US20100177814A1 - Equalizer and semiconductor device - Google Patents
Equalizer and semiconductor device Download PDFInfo
- Publication number
- US20100177814A1 US20100177814A1 US12/730,061 US73006110A US2010177814A1 US 20100177814 A1 US20100177814 A1 US 20100177814A1 US 73006110 A US73006110 A US 73006110A US 2010177814 A1 US2010177814 A1 US 2010177814A1
- Authority
- US
- United States
- Prior art keywords
- signal
- equalizer
- voltage
- output
- pass filter
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
- H04L25/03878—Line equalisers; line build-out devices
Definitions
- the present invention relates to an equalizer and a semiconductor device, that restore a waveform of a transmission signal, which is changed due to attenuation of signals on a transmission path, to an original waveform.
- FIG. 1 of Japanese Patent Application Laid-Open No. 2004-120468 shows an example of a circuit configuration of an equalizer for restoring a waveform of transmission signal, which is changed due to attenuation of the signal on the transmission path, to an original waveform.
- This figure shows an equalizer 18 , at the reception end, which includes a high-pass filter 22 , an amplifier 24 and an adder 20 .
- this equalizer 18 there is adopted a circuit configuration in which a high frequency component of a reception signal extracted with the high-pass filter 22 is amplified with the amplifier 24 and the amplified high frequency component is added to the reception signal with the adder 20 .
- Such a circuit configuration compensates the attenuated high frequency component.
- the noise ratio increases while the ratio of the high frequency component signal increases. This means a decrease in S/N ratio of the reception signal.
- the high-pass filter is normally configured by a capacitive element on the transmission path and a resistive element suspended from one end of the capacitive element.
- the high-pass filter since one and the other ends of the capacitive element on the transmission path are insulated from each another, it is difficult to conduct a disconnection test, by a direct current signal, on whether disconnection of the path among the reception end (b), the high-pass filter 22 and the amplifier 24 has not occurred in the equalizer. Therefore, it is necessary to extra test circuit for the purpose of conducting a disconnection test by a direct current signal.
- the amplifier 24 is provided on the path for signal transmission subsequent to the high-pass filter and prior to the adder 20 .
- the signal transmitted through the high-pass filter 22 and the amplifier 24 lags behind an original reception signal directly inputted into the adder 20 , in reaching the adder 20 because of signal delay in circuits on the path. Therefore, the simultaneity of the two signals to be added to each other with the adder 20 is impaired. This makes it difficult to truly regenerate a transmission signal, causing a decrease in reproducibility of a transmission signal.
- An object of the present invention to provide an equalizer and a semiconductor device, that can suppress a decrease in S/N ratio of a reception signal, can facilitate a disconnection test by a direct current signal, and are excellent in reproducibility of a transmission signal.
- an equalizer includes a low-pass filter, a subtraction unit, an addition unit and an amplifier.
- the low-pass filter receives a reception signal.
- the subtraction unit subtracts from the reception signal an output signal from the low-pass filter.
- the addition unit adds the reception signal to an output signal from the subtraction unit.
- the amplifier amplifies an output signal from the addition unit.
- the low frequency component of the reception signal is also amplified in the equalizer of the first aspect, thereby enabling suppression of a decrease in S/N ratio of the reception signal.
- the equalizer of the first aspect adopts a low-pass filter rather than a high-pass filter.
- the low-pass filter is normally configured by a resistive element on a transmission path and a capacitive element suspended from one end of the resistive element, and one end of the resistive element on the transmission path is not insulated from the other end thereof. It is thereby possible to facilitate a disconnection test, by a direct current signal, on whether disconnection of the path among the reception end, the low-pass filter, the subtraction unit, the addition unit and the amplifier has not occurred. Furthermore, in the equalizer of the first aspect, the amplifier is arranged on the path subsequent to the addition unit. This prevents delay in a signal, which passes through the low-pass filter and the subtraction unit to reach the addition unit, due to the amplifier. Thereby, the simultaneity is easily maintained between the reception signal and the output signal from the subtraction unit which are computed in the addition unit. The equalizer is thus excellent in reproducibility of a transmission signal.
- an equalizer includes a low-pass filter, a subtraction unit and amplifier.
- the low-pass filter receives a reception signal.
- the subtraction unit subtracts from the reception signal an output signal from the low-pass filter.
- the amplifier amplifies an output signal from the subtraction unit.
- a signal mainly composed of a high frequency component, obtained by subtracting from the reception signal the output signal from the low-pass filter, is amplified.
- the low frequency component of the reception signal is also amplified in the equalizer of the second aspect. It is thereby possible to suppress a decrease in S/N ratio of the reception signal.
- the equalizer of the second aspect adopts a low-pass filter rather than a high-pass filter.
- the low-pass filter is normally configured by a resistive element on a transmission path and a capacitive element suspended from one end of the resistive element, and one end of the resistive element on the transmission path is not insulated from the other end thereof. It is thereby possible to facilitate a disconnection test, by a direct current signal, on whether disconnection of the path among the reception end, the low-pass filter, the subtraction unit and the amplifier has not occurred. Furthermore, in the equalizer of the second aspect, the amplifier is arranged on the path subsequent to the subtraction unit. This prevents delay in a signal, which passes through the low-pass filter to reach the subtraction unit, due to the amplifier.
- the equalizer of the second aspect can exert the same effect as the equalizer of the first aspect with a simpler circuit configuration than the equalizer of the first aspect.
- an equalizer includes a signal conversion unit, a subtraction unit, an addition unit and amplifier.
- the signal conversion unit converts a reception signal into a signal in direct proportion to the reception signal to output the converted signal.
- the subtraction unit subtracts from the reception signal the output signal from the signal conversion unit.
- the addition unit adds the reception signal to an output signal from the subtraction unit.
- the amplifier amplifies an output signal from the addition unit.
- An input/output gain of the signal conversion unit is constant in a region not higher than a predetermined frequency in a frequency band of a signal component contained in the reception signal, and gradually decreases in a region exceeding the predetermined frequency.
- an input/output gain of the signal conversion unit is constant in a region not higher than a predetermined frequency in a frequency band of a signal component contained in the reception signal, and gradually decreases in a region exceeding the predetermined frequency. Therefore, since the signal conversion unit serves an equivalent function to that of the low-pass filter, as in the case of the equalizer of the first aspect, the equalizer of the third aspect can suppress a decrease in S/N ratio of the reception signal. Further, as in the equalizer of the first aspect, there occurs no delay in the signal, which passes through the signal conversion unit and the subtraction unit to reach the addition unit, due to the amplifier.
- the equalizer is thus excellent in reproducibility of a transmission signal. Moreover, it is also possible to convert the signal into a signal suitable for signal processing of the equalizer.
- an equalizer includes a signal conversion unit, a subtraction unit and an amplifier.
- the signal conversion unit converts a reception signal into a signal in direct proportion to the reception signal to output the converted signal.
- the subtraction unit subtracts from the reception signal the output signal from the signal conversion unit.
- the amplifier amplifies an output signal from the addition unit.
- An input/output gain of the signal conversion unit is constant in a region not higher than a predetermined frequency in a frequency band of a signal component contained in the reception signal, and gradually decreases in a region exceeding the predetermined frequency.
- an input/output gain of the signal conversion unit is constant in a region not higher than a predetermined frequency in a frequency band of a signal component contained in the reception signal, and gradually decreases in a region exceeding the predetermined frequency. Therefore, since the signal conversion unit serves an equivalent function to that of the low-pass filter, as in the case of the equalizer of the second aspect, the equalizer of the fourth aspect can suppress a decrease in S/N ratio of the reception signal. Further, as in the equalizer of the second aspect, there occurs no delay in the signal, which passes through the signal conversion unit to reach the subtraction unit, due to the amplifier. Thereby, the simultaneity is easily maintained between the reception signal and the output signal from the signal conversion unit which are computed in the subtraction unit.
- the equalizer is thus excellent in reproducibility of a transmission signal.
- the equalizer of the fourth aspect can exert the same effect as the equalizer of the first aspect with a simpler circuit configuration than the equalizer of the first aspect, as in the case of the equalizer of the second aspect. Moreover, it is also possible to convert the signal into a signal suitable for signal processing of the equalizer.
- FIG. 1 shows a semiconductor circuit device including a transmission circuit and a semiconductor circuit device including a reception circuit, which are connected to each another through transmission paths;
- FIG. 2 is a circuit diagram showing a principle of an equalizer according to the present invention.
- FIG. 3 is a Bode diagram showing an operational principle of the equalizer of FIG. 2 ;
- FIG. 4 is a circuit diagram showing an equalizer capable of realizing an equivalent function to that of the equalizer of FIG. 2 with a simpler configuration than the equalizer of FIG. 2 ;
- FIG. 5 is a circuit diagram showing an equalizer according to a first embodiment
- FIG. 6 shows one example of a detailed configuration of a low-pass filter
- FIG. 7 shows another example of the detailed configuration of the low-pass filter
- FIG. 8 shows one example of a detailed configuration of a voltage-current signal conversion unit
- FIG. 9 is a graph showing the relation between a voltage signal to be inputted into an input end and each of current signals to be outputted in the voltage-current signal conversion unit;
- FIG. 10 is a graph showing the relation between a frequency component contained in a voltage signal to be inputted into the input end and an input/output gain in the voltage-current signal conversion unit;
- FIG. 11 shows one example of a detailed configuration of an amplifier pre-stage unit
- FIG. 12 shows another example of the detailed configuration of the amplifier pre-stage unit
- FIG. 13 is a circuit diagram showing an equalizer according to a third embodiment
- FIG. 14 shows one example of a detailed configuration of a voltage-voltage signal conversion unit
- FIG. 15 is a graph showing the relation between a voltage signal to be inputted into the input end and each of voltages to be outputted in the voltage-voltage signal conversion unit;
- FIG. 16 is a graph showing the relation between a frequency component contained in a voltage signal to be inputted into the input end and an input/output gain in the voltage-voltage signal conversion unit;
- FIG. 17 shows one example of a detailed configuration of an amplifier pre-stage unit
- FIG. 18 shows an equalizer according to a fourth embodiment
- FIG. 19 is a sectional view showing a semiconductor chip including a semiconductor substrate on which an equalizer including an inductor element is formed;
- FIG. 20 is a top view showing the semiconductor substrate on which the equalizer is formed
- FIG. 21 is a sectional view showing the semiconductor substrate on which the equalizer is formed.
