation of the impedance of the transmission channel (S 11 ), where
1= S 21 2 +S 11 2 .
20. The computer-readable medium of claim 19 wherein in determining the time domain representation of the impedance of the transmission channel, the method further comprises:
performing an Inverse FFT on the frequency domain representation of the impedance of the transmission channel (S 11 ), such that
y
(
n
)
=
1
N
∑
n
=
0
N
-
1
y
(
k
)
·
e
i
(
2
π
N
)
kn
.