- FIG. 22 is a top view of a semiconductor substrate of a semiconductor device according to a fifth embodiment.
- FIG. 23 is a sectional view of the semiconductor substrate of the semiconductor device according to the fifth embodiment.
- FIG. 24 is a circuit diagram showing an equalizer according to a sixth embodiment.
- FIG. 1 shows a semiconductor circuit device 100 including a transmission circuit 101 and a semiconductor circuit device 300 including a reception circuit 301 , the devices being connected to each other through transmission paths 201 a, 201 b.
- the semiconductor circuit devices 100 and 300 are, for example, semiconductor IC (Integrated Circuit) chips mounted on a print substrate (not shown), and the transmission paths 201 a, 201 b are, for example, printed wiring on the print substrate (not shown).
- a positive logic signal and a negative logic signal in a complementary relation to the positive logic signal are outputted from the transmission circuit 101 .
- the positive logic signal from the transmission circuit 101 is transmitted to the reception circuit 301 through the positive logic-side transmission path 201 a.
- the negative logic signal from the transmission circuit 101 is transmitted to the reception circuit 301 through the negative logic-side transmission path 201 b.
- the present invention is an equalizer to function as this reception circuit 301 .
- FIG. 1 shows a transmission signal (e.g., a signal of aligned information of “High”-“Low”-“High” . . .) outputted from the transmission circuit 101 , and a reception end signal received in the reception circuit 301 .
- the waveform of the reception end signal appears to be dull owing to a noise or the like received on the transmission paths 201 a and 201 b.
- FIG. 2 is a circuit diagram showing the principle of the equalizer according to the present invention. As shown in FIG. 2 , this equalizer includes a low-pass filter 4 , a subtraction unit 5 , an addition unit 6 and an amplifier 2 .
- FIG. 3 is a Bode diagram showing the operational principle of the equalizer of FIG. 2 .
- the operational principle of the equalizer according to the present invention is described using FIG. 3 .
- the low-pass filter 4 receives a reception signal (having a frequency characteristic shown with a graph of CH 1 in FIG. 3 ) given from a reception end 1 , the low-pass filter 4 outputs a signal (having a frequency characteristic shown with a graph of CH 2 in FIG. 3 ) obtained by removing the high frequency component from the reception signal.
- the subtraction unit 5 subtracts from the reception signal the output signal from the low-pass filter 4 . Thereby, an output signal from the subtraction unit 5 has a frequency characteristic shown with a graph of CH 3 in FIG. 3 .
- the addition unit 6 adds a reception signal from the reception end 1 to the output signal from the subtraction unit 5 .
- an output signal from the addition unit 6 has a frequency characteristic of emphasizing the high frequency component as shown with a graph of CH 4 in FIG. 3 .
- the amplifier 2 amplifies the output signal from the addition unit 6 , and transmits it to an output end 3 .
- a signal (CH 3 ) mainly composed of a high frequency component, obtained by subtracting an output signal (CH 2 ) from the low-pass filter 4 from a reception signal (CH 1 ) is added to the reception signal (CH 1 ), and the signal (CH 4 ) after addition is amplified.
- the low frequency component of the reception signal (CH 1 ) is amplified as well as the high frequency component (CH 3 ) in the equalizer of FIG. 2 . It is thereby possible to suppress a decrease in S/N ratio of the reception signal.
- the equalizer of FIG. 2 adopts the low-pass filter 4 rather than a high-pass filter.
- the low-pass filter is normally configured by a resistive element on a transmission path and a capacitive element suspended from one end of the resistive element, and one end of the resistive element on the transmission path is not insulated from the other end thereof. It is thereby possible to facilitate a disconnection test, by a direct current signal, on whether disconnection of the path among the reception end 1 , the low-pass filter 4 , the subtraction unit 5 , the addition unit 6 and the amplifier 2 has not occurred.
- the amplifier 2 is arranged on the path subsequent to the addition unit 6 . This prevents delay in a signal, which passes through the low-pass filter 4 and the subtraction unit 5 to reach the addition unit 6 , due to the amplifier 2 . Thereby, the simultaneity is easily maintained between the reception signal from the reception end 1 and the output signal from the subtraction unit 5 which are computed in the addition unit 6 .
- the equalizer is thus excellent in reproducibility of a transmission signal.
- FIG. 4 shows an equalizer capable of realizing an equivalent function to that of the equalizer of FIG. 2 with a simpler configuration than the equalizer of FIG. 2 .
- this equalizer includes the low-pass filter 4 , the subtraction unit 5 and the amplifier 2 having the configuration of the equalizer of FIG. 2 with the addition unit 6 omitted therefrom.
- the low-pass filter 4 receives a signal (graph of CH 1 in FIG. 3 ) given from the reception end 1 and outputs a signal (graph of CH 2 in FIG. 3 ) obtained by removing the high frequency component from the reception signal.
- a value of an input/output gain in a pass band in the low-pass filter 4 is set to a value less than one time.
- the subtraction unit 5 subtracts from the reception signal the output signal from the low-pass filter 4 .
- the output signal from the subtraction unit 5 has a frequency characteristic shown with the graph of CH 3 in FIG. 3 .
- the value of the input/output gain of the pass band in the low-pass filter 4 is smaller than one time, the low frequency component is not completely lost in the output signal from the subtraction unit 5 . Consequently, the output signal from the subtraction unit 5 has a frequency characteristic of emphasizing the high frequency component.
- the amplifier 2 then amplifies the output signal from the subtraction unit 5 , and transmits it to the output end 3 .
- the low frequency component of the reception signal is amplified as well as the high frequency component in the equalizer of FIG. 4 , as in the equalizer of FIG. 2 . It is thereby possible to suppress a decrease in S/N ratio of the reception signal.
- the equalizer of FIG. 4 also adopts the low-pass filter 4 rather than a high-pass filter. It is thereby possible to facilitate a disconnection test, by a direct current signal, on whether disconnection of the path among the reception end 1 , the low-pass filter 4 , the subtraction unit 5 and the amplifier 2 has not occurred.
- the amplifier 2 is arranged on the path subsequent to the subtraction unit 5 . This prevents delay in a signal, which passes through the low-pass filter 4 to reach the subtraction unit 5 , due to the amplifier 2 . Thereby, the simultaneity is easily maintained between the reception signal from the reception end 1 and the output signal from the low-pass filter 4 which are computed in the subtraction unit 5 .
- the equalizer is thus excellent in reproducibility of a transmission signal.
- the equalizer of FIG. 4 can exert the same effect as the equalizer of FIG. 2 with a simpler circuit configuration than the equalizer of FIG. 2 .
- FIG. 5 is a circuit diagram showing an equalizer according to this embodiment, which functions as a reception circuit 301 a.
- This equalizer also includes the low-pass filter 4 and the amplifier 2 which are shown in FIG. 4 .
- adders 51 a, 51 b which are shown in FIG. 5 correspond to the addition unit 5 of FIG. 4 .
- the equalizer of FIG. 5 includes voltage-current signal conversion units 60 a, 60 b, and resistors 52 a, 52 b. It is to be noted that the adders 51 a, 51 b and the resistors 52 a, 52 b configure an amplifier pre-stage unit 50 .
- a reception end la for receiving a voltage signal of a positive logic and a reception end lb for receiving a voltage signal of a negative logic are clearly specified.
- two respective signal paths arranged after the reception ends are shown with two lines.
- a voltage signal S 1 of the positive logic and a voltage signal S 2 of the negative logic, which are received at the reception ends 1 a , 1 b , respectively, are inputted into the low-pass filter 4 , and also inputted into the voltage-current signal conversion unit 60 a.
- the voltage-current signal conversion unit 60 a converts the received voltage signals S 1 , S 2 into current signals S 6 , S 7 in direct proportion to the voltage signals, and the converted signals are respectively outputted to the adders 51 a, 51 b as the subtraction unit 5 .
- the voltage-current signal conversion unit 60 b converts output voltage signals S 4 , S 5 from the low-pass filter 4 into current signals S 8 , S 9 in direct proportion to the output voltage signals, and the converted current signals S 8 , S 9 are respectively outputted to the adders 51 a, 51 b as the subtraction unit 5 .
- the current signal S 6 of the positive logic from the voltage-current signal conversion unit 60 a is added to the current signal S 8 from the voltage-current signal conversion unit 60 b.
- the current signal S 7 of the negative logic from the voltage-current signal conversion unit 60 a is added to the current signal S 9 from the voltage-current signal conversion unit 60 b.
- the adders 51 a, 51 b function as the subtraction unit.
- the voltage signal S 1 of the positive logic and the voltage signal S 2 of the negative logic having passed through the low-pass filter 4 and reversed in polarity, are respectively inputted as the voltage signals S 5 , S 4 into the positive signal input end (indicated as “+”) and the negative signal input end (indicated as “ ⁇ ”).
- the current signals are reconverted into voltage signals S 10 , S 11 .
- the voltage signals S 10 , S 11 are respectively inputted into the input ends of the amplifier 2 .
- Output signals S 13 , S 14 from the amplifier 2 are respectively given to output ends 3 a, 3 b.
- one ends of the resistors 52 a, 52 b are connected to the input ends of the amplifier 2 , and a power supply potential VDD is supplied to the other ends of the resistors 52 a, 52 b.
- values of the resistors 52 a , 52 b are appropriately set so as to adjust the strengths of the signals to be inputted into the amplifier 2 .
- a frequency characteristic of the equalizer according to a characteristic of the transmission path for the reception signal.
- FIG. 6 shows one example of a specific configuration of the low-pass filter 4 . It should be noted that the low-pass filter of FIG. 6 is differentiated by provision of a symbol “ 4 a”.
- the low-pass filter 4 a is configured by four Nch-MOS transistors (N-channel metal oxide semiconductor transistor) N 1 a , N 2 a, N 1 b , N 2 b.
- the voltage signal S 1 of the positive logic is given to one of a drain and a source of the Nch-MOS transistor N 1 a . Further, the voltage signal S 2 of the negative logic is given to one of a drain and a source of the Nch-MOS transistor N 1 b.
- the other one of the drain and the source of the Nch-MOS transistor N 1 a outputs the voltage signal S 5 of the positive logic. Further, the other one of the drain and the source of the Nch-MOS transistor N 1 a is connected to a gate of the Nch-MOS transistor N 2 a. The other one of the drain and the source of the Nch-MOS transistor N 1 b outputs the voltage signal S 4 of the negative logic. Further, the other one of the drain and the source of the Nch-MOS transistor N 1 b is connected to a gate of the Nch-MOS transistor N 2 b.
- a source and a drain of the Nch-MOS transistor N 2 a are short-circuited, and supplied with a ground potential GND.
- a source and a drain of the Nch-MOS transistor N 2 b are short-circuited, and supplied with a ground potential GND.
- the voltage signal S 3 is given from the outside to the gates of the Nch-MOS transistors N 1 a and N 1 b .
- the strength of the voltage signal S 3 is variable in the triode region (linear region) of the Nch-MOS transistors N 1 a and N 1 b.
- both the Nch-MOS transistors N 2 a and N 2 b function as capacitive elements
- both the Nch-MOS transistors N 1 a and N 1 b function as variable resistive elements.
- the low-pass filter 4 a is a typical low-pass filter configured by a resistive element on a transmission path and a capacitive element suspended from one end of the resistive element.
- a cutoff frequency of the low-pass filter 4 a is variable.
- FIG. 7 shows another example of the detailed configuration of the low-pass filter 4 .
- the low-pass filter of FIG. 7 is differentiated by provision of a symbol “ 4 b”.
- the low-pass filter 4 b In the low-pass filter 4 b, the power supply potential VDD is given, in place of the voltage signal S 3 , to the gate of the Nch-MOS transistors N 1 a , N 1 b . Except for this respect, the low-pass filter 4 b has the same configuration as the low-pass filter 4 a of FIG. 6 .
- a cutoff frequency is a fixed value. Therefore, appropriate designing of the gate size, an injection amount of the impurity in the channel region, and the like in terms of the Nch-MOS transistors N 1 a , N 1 b enables configuration of a control-free low-pass filter.
- FIG. 8 shows one example of a detailed configuration of the voltage-current signal conversion unit 60 a capable of converting the received voltage signals S 1 , S 2 into the current signals S 6 , S 7 in direct proportion to the voltage signals, and the voltage-current signal conversion unit 60 b capable of converting the received voltage signals S 4 , S 5 into the current signals S 8 , S 9 in direct proportion to the voltage signals.
- an input/output gain of the voltage-current signal conversion unit 60 a is constant in a frequency band of a signal component contained in the received voltage signals S 1 , S 2
- the input/output gain of the voltage-current signal conversion unit 60 b is constant in a frequency band of a signal component contained in the received voltage signals S 4 , S 5 .
- the voltage-current signal conversion units 60 a, 60 b have an equivalent circuit configuration except for the following differences.
- the input signals are the signals S 1 , S 2 and the output signals are the current signals S 6 , S 7 in the voltage-current signal conversion unit 60 a, whereas the input signals are the voltage signals S 4 , S 5 and the output signals are the current signals S 8 , S 9 in the voltage-current signal conversion unit 60 b . Therefore, only the circuit configuration of the voltage-current signal conversion unit 60 a is shown in FIG. 8 , with each signal in the case of the circuit configuration of the voltage-current signal conversion unit 60 b shown in parenthesis. Also in the following description, each signal in the case of the voltage-current signal conversion unit 60 b is shown in parenthesis, as in FIG. 8 .
- the voltage-current signal conversion unit 60 a ( 60 b ) is configured by six Pch-MOS transistors P 1 to P 6 and two Nch-MOS transistors N 3 , N 4 .
- the voltage signal S 1 of the positive logic (the voltage signal S 4 of the negative logic) is given to a gate of the Pch-MOS transistor P 5 .
- the voltage signal S 2 of the negative logic (the voltage signal S 5 of the positive logic) is given to a gate of the Pch-MOS transistor P 6 .
- a drain of the Pch-MOS transistor P 1 is connected to a source of the Pch-MOS transistor P 5 . Further, a drain of the Pch-MOS transistor P 2 is connected to a source of the Pch-MOS transistor P 6 .
- the power supply potential VDD is supplied to both sources of the Pch-MOS transistors P 1 and P 2 .
- a bias potential “bias” is supplied to both gates of the Pch-MOS transistors P 1 , P 2 .
- a drain of the Pch-MOS transistor P 3 is connected to a drain of the Pch-MOS transistor P 4 and, also, to the source of the Pch-MOS transistor P 5 . Further, a source of the Pch-MOS transistor P 3 is connected to a source of the Pch-MOS transistor P 4 and, also, to the source of the Pch-MOS transistor P 6 .
- a gate of the Pch-MOS transistor P 3 is connected to the gate of the Pch-MOS transistor P 5
- a gate of the Pch-MOS transistor P 4 is connected to a gate of the Pch-MOS transistor P 6 .
- the drain of the Pch-MOS transistor P 5 is connected to a drain of the Nch-MOS transistor N 3 and, also, to a gate of the Nch-MOS transistor N 3 . Further, a drain of the Pch-MOS transistor P 6 is connected to a drain of the Nch-MOS transistor N 4 and, also, to a gate of the Nch-MOS transistor N 4 .
- the ground potential GND is supplied to both sources of the Nch-MOS transistors N 3 , N 4 .
- Respective currents In, Ip in the drains of the Nch-MOS transistors N 3 , N 4 are current signals S 6 (S 8 ), S 7 (S 9 ) as output signals from the voltage-current signal conversion unit 60 a ( 60 b ).
- FIG. 9 is a graph showing the relation between the voltage signal S 1 (S 4 ) to be inputted into an input end PI and each of the currents In, Ip, i.e., the current signals S 6 (S 8 ), S 7 (S 9 ), in the voltage-current signal conversion unit 60 a ( 60 b ).
- the gate size, an injection amount of the impurity in the channel region, and the like, in terms of the Pch-MOS transistors P 1 to P 6 and Nch-MOS transistors N 3 , N 4 are appropriately designed so as to obtain the currents In, Ip, i.e., the current signals S 6 (S 8 ), S 7 (S 9 ), in direct proportion to the voltage signal S 1 (S 4 ), as shown in FIG. 9 .
- FIG. 10 is a graph showing the relation between a frequency component contained in the voltage signals S 1 (S 4 ), S 2 (S 5 ), and an input/output gain between the voltage signals S 1 (S 4 ), S 2 (S 5 ) and the current signals S 6 (S 8 ), S 7 (S 9 ), in the voltage-current signal conversion unit 60 a ( 60 b ).
- the gate size, an injection amount of the impurity in the channel region, and the like, in terms of the Pch-MOS transistors P 1 to P 6 and the Nch-MOS transistors N 3 , N 4 are appropriately designed so as to keep the input/output gain of the voltage-current signal conversion unit 60 a ( 60 b ) constant in a frequency band of a signal component contained in the received voltage signals S 1 (S 4 ), S 2 (S 5 ), as shown in FIG. 10 .
- the voltage-current signal conversion unit 60 a ( 60 b ) converts the voltage signals S 1 (S 4 ), S 2 (S 5 ) to be inputted, in a state where the input/output gain is constant in a frequency band of a signal component contained in the voltage signals S 1 (S 4 ), S 2 (S 5 ), into the current signals S 6 (S 8 ), S 7 (S 9 ), and then outputs the converted current signals S 6 (S 8 ), S 7 (S 9 ).
- a reception signal can be reduced to a signal suitable for the operating power supply voltage, and thus the reception signal can be converted into a signal suitable for signal processing in the equalizer.
- the signal can be magnified to a signal with a large amplitude, and thus can be converted into a signal suitable for signal processing in the equalizer.
- FIG. 11 shows the detailed configuration of the amplifier pre-stage unit 50 . It is to be noted that the amplifier pre-stage unit of FIG. 11 is differentiated by provision of a symbol “ 50 a”.
- the amplifier pre-stage unit 50 a is configured by two groups each including two Pch-MOS transistors P 7 a, P 7 b and four Nch-MOS transistors N 5 a, N 6 a, N 5 b, N 6 b. Among them, two groups each including the Pch-MOS transistors P 7 a, P 7 b and the Nch-MOS transistors N 5 a, N 5 b , configure the resistors 52 a, 52 b, while two groups each including the Nch-MOS transistors N 6 a, N 6 b configure the adders 51 a, 51 b.
- the adder 51 a adds the current signal S 6 to the current signal S 8 , and reconverts the addition result into the voltage signal S 10 , to be outputted. Further, the adder 51 b adds the current signal S 7 to the current signal S 9 , and reconverts the addition result into the voltage signal S 11 , to be outputted.
- the resistors 52 a, 52 b have an equivalent circuit configuration
- the adders 51 a, 51 b have an equivalent circuit configuration, except for the following differences.
- the input signals are the current signals S 6 , S 8 and the output signal is the voltage signal S 10 in the resistor 52 a and the adder 51 a
- the input signals are the current signals S 7 , S 9 and the output signal is the voltage signal S 11 in the resistor 52 b and the adder 51 b. Therefore, only the circuit configurations of the resistor 52 a and the adder 51 a are shown in FIG. 11 , with each signal in the case of the circuit configuration of the resistor 52 b and the adder 51 b shown in parenthesis. Also in the following description, respective signals in the case of the resistor 52 b and the adder 51 b are shown in parenthesis, as in FIG. 11 .
- the current signal S 6 of the positive logic (current signal S 7 of the negative logic) is given to a gate of the Nch-MOS transistor N 6 a. Further, the current signal S 8 of the negative logic (the current signal S 9 of the positive logic) is given to a gate of the Nch-MOS transistor N 6 b.
- the ground potential GND is supplied to both sources of the Nch-MOS transistors N 6 a, N 6 b.
- a drain of the Nch-MOS transistor N 6 a is connected to a source of the Nch-MOS transistor N 5 a while a source of the Nch-MOS transistor N 5 b is connected to a drain of the Nch-MOS transistor N 6 b.
- a drain of the Pch-MOS transistor P 7 a is connected to a drain of the Nch-MOS transistor N 5 a while a drain of the Pch-MOS transistor N 7 b is connected to a drain of the Nch-MOS transistor N 5 b.
- the power supply potential VDD is supplied to both sources of the Pch-MOS transistors P 7 a , P 7 b.
- Drains of the Pch-MOS transistors P 7 a, P 7 b are also connected to each other, and a voltage at this node is the voltage signal S 10 (S 11 ) as an output signal from the amplifier pre-stage unit 50 a to the amplifier 2 .
- the voltage signal S 12 a (S 12 b ) is given from the outside to gates of the Pch-MOS transistors P 7 a and P 7 b. Moreover, a voltage signal /S 12 a (/S 12 b ) as a reversed signal of the voltage signal S 12 a (S 12 b ) is also given from the outside to gates of the Nch-MOS transistors N 5 a and N 5 b.
- the strengths of the voltage signals S 12 a (S 12 b ), /S 12 a (/S 12 b ) are variable in the triode region (linear region) of the Pch-MOS transistors P 7 a and P 7 b as well as the Nch-MOS transistors N 5 a and N 5 b. It is to be noted that the voltage signal S 12 is composed of the voltage signals S 12 a and S 12 b.
- variable resistors 52 a, 52 b allow adjustment of the strengths of the signals S 10 , S 11 to be inputted into the amplifier 2 , and it is thus possible to set the frequency characteristic of the equalizer according to characteristics of the transmission paths 201 a, 201 b for the reception signals.
- FIG. 12 shows another example of the detailed configuration of the amplifier pre-stage unit 50 .
- the amplifier pre-stage unit of FIG. 12 is differentiated by provision of a symbol “ 50 b”.
- the Nch-MOS transistors N 5 a and N 5 b in the amplifier pre-stage unit 50 a of FIG. 11 are omitted.
- the drain of the Nch-MOS transistor N 6 a is directly connected with the drain of the Pch-MOS transistor P 7 a
- the drain of the Nch-MOS transistor N 6 b is directly connected with the drain of the Pch-MOS transistor P 7 b.
- the ground potential GND is supplied, in place of the voltage signal S 12 a, to the gates of the Pch-MOS transistors P 7 a and P 7 b .
- the Pch-MOS transistors P 7 a and P 7 b configure a resistor 52 c ( 52 d ), and the Nch-MOS transistors N 6 a and N 6 b configure an adder 51 c ( 51 d ).
- the amplifier pre-stage unit 50 b of FIG. 12 has the same configuration as the amplifier pre-stage unit 50 a of FIG. 11 .
- resistance values of the resistors 52 c, 52 d are fixed values. Therefore, appropriate designing of the gate size, an injection amount of the impurity in the channel region, and the like in terms of the Pch-MOS transistors P 7 a and P 7 b enables configuration of a control-free resistor.
- a second embodiment is a modification of the equalizer according to the first embodiment, obtained by omitting the low-pass filter 4 in the first embodiment and instead providing a low-pass filter function to the voltage-current signal conversion unit 60 b.
- the gate size, an injection amount of the impurity in the channel region, and the like in terms of each MOS transistor of the voltage-current signal conversion unit 60 b are designed such that the input/output gain of the voltage-current signal conversion unit 60 b is constant in a region not higher than a predetermined frequency in the frequency band of the signal component contained in the received voltage signals S 4 , S 5 , and gradually decreases in a region exceeding the predetermined frequency.
- the voltage-current signal conversion unit 60 b serves an equivalent function to the low-pass filter 4 , it is possible to suppress a decrease in S/N ratio of the reception signal, as in the case of the equalizer according to the first embodiment.
- the equalizer according to the first embodiment there occurs no delay in the signal, which passes through the voltage-current signal conversion unit 60 b to reach the subtraction unit 5 , due to the amplifier 2 , thereby making it easy to maintain the simultaneity between the reception signal and the output signal from the voltage-current signal conversion unit 60 b which are computed in the subtraction unit 5 .
- the equalizer of this embodiment is thus excellent in reproducibility of a transmission signal.
- the equalizer according to this embodiment can exert the same effect as the equalizer of FIG. 2 with a simpler circuit configuration than the equalizer of FIG. 2 .
- the equalizer can convert the signal into a signal suitable for signal processing of the equalizer by means of the voltage-current signal conversion unit 60 b.
- a third embodiment is also a modification of the equalizer according to the first embodiment, where voltage-voltage signal conversion units are adopted in place of the voltage-current signal conversion units 60 a, 60 b in the first embodiment.
- FIG. 13 is a circuit diagram showing the equalizer according to this embodiment which functions as a reception circuit 301 b.
- This equalizer also includes adders 51 e, 51 f similar to the adders 51 a, 51 b, resisters 52 e , 52 f similar to the resistors 52 a, 52 b, the low-pass filter 4 , and the amplifier 2 . It should be noted that the adders 51 e, 51 f and the resistors 52 e, 52 f configure an amplifier pre-stage unit 50 c.
- the equalizer of FIG. 13 includes voltage-voltage signal conversion units 61 a, 61 b in place of the voltage-current signal conversion units 60 a, 60 b.
- the voltage-voltage signal conversion unit 61 a converts the received voltage signals S 1 , S 2 into voltage signals S 6 a, S 7 a in direct proportion to the received voltage signals, and outputs the converted signals respectively to the adders 51 e, 51 f as the subtraction unit 5 .
- the voltage-voltage signal conversion unit 61 b converts the output voltage signals S 4 , S 5 from the low-pass filter 4 into voltage signals S 8 a, S 9 a in direct proportion to the output voltage signals, and outputs the converted signals S 8 a, S 9 a respectively to the adders 51 e, 51 f as the subtraction unit 5 .
- a voltage signal S 6 a of the positive logic from the voltage-voltage signal conversion unit 61 a is added to a voltage signal S 8 a from the voltage-voltage signal conversion unit 61 b.
- a voltage signal S 7 a of the negative logic from the voltage-voltage signal conversion unit 61 a is added to a voltage signal S 9 a from the voltage-voltage signal conversion unit 61 b.
- Voltage signals S 10 a, S 11 a as the addition results are outputted from the adders 51 e, 51 f.
- the voltage signals S 10 a, S 11 a are respectively inputted into the input ends of the amplifier 2 .
- the output signals S 13 , S 14 from the amplifier 2 are respectively given to the output ends 3 a, 3 b.
- one ends of the resistors 52 e, 52 f are connected to the input ends of the amplifier 2 , and the power supply potential VDD is supplied to the other ends of the resistors 52 e, 52 f.
- FIG. 14 shows one example of a detailed configuration of the voltage-voltage signal conversion unit 61 a capable of converting the received voltage signals S 1 , S 2 into the voltage signals S 6 a , S 7 a in direct proportion to the received voltage signals, and the voltage-voltage signal conversion unit 61 b capable of converting the received voltage signals S 4 , S 5 into the voltage signals S 8 a, S 9 a in direct proportion to the received voltage signals.
- an input/output gain of the voltage-voltage signal conversion unit 61 a is constant in a frequency band of a signal component contained in the received voltage signals S 1 , S 2
- the input/output gain of the voltage-voltage signal conversion unit 61 b is constant in a frequency band of a signal component contained in the received voltage signals S 4 , S 5 .
- the voltage-voltage signal conversion units 61 a, 61 b have an equivalent circuit configuration except for the following differences.
- the input signal are the voltage signals S 1 , S 2 and the output signals are the voltage signals S 6 a , S 7 a in the voltage-voltage signal conversion unit 61 a
- the input signals are the voltage signals S 4 , S 5
- the output signals are the voltage signals S 8 a, S 9 a in the voltage-voltage signal conversion unit 61 b. Therefore, only the circuit configuration of the voltage-voltage signal conversion unit 61 a is shown in FIG. 14 , with each signal in the case of the circuit configuration of the voltage-voltage signal conversion unit 61 b shown in parenthesis. Also in the following description, each signal in the case of the voltage-voltage signal conversion unit 61 b is shown in parenthesis, as in FIG. 14 .
- the voltage-voltage signal conversion unit 61 a ( 61 b ) is configured by six Pch-MOS transistors P 1 to P 6 and two Nch-MOS transistors N 3 , N 4 .
- the voltage signal S 1 of the positive logic (voltage signal S 4 of the negative logic) is given to a gate of the Pch-MOS transistor P 5 . Further, the voltage signal S 2 of the negative logic (the voltage signal S 5 of the positive logic) is given to a gate of the Pch-MOS transistor P 6 .
- a drain of the Pch-MOS transistor P 1 is connected to a source of the Pch-MOS transistor P 5 . Further, a drain of the Pch-MOS transistor P 2 is connected to a source of the Pch-MOS transistor P 6 .
- the power supply potential VDD is supplied to both sources of the Pch-MOS transistors P 1 and P 2 .
- the bias potential “bias” is supplied to both gates of the Pch-MOS transistors P 1 , P 2 .
- a drain of the Pch-MOS transistor P 3 is connected to a drain of the Pch-MOS transistor P 4 and, also, to the source of the Pch-MOS transistor P 5 . Further, a source of the Pch-MOS transistor P 3 is connected to a source of the Pch-MOS transistor P 4 and, also, to the source of the Pch-MOS transistor P 6 .
- a gate of the Pch-MOS transistor P 3 is connected to a gate of the Pch-MOS transistor P 5
- a gate of the Pch-MOS transistor P 4 is connected to a gate of the Pch-MOS transistor P 6 .
- a drain of the Pch-MOS transistor P 5 is connected to a drain of the Nch-MOS transistor N 3 .
- the power supply potential VDD is supplied to a gate of the Nch-MOS transistor N 3 .
- a drain of the Pch-MOS transistor P 6 is connected to a drain of the Nch-MOS transistor N 4 .
- the power supply potential VDD is supplied to a gate of the Nch-MOS transistor N 4 .
- the ground potential GND is supplied to both sources of the Nch-MOS transistors N 3 , N 4 .
- Respective voltages NO, PO in the drains of the Nch-MOS transistors N 3 , N 4 are voltage signals S 6 a (S 8 a ), S 7 a (S 9 a ) as output signals from the voltage-voltage signal conversion unit 61 a ( 61 b ).
- FIG. 15 is a graph showing the relation between the voltage signal S 1 (S 4 ) to be inputted into the input end PI and each of the voltages NO, PO, i.e., the voltage signals S 6 a (S 8 a ), S 7 a (S 9 a ), in the voltage-voltage signal conversion unit 61 a ( 61 b ).
- the gate size, an injection amount of the impurity in the channel region, and the like, in terms of the Nch-MOS transistors N 3 , N 4 and the Pch-MOS transistors P 1 to P 6 are appropriately designed so as to obtain the voltages NO, PO, i.e., the voltage signals S 6 a (S 8 a ), S 7 a (S 9 a ), in direct proportion to the voltage signal S 1 (S 4 ), as shown in FIG. 15 .
- FIG. 16 is a graph showing the relation between a frequency component contained in the voltage signals S 1 (S 4 ), S 2 (S 5 ), and an input/output gain between the voltage signals S 1 (S 4 ), S 2 (S 5 ) and the voltage signals S 6 a (S 8 a ), S 7 a (S 9 a ), in the voltage-voltage signal conversion unit 61 a ( 61 b ).
- the gate size, an injection amount of the impurity in the channel region, and the like, in terms of the Pch-MOS transistors P 1 to P 6 and the Nch-MOS transistors N 3 , N 4 are appropriately designed so as to keep the input/output gain of the voltage-voltage signal conversion unit 61 a ( 61 b ) constant in a frequency band of a signal component contained in the received voltage signals S 1 (S 4 ), S 2 (S 5 ), as shown in FIG. 16 .
- the voltage-voltage signal conversion unit 61 a ( 61 b ) converts the voltage signals S 1 (S 4 ), S 2 (S 5 ) to be inputted, in a state where the input/output gain is constant in a frequency band of a signal component contained in the voltage signals S 1 (S 4 ), S 2 (S 5 ), into the voltage signals S 6 a (S 8 a ), S 7 a (S 9 a ), and then outputs the converted voltage signals S 6 a (S 8 a ), S 7 a (S 9 a ).
- a reception signal can be reduced, or magnified to a signal with a large amplitude, and thus can be converted into a signal suitable for signal processing in the equalizer.
- FIG. 17 is an example showing the detailed configuration of the amplifier pre-stage unit 50 . It is to be noted that the amplifier pre-stage unit of FIG. 17 is differentiated by provision of a symbol “ 50 c”.
- the amplifier pre-stage unit 50 c is configured by two groups each including a Pch-MOS transistors P 7 c and two Nch-MOS transistors N 6 c , N 6 d. Among them, two groups of the Pch-MOS transistor P 7 c configure the resistors 52 e, 52 f, while two groups each including the Nch-MOS transistors N 6 c, N 6 d configure the adders 51 e, 51 f.
- the adder 51 e adds the voltage signal S 6 a to the voltage signal S 8 a, and reconverts the addition result into the voltage signal S 10 a , to be outputted. Further, the adder 51 f adds the voltage signal S 7 a to the voltage signal S 9 a, and reconverts the addition result into the voltage signal S 11 a , to be outputted.
- the resistors 52 e, 52 f have an equivalent circuit configuration
- the adders 51 e, 51 f have an equivalent circuit configuration, except for the following differences.
- the input signals are the voltage signals S 6 a, S 8 a and the output signal is the voltage signal S 10 a in the resistor 52 e and the adder 51 e
- the input signals are the voltage signals S 7 a, S 9 a and the output signal is the voltage signal S 11 a in the resistor 52 f and the adder 51 f. Therefore, only the circuit configurations of the resistor 52 e and the adder 51 e are shown in FIG. 17 , with each signal in the case of the circuit configuration of the resistor 52 f and the adder 51 f shown in parenthesis. Also in the following description, each signal in the case of the resistor 52 f and the adder 51 f is shown in parenthesis, as in FIG. 17 .
- the voltage signal S 6 a of the positive logic (voltage signal S 7 a of the negative logic) is given to a gate of the Nch-MOS transistor N 6 c. Further, the voltage signal S 8 a of the negative logic (the voltage signal S 9 a of the positive logic) is given to a gate of the Nch-MOS transistor N 6 d.
- the ground potential GND is supplied to both sources of the Nch-MOS transistors N 6 c, N 6 d.
- a drain of the Nch-MOS transistor Plc is connected to both drains of the Nch-MOS transistor N 6 c, N 6 d. Further, the power supply potential VDD is supplied to a source of the Pch-MOS transistor P 7 c.
- a voltage of a drain of the Pch-MOS transistor P 7 c is the voltage signal S 10 a (S 11 a ) as an output signal from the amplifier pre-stage unit 50 c to the amplifier 2 .
- resistance values of the resistors 52 e, 52 f are fixed values. Therefore, appropriate designing of the gate size, an injection amount of the impurity in the channel region, and the like in terms of the Pch-MOS transistors P 7 c enables configuration of a control-free resistor.
- the signal S 12 shown in FIG. 5 may be given to a gate of the Pch-MOS transistor P 7 c so as to make the resistors 52 e, 52 f variable, as in the case of the first embodiment.
- this embodiment may be applied to the equalizer according to the second embodiment as well as the equalizer according to the first embodiment.
- a fourth embodiment is also a modification of the equalizer according to the first embodiment, where the voltage signal S 12 for determining resistance values of the variable resistors 52 a, 52 b and the voltage signal S 3 for determining a cutoff frequency of the low-pass filter 4 a in the first embodiment are automatically generated to be the optimum values.
- FIG. 18 shows an equalizer according to this embodiment.
- the equalizer according to this embodiment further includes a bit error rate tester (BERT) for measuring bit error rates of the reception signals S 1 , S 2 upon receiving the output signals S 13 , S 14 from the amplifier 2 , or an eye pattern detector 70 for detecting eye patterns of the reception signals S 1 , S 2 upon receiving the output signals S 13 , S 14 from the amplifier 2 .
- BERT bit error rate tester
- the bit error rate tester or the eye detector 70 transmits a resistance value adjustment signal as the voltage signal S 12 for adjusting resistance values of the variable resistors 52 a, 52 b, and further transmits a cutoff frequency adjustment signal as the voltage signal S 3 for adjusting a cutoff frequency of the low-pass filter 4 a.
- the output signals S 3 and S 12 as the resistance value adjustment signal and the cutoff frequency adjustment signal are generated so as to minimize the bit error rates of the reception signals S 1 , S 2 .
- the output signals S 3 and S 12 as the resistance value adjustment signal and the cutoff frequency adjustment signal are generated so as to maximize the areas of the eye patterns of the reception signals S 1 , S 2 .
- bit error rate tester or the eye pattern detector 70 outputs both the resistance value adjustment signal and the cutoff frequency adjustment signal
- another configuration example may be to output either of the two signals.
- this embodiment may be applied to the equalizer according to the second or third embodiment as well as the equalizer according to the first embodiment.
- a fifth embodiment is a semiconductor device on which an equalizer according to any one of the first to fourth embodiments is mounted.
- An equalizer including an inductor element is shown in each of FIGS. 3 , 9 and 12 of Japanese Patent Application Laid-Open No. 2003-168944. The case of mounting such an equalizer on a semiconductor device is considered.
- FIG. 19 is a sectional view showing a semiconductor chip as a semiconductor device including a semiconductor substrate on which an equalizer including an inductor element is formed, and a flip chip package containing the semiconductor substrate.
- This semiconductor chip has a package substrate Sp with a solder ball SB formed on the one-side main face thereof, a package ring PR formed on the periphery of the other-side main surface of the package substrate Sp, and a heat sink RB bonded to the package ring PR.
- the package substrate Sp, the package ring PR and the heat sink RB configure the flip chip package.
- One-side main face of a semiconductor substrate Ss on which the equalizer is formed is bonded to the heat sink RB via a resin RS 2 .
- An interlayer insulation film IL is formed on the other-side main face of the semiconductor substrate Ss.
- a plurality of bumps BP are formed on the surface of the interlayer insulation film IL.
- the plurality of bumps BP are electrically connected to respective prescribed portions on the package substrate Sp.
- a resin RS 1 is formed on the peripheries of the plurality of bumps BP and on the surface and in the vicinity of the interlayer insulation film IL, so as to solidify the connection between the plurality of bumps BP and the package substrate Sp.
- FIG. 20 is a top view seen from the main face (interlayer insulation film IL forming side) of the semiconductor substrate Ss on which the equalizer is formed. Further, FIG. 21 is a sectional view taken along a cutting line XXI-XXI in FIG. 20 .
- an equalizer EQ is formed on the surface of the semiconductor substrate Ss.
- the interlayer insulation film IL is formed so as to cover the surface of the semiconductor substrate Ss and the equalizer EQ.
- Conductive pads PD are formed on the interlayer insulation film IL, and each of the bumps BP is formed on each of the pads PD. It should be noted that the pads PD and the equalizer EQ are electrically connected to each other through wiring WR in the interlayer insulation film IL.
- the equalizer EQ includes a ring-like inductor element ID.
- the inductor element ID generates a magnetic field MF at the time of operation of the circuit.
- the magnetic field MF generated by inductor element ID is interfered with the bumps BP of the flip chip package, leading to fluctuation in frequency characteristic of the equalizer EQ.
- variations tend to occur in circuit characteristic among the equalizers EQ due to the positional relation between the inductor element ID in each of the equalizers EQ and the bumps BP.
- the equalizer since an equalizer according to any one of the first to fourth embodiments is mounted on the semiconductor device according to this embodiment, the equalizer includes no inductor element (see configurations of the low-pass filters in FIGS. 6 and 7 ). Therefore, the above-mentioned problems that may arise in the case of mounting an equalizer including an inductor element on a semiconductor device will not arise in the semiconductor device according to this embodiment.
- FIG. 22 is a top view of the semiconductor substrate of the semiconductor device according to this embodiment.
- FIG. 23 is a sectional view taken along a cutting line XXIII-XXIII in FIG. 22 . Symbols used in FIGS. 22 and 23 are respectively the same as those shown in FIGS. 20 and 21 , and descriptions of those symbols are thus omitted.
- the inductor element ID is not formed in the equalizer EQ.
- This equalizer EQ is an equalizer according to any one of the first to fourth embodiments. Namely, the equalizer EQ having no inductor element is formed on the surface of the semiconductor substrate Ss, as in the cases of the low-pass filters 4 a and 4 b in FIGS. 6 and 7 .
- the flip chip package configured by the package substrate Sp, the package ring PR and the heat sink RB, contains the semiconductor substrate Ss.
- a sixth embodiment is also a modification of the equalizer according to the first embodiment, where capacitors are provided between the reception end 1 a , 1 b and the input end of the amplifier 2 in the first embodiment.
- FIG. 24 is a circuit diagram showing an equalizer according to this embodiment, which functions as the reception circuit 301 c.
- This equalizer also includes the adders 51 a, 51 b, the resistors 52 a, 52 b, the voltage-current signal conversion units 60 a, 60 b, the low-pass filter 4 , and the amplifier 2 , which are shown in FIG. 5 .
- the equalizer of FIG. 24 includes capacitors CP a , CP b respectively on the transmission path for the positive logic signal and the transmission path for the negative logic signal, the respective capacitors having the one electrodes for receiving the reception signals S 1 , S 2 , and the other electrodes for receiving the signals S 10 , S 11 to be inputted into the amplifier 2 .
- the capacitors CP a , CP b function as bypasses for transmitting a high frequency component of a reception signal to the amplifier 2 , thereby allowing improvement in high frequency characteristic of the equalizer.
- this embodiment may be applied to the equalizers according to the second to fifth embodiments as well as the equalizer according to the first embodiment.
- the description has been given based upon the equalizer of FIG. 4 .
- the output signals from the adders 51 a, 51 b as the subtraction unit 5 are added to the output signal S 6 , S 7 from the voltage-current signal conversion unit 60 a, and the addition results are inputted into the amplifier 2 .
- the low-pass filter 4 may be deleted from the foregoing configuration obtained by applying the first embodiment to the equalizer of FIG. 2 .
- the gate size, an injection amount of the impurity in the channel region, and the like, in terms of the transistors in the voltage-current signal conversion unit 60 b are designed such that the input/output gain of the voltage-current signal conversion unit 60 b is constant in a region not higher than a predetermined frequency in a frequency band of a signal component included in the received voltage signals S 1 (S 4 ), S 2 (S 5 ) and gradually decreases in a region exceeding the predetermined frequency.
- the voltage-current signal conversion units 60 a, 60 b may be changed to the voltage-voltage signal conversion units 61 a, 61 b described in the third embodiment.
- bit error rate tester or the eye pattern detector 70 described in the fourth embodiment may be added to the foregoing configuration where the first or second embodiment is applied to the equalizer of FIG. 2 .
- a semiconductor device having the flip chip package described in the fifth embodiment may be adopted in the foregoing configuration where the first or second embodiment is applied to the equalizer of FIG. 2 .
- the capacitors CP a , CP b described in the sixth embodiment may be added to the foregoing configuration where the first or second embodiment is applied to the equalizer of FIG. 2 .
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Amplifiers (AREA)
- Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)
- Dc Digital Transmission (AREA)
Abstract
The present invention provides an equalizer and a semiconductor device, that can suppress a decrease in S/N ratio of a reception signal, can facilitate a disconnection test by a direct current signal, and are excellent in reproducibility of a transmission signal. A low-pass filter receives a reception signal supplied from a reception end to output a signal obtained by removing a high frequency component from the reception signal. A subtraction unit subtracts an output signal from the low-pass filter from the reception signal. An addition unit adds the reception signal from the reception end to an output signal from the subtraction unit. Thus, an output signal from the addition unit has a frequency characteristic of emphasizing the high frequency component. Then, an amplifier amplifies the output signal from the addition unit to transmit it to an output end.
Description
- This application is a Continuation of and claims the benefit of priority under 35 U.S.C. §120 from U.S. Ser. No. 11/282,647, filed Nov. 21, 2005, the entire contents of which are incorporated herein by reference. U.S. Ser. No. 11/282,647 claims the benefit of priority under 35 U.S.C. §119 from Japanese Patent Application Nos. JP 2004-344000, filed Nov. 29, 2004 and JP 2005-274904, filed Sep. 22, 2005.
- 1. Field of the Invention
- The present invention relates to an equalizer and a semiconductor device, that restore a waveform of a transmission signal, which is changed due to attenuation of signals on a transmission path, to an original waveform.
- 2. Description of the Background Art
- On a transmission path such as a printed wiring on a print substrate, the higher the frequency component contained in a transmission signal, the greater the increasing amount of attenuation of the signal. For this reason, the waveform of a reception signal that has reached a signal reception end through the transmission path is distorted with a high frequency component attenuated, as compared with a transmission signal at a transmission end.
- Especially in the case of a digital signal, when it becomes impossible to properly recognize “High” or “Low” of a signal at the reception end due to the attenuation of the signal on the transmission path, it then becomes difficult to exchange the signal between a semiconductor integrated circuit device on the transmission side and a semiconductor integrated circuit device on the reception side.
-
FIG. 1 of Japanese Patent Application Laid-Open No. 2004-120468 shows an example of a circuit configuration of an equalizer for restoring a waveform of transmission signal, which is changed due to attenuation of the signal on the transmission path, to an original waveform. This figure shows an equalizer 18, at the reception end, which includes a high-pass filter 22, an amplifier 24 and an adder 20. - In this equalizer 18, there is adopted a circuit configuration in which a high frequency component of a reception signal extracted with the high-pass filter 22 is amplified with the amplifier 24 and the amplified high frequency component is added to the reception signal with the adder 20. Such a circuit configuration compensates the attenuated high frequency component.
- It is to be noted that “A CMOS 3.5 Gbps Continuous-time Adaptive Cable Equalizer with Joint Adaptation Method of Low-Frequency Gain and High-Frequency Boosting” (Jong-Sang Choi et al., 2003 Symposium on VLSI Circuits Digest of Technical Papers 4-89114-034-8, pp. 103-106) also shows a similar circuit configuration to that of Japanese Patent Application Laid-Open No. 2004-120468. Further, in addition to Japanese Patent Application Laid-Open No. 2004-120468 and the above non-patent document, Japanese Patent Application Laid-Open No. 09-167944 (1997) and Japanese Patent Application Laid-Open No. 2003-168944 are related to the present invention.
- In the case of the circuit configuration of the equalizer 18 in Japanese Patent Application Laid-Open No. 2004-120468, since only the high frequency component is amplified with the high-pass filter 22 and the amplifier 24, high frequency other than the transmission signal (e.g., noise at the time of switching, etc.) is also amplified. In the meantime, the original reception signal is not amplified, and added with the amplified high frequency component.
- Hence in the reception signal added with the high frequency component, the noise ratio increases while the ratio of the high frequency component signal increases. This means a decrease in S/N ratio of the reception signal.
- Further, the high-pass filter is normally configured by a capacitive element on the transmission path and a resistive element suspended from one end of the capacitive element. In the case of adopting such a high-pass filter in the equalizer 18, since one and the other ends of the capacitive element on the transmission path are insulated from each another, it is difficult to conduct a disconnection test, by a direct current signal, on whether disconnection of the path among the reception end (b), the high-pass filter 22 and the amplifier 24 has not occurred in the equalizer. Therefore, it is necessary to extra test circuit for the purpose of conducting a disconnection test by a direct current signal.
- Further, in the case of the circuit configuration of the equalizer 18 in Japanese Patent Application Laid-Open No. 2004-120468, the amplifier 24 is provided on the path for signal transmission subsequent to the high-pass filter and prior to the adder 20. In this case, the signal transmitted through the high-pass filter 22 and the amplifier 24 lags behind an original reception signal directly inputted into the adder 20, in reaching the adder 20 because of signal delay in circuits on the path. Therefore, the simultaneity of the two signals to be added to each other with the adder 20 is impaired. This makes it difficult to truly regenerate a transmission signal, causing a decrease in reproducibility of a transmission signal.
- An object of the present invention to provide an equalizer and a semiconductor device, that can suppress a decrease in S/N ratio of a reception signal, can facilitate a disconnection test by a direct current signal, and are excellent in reproducibility of a transmission signal.
- According to a first aspect of the present invention, an equalizer includes a low-pass filter, a subtraction unit, an addition unit and an amplifier.
- The low-pass filter receives a reception signal.
- The subtraction unit subtracts from the reception signal an output signal from the low-pass filter.
- The addition unit adds the reception signal to an output signal from the subtraction unit.
- The amplifier amplifies an output signal from the addition unit.
- According to the first aspect of the present invention, a signal mainly composed of a high frequency component, obtained by subtracting from the reception signal the output signal from the low-pass filter, is added to the reception signal, and the added signal is amplified. Thus, as compared with an equalizer where only the high frequency component of the reception signal is amplified and the amplified signal is added to the reception signal, the low frequency component of the reception signal is also amplified in the equalizer of the first aspect, thereby enabling suppression of a decrease in S/N ratio of the reception signal. Further, the equalizer of the first aspect adopts a low-pass filter rather than a high-pass filter. The low-pass filter is normally configured by a resistive element on a transmission path and a capacitive element suspended from one end of the resistive element, and one end of the resistive element on the transmission path is not insulated from the other end thereof. It is thereby possible to facilitate a disconnection test, by a direct current signal, on whether disconnection of the path among the reception end, the low-pass filter, the subtraction unit, the addition unit and the amplifier has not occurred. Furthermore, in the equalizer of the first aspect, the amplifier is arranged on the path subsequent to the addition unit. This prevents delay in a signal, which passes through the low-pass filter and the subtraction unit to reach the addition unit, due to the amplifier. Thereby, the simultaneity is easily maintained between the reception signal and the output signal from the subtraction unit which are computed in the addition unit. The equalizer is thus excellent in reproducibility of a transmission signal.
- According to a second aspect of the present invention, an equalizer includes a low-pass filter, a subtraction unit and amplifier.
- The low-pass filter receives a reception signal.
- The subtraction unit subtracts from the reception signal an output signal from the low-pass filter.
- The amplifier amplifies an output signal from the subtraction unit.
- According to the second aspect of the present invention, a signal mainly composed of a high frequency component, obtained by subtracting from the reception signal the output signal from the low-pass filter, is amplified. Thus, as compared with the case of amplifying only the high frequency component of the reception signal and then adding the amplified signal to the reception signal, the low frequency component of the reception signal is also amplified in the equalizer of the second aspect. It is thereby possible to suppress a decrease in S/N ratio of the reception signal. Further, the equalizer of the second aspect adopts a low-pass filter rather than a high-pass filter. The low-pass filter is normally configured by a resistive element on a transmission path and a capacitive element suspended from one end of the resistive element, and one end of the resistive element on the transmission path is not insulated from the other end thereof. It is thereby possible to facilitate a disconnection test, by a direct current signal, on whether disconnection of the path among the reception end, the low-pass filter, the subtraction unit and the amplifier has not occurred. Furthermore, in the equalizer of the second aspect, the amplifier is arranged on the path subsequent to the subtraction unit. This prevents delay in a signal, which passes through the low-pass filter to reach the subtraction unit, due to the amplifier. Thereby, the simultaneity is easily maintained between the reception signal and the output signal from the low-pass filter which are computed in the subtraction unit. The equalizer is thus excellent in reproducibility of a transmission signal. Furthermore, since requiring no adder, the equalizer of the second aspect can exert the same effect as the equalizer of the first aspect with a simpler circuit configuration than the equalizer of the first aspect.
- According to a third aspect of the present invention, an equalizer includes a signal conversion unit, a subtraction unit, an addition unit and amplifier.
- The signal conversion unit converts a reception signal into a signal in direct proportion to the reception signal to output the converted signal.
- The subtraction unit subtracts from the reception signal the output signal from the signal conversion unit.
- The addition unit adds the reception signal to an output signal from the subtraction unit.
- The amplifier amplifies an output signal from the addition unit.
- An input/output gain of the signal conversion unit is constant in a region not higher than a predetermined frequency in a frequency band of a signal component contained in the reception signal, and gradually decreases in a region exceeding the predetermined frequency.
- According to the third aspect of the present invention, an input/output gain of the signal conversion unit is constant in a region not higher than a predetermined frequency in a frequency band of a signal component contained in the reception signal, and gradually decreases in a region exceeding the predetermined frequency. Therefore, since the signal conversion unit serves an equivalent function to that of the low-pass filter, as in the case of the equalizer of the first aspect, the equalizer of the third aspect can suppress a decrease in S/N ratio of the reception signal. Further, as in the equalizer of the first aspect, there occurs no delay in the signal, which passes through the signal conversion unit and the subtraction unit to reach the addition unit, due to the amplifier. Thereby, the simultaneity is easily maintained between the reception signal and the output signal from the subtraction unit which are computed in the addition unit. The equalizer is thus excellent in reproducibility of a transmission signal. Moreover, it is also possible to convert the signal into a signal suitable for signal processing of the equalizer.
- According to a fourth aspect of the present invention, an equalizer includes a signal conversion unit, a subtraction unit and an amplifier.
- The signal conversion unit converts a reception signal into a signal in direct proportion to the reception signal to output the converted signal.
- The subtraction unit subtracts from the reception signal the output signal from the signal conversion unit.
- The amplifier amplifies an output signal from the addition unit.
- An input/output gain of the signal conversion unit is constant in a region not higher than a predetermined frequency in a frequency band of a signal component contained in the reception signal, and gradually decreases in a region exceeding the predetermined frequency.
- According to the fourth aspect of the present invention, an input/output gain of the signal conversion unit is constant in a region not higher than a predetermined frequency in a frequency band of a signal component contained in the reception signal, and gradually decreases in a region exceeding the predetermined frequency. Therefore, since the signal conversion unit serves an equivalent function to that of the low-pass filter, as in the case of the equalizer of the second aspect, the equalizer of the fourth aspect can suppress a decrease in S/N ratio of the reception signal. Further, as in the equalizer of the second aspect, there occurs no delay in the signal, which passes through the signal conversion unit to reach the subtraction unit, due to the amplifier. Thereby, the simultaneity is easily maintained between the reception signal and the output signal from the signal conversion unit which are computed in the subtraction unit. The equalizer is thus excellent in reproducibility of a transmission signal. The equalizer of the fourth aspect can exert the same effect as the equalizer of the first aspect with a simpler circuit configuration than the equalizer of the first aspect, as in the case of the equalizer of the second aspect. Moreover, it is also possible to convert the signal into a signal suitable for signal processing of the equalizer.
- These and other objects, features, aspects and advantages of the present invention will become more apparent from the following detailed description of the present invention when taken in conjunction with the accompanying drawings.
-
FIG. 1 shows a semiconductor circuit device including a transmission circuit and a semiconductor circuit device including a reception circuit, which are connected to each another through transmission paths; -
FIG. 2 is a circuit diagram showing a principle of an equalizer according to the present invention; -
FIG. 3 is a Bode diagram showing an operational principle of the equalizer ofFIG. 2 ; -
FIG. 4 is a circuit diagram showing an equalizer capable of realizing an equivalent function to that of the equalizer ofFIG. 2 with a simpler configuration than the equalizer ofFIG. 2 ; -
FIG. 5 is a circuit diagram showing an equalizer according to a first embodiment; -
FIG. 6 shows one example of a detailed configuration of a low-pass filter; -
FIG. 7 shows another example of the detailed configuration of the low-pass filter; -
FIG. 8 shows one example of a detailed configuration of a voltage-current signal conversion unit; -
FIG. 9 is a graph showing the relation between a voltage signal to be inputted into an input end and each of current signals to be outputted in the voltage-current signal conversion unit; -
FIG. 10 is a graph showing the relation between a frequency component contained in a voltage signal to be inputted into the input end and an input/output gain in the voltage-current signal conversion unit; -
FIG. 11 shows one example of a detailed configuration of an amplifier pre-stage unit; -
FIG. 12 shows another example of the detailed configuration of the amplifier pre-stage unit; -
FIG. 13 is a circuit diagram showing an equalizer according to a third embodiment; -
FIG. 14 shows one example of a detailed configuration of a voltage-voltage signal conversion unit; -
FIG. 15 is a graph showing the relation between a voltage signal to be inputted into the input end and each of voltages to be outputted in the voltage-voltage signal conversion unit; -
FIG. 16 is a graph showing the relation between a frequency component contained in a voltage signal to be inputted into the input end and an input/output gain in the voltage-voltage signal conversion unit; -
FIG. 17 shows one example of a detailed configuration of an amplifier pre-stage unit; -
FIG. 18 shows an equalizer according to a fourth embodiment; -
FIG. 19 is a sectional view showing a semiconductor chip including a semiconductor substrate on which an equalizer including an inductor element is formed; -
FIG. 20 is a top view showing the semiconductor substrate on which the equalizer is formed; -
FIG. 21 is a sectional view showing the semiconductor substrate on which the equalizer is formed; -
FIG. 22 is a top view of a semiconductor substrate of a semiconductor device according to a fifth embodiment; -
FIG. 23 is a sectional view of the semiconductor substrate of the semiconductor device according to the fifth embodiment; and -
FIG. 24 is a circuit diagram showing an equalizer according to a sixth embodiment. - Prior to the description of an equalizer according to the present invention, the principle of the present invention is described with reference to
FIGS. 1 to 4 . -
FIG. 1 shows asemiconductor circuit device 100 including atransmission circuit 101 and asemiconductor circuit device 300 including areception circuit 301, the devices being connected to each other throughtransmission paths semiconductor circuit devices transmission paths - A positive logic signal and a negative logic signal in a complementary relation to the positive logic signal are outputted from the
transmission circuit 101. The positive logic signal from thetransmission circuit 101 is transmitted to thereception circuit 301 through the positive logic-side transmission path 201 a. The negative logic signal from thetransmission circuit 101 is transmitted to thereception circuit 301 through the negative logic-side transmission path 201 b. The present invention is an equalizer to function as thisreception circuit 301. -
FIG. 1 shows a transmission signal (e.g., a signal of aligned information of “High”-“Low”-“High” . . .) outputted from thetransmission circuit 101, and a reception end signal received in thereception circuit 301. The waveform of the reception end signal appears to be dull owing to a noise or the like received on thetransmission paths -
FIG. 2 is a circuit diagram showing the principle of the equalizer according to the present invention. As shown inFIG. 2 , this equalizer includes a low-pass filter 4, asubtraction unit 5, anaddition unit 6 and anamplifier 2. - Further,
FIG. 3 is a Bode diagram showing the operational principle of the equalizer ofFIG. 2 . The operational principle of the equalizer according to the present invention is described usingFIG. 3 . - Receiving a reception signal (having a frequency characteristic shown with a graph of CH1 in
FIG. 3 ) given from areception end 1, the low-pass filter 4 outputs a signal (having a frequency characteristic shown with a graph of CH2 inFIG. 3 ) obtained by removing the high frequency component from the reception signal. - The
subtraction unit 5 subtracts from the reception signal the output signal from the low-pass filter 4. Thereby, an output signal from thesubtraction unit 5 has a frequency characteristic shown with a graph of CH3 inFIG. 3 . - The
addition unit 6 adds a reception signal from thereception end 1 to the output signal from thesubtraction unit 5. With this addition, an output signal from theaddition unit 6 has a frequency characteristic of emphasizing the high frequency component as shown with a graph of CH4 inFIG. 3 . Subsequently, theamplifier 2 amplifies the output signal from theaddition unit 6, and transmits it to anoutput end 3. - As thus described, according to the equalizer shown in
FIG. 2 , a signal (CH3) mainly composed of a high frequency component, obtained by subtracting an output signal (CH2) from the low-pass filter 4 from a reception signal (CH1) is added to the reception signal (CH1), and the signal (CH4) after addition is amplified. As compared with an equalizer where only the high frequency component of the reception signal is amplified and the amplified signal is added to the reception signal as in the above-described technique of Japanese Patent Application Laid-Open No. 2004-120468, the low frequency component of the reception signal (CH1) is amplified as well as the high frequency component (CH3) in the equalizer ofFIG. 2 . It is thereby possible to suppress a decrease in S/N ratio of the reception signal. - Further, the equalizer of
FIG. 2 adopts the low-pass filter 4 rather than a high-pass filter. The low-pass filter is normally configured by a resistive element on a transmission path and a capacitive element suspended from one end of the resistive element, and one end of the resistive element on the transmission path is not insulated from the other end thereof. It is thereby possible to facilitate a disconnection test, by a direct current signal, on whether disconnection of the path among thereception end 1, the low-pass filter 4, thesubtraction unit 5, theaddition unit 6 and theamplifier 2 has not occurred. - Furthermore, in the equalizer of
FIG. 2 , theamplifier 2 is arranged on the path subsequent to theaddition unit 6. This prevents delay in a signal, which passes through the low-pass filter 4 and thesubtraction unit 5 to reach theaddition unit 6, due to theamplifier 2. Thereby, the simultaneity is easily maintained between the reception signal from thereception end 1 and the output signal from thesubtraction unit 5 which are computed in theaddition unit 6. The equalizer is thus excellent in reproducibility of a transmission signal. - It is to be noted that
FIG. 4 shows an equalizer capable of realizing an equivalent function to that of the equalizer ofFIG. 2 with a simpler configuration than the equalizer ofFIG. 2 . As shown inFIG. 4 , this equalizer includes the low-pass filter 4, thesubtraction unit 5 and theamplifier 2 having the configuration of the equalizer ofFIG. 2 with theaddition unit 6 omitted therefrom. - Also in the equalizer of
FIG. 4 , the low-pass filter 4 receives a signal (graph of CH1 inFIG. 3 ) given from thereception end 1 and outputs a signal (graph of CH2 inFIG. 3 ) obtained by removing the high frequency component from the reception signal. However, a value of an input/output gain in a pass band in the low-pass filter 4 is set to a value less than one time. - The
subtraction unit 5 subtracts from the reception signal the output signal from the low-pass filter 4. Thereby, the output signal from thesubtraction unit 5 has a frequency characteristic shown with the graph of CH3 inFIG. 3 . However, since the value of the input/output gain of the pass band in the low-pass filter 4 is smaller than one time, the low frequency component is not completely lost in the output signal from thesubtraction unit 5. Consequently, the output signal from thesubtraction unit 5 has a frequency characteristic of emphasizing the high frequency component. - The
amplifier 2 then amplifies the output signal from thesubtraction unit 5, and transmits it to theoutput end 3. - As seen from the graphs of CH3 and CH4 in
FIG. 3 , so long as the low frequency component is not completely lost, it is possible to obtain the graph of CH4 by amplifying the graph of CH3 without addition of the graph of CH3 to the graph of CH1. - Hence, with the output signal from the
subtraction unit 5 directly amplified with theamplifier 2, a signal component other than the high frequency component as the main component is also amplified, without addition of the reception signal from theaddition unit 6 to the output signal from thesubtraction unit 5. - Namely, as compared with an equalizer where only the high frequency component of the reception signal is amplified and the amplified signal is added to the reception signal as in the above-described technique of Japanese Patent Application Laid-Open No. 2004-120468, the low frequency component of the reception signal is amplified as well as the high frequency component in the equalizer of
FIG. 4 , as in the equalizer ofFIG. 2 . It is thereby possible to suppress a decrease in S/N ratio of the reception signal. - Further, the equalizer of
FIG. 4 also adopts the low-pass filter 4 rather than a high-pass filter. It is thereby possible to facilitate a disconnection test, by a direct current signal, on whether disconnection of the path among thereception end 1, the low-pass filter 4, thesubtraction unit 5 and theamplifier 2 has not occurred. - Further, also in the equalizer of
FIG. 4 , theamplifier 2 is arranged on the path subsequent to thesubtraction unit 5. This prevents delay in a signal, which passes through the low-pass filter 4 to reach thesubtraction unit 5, due to theamplifier 2. Thereby, the simultaneity is easily maintained between the reception signal from thereception end 1 and the output signal from the low-pass filter 4 which are computed in thesubtraction unit 5. The equalizer is thus excellent in reproducibility of a transmission signal. - Furthermore, since not requiring the
addition unit 6, the equalizer ofFIG. 4 can exert the same effect as the equalizer ofFIG. 2 with a simpler circuit configuration than the equalizer ofFIG. 2 . - In the following first to fourth embodiments, more specific description of the circuit configuration is given based upon the equalizer of
FIG. 4 . -
FIG. 5 is a circuit diagram showing an equalizer according to this embodiment, which functions as areception circuit 301 a. This equalizer also includes the low-pass filter 4 and theamplifier 2 which are shown inFIG. 4 . Further,adders FIG. 5 correspond to theaddition unit 5 ofFIG. 4 . Moreover, the equalizer ofFIG. 5 includes voltage-currentsignal conversion units resistors adders resistors pre-stage unit 50. - In this embodiment, for the purpose of specifically showing a transmission path for a positive logic signal and a transmission path for a negative logic signal, a reception end la for receiving a voltage signal of a positive logic and a reception end lb for receiving a voltage signal of a negative logic are clearly specified. In correspondence to these reception ends, two respective signal paths arranged after the reception ends are shown with two lines.
- In
FIG. 5 , a voltage signal S1 of the positive logic and a voltage signal S2 of the negative logic, which are received at the reception ends 1 a, 1 b, respectively, are inputted into the low-pass filter 4, and also inputted into the voltage-currentsignal conversion unit 60 a. - The voltage-current
signal conversion unit 60 a converts the received voltage signals S1, S2 into current signals S6, S7 in direct proportion to the voltage signals, and the converted signals are respectively outputted to theadders subtraction unit 5. - Further, the voltage-current
signal conversion unit 60 b converts output voltage signals S4, S5 from the low-pass filter 4 into current signals S8, S9 in direct proportion to the output voltage signals, and the converted current signals S8, S9 are respectively outputted to theadders subtraction unit 5. - In the
adder 51 a, the current signal S6 of the positive logic from the voltage-currentsignal conversion unit 60 a is added to the current signal S8 from the voltage-currentsignal conversion unit 60 b. In theadder 51 b, the current signal S7 of the negative logic from the voltage-currentsignal conversion unit 60 a is added to the current signal S9 from the voltage-currentsignal conversion unit 60 b. - It is described here that the
adders signal conversion unit 60 b, the voltage signal S1 of the positive logic and the voltage signal S2 of the negative logic, having passed through the low-pass filter 4 and reversed in polarity, are respectively inputted as the voltage signals S5, S4 into the positive signal input end (indicated as “+”) and the negative signal input end (indicated as “−”). - Since the positive logic and the negative logic of the converted current signals S9, S8 do not differ from those of the voltage signals S5, S4 before conversion, eventually, the current signal S6 of the positive logic is added to the current signal S8 of the negative logic, while the current signal S7 of the negative logic is added to the current signal S9 of the positive logic. Namely, such additions correspond to subtractions of signals S5, S4 after passage through the low-
pass filter 4 respectively from the received voltage signals S1, S2. - In the
adders amplifier 2. Output signals S13, S14 from theamplifier 2 are respectively given to output ends 3 a, 3 b. - It is to be noted that one ends of the
resistors amplifier 2, and a power supply potential VDD is supplied to the other ends of theresistors - With
such resistors resistors amplifier 2. Hence it is possible to set a frequency characteristic of the equalizer according to a characteristic of the transmission path for the reception signal. -
FIG. 6 shows one example of a specific configuration of the low-pass filter 4. It should be noted that the low-pass filter ofFIG. 6 is differentiated by provision of a symbol “4 a”. - The low-
pass filter 4 a is configured by four Nch-MOS transistors (N-channel metal oxide semiconductor transistor) N1 a, N2 a, N1 b, N2 b. - The voltage signal S1 of the positive logic is given to one of a drain and a source of the Nch-MOS transistor N1 a. Further, the voltage signal S2 of the negative logic is given to one of a drain and a source of the Nch-MOS transistor N1 b.
- The other one of the drain and the source of the Nch-MOS transistor N1 a outputs the voltage signal S5 of the positive logic. Further, the other one of the drain and the source of the Nch-MOS transistor N1 a is connected to a gate of the Nch-MOS transistor N2 a. The other one of the drain and the source of the Nch-MOS transistor N1 b outputs the voltage signal S4 of the negative logic. Further, the other one of the drain and the source of the Nch-MOS transistor N1 b is connected to a gate of the Nch-MOS transistor N2 b.
- Further, a source and a drain of the Nch-MOS transistor N2 a are short-circuited, and supplied with a ground potential GND. Similarly, a source and a drain of the Nch-MOS transistor N2 b are short-circuited, and supplied with a ground potential GND.
- The voltage signal S3 is given from the outside to the gates of the Nch-MOS transistors N1 a and N1 b. The strength of the voltage signal S3 is variable in the triode region (linear region) of the Nch-MOS transistors N1 a and N1 b.
- With the above-mentioned connection configuration, both the Nch-MOS transistors N2 a and N2 b function as capacitive elements, and both the Nch-MOS transistors N1 a and N1 b function as variable resistive elements. Namely, the low-
pass filter 4 a is a typical low-pass filter configured by a resistive element on a transmission path and a capacitive element suspended from one end of the resistive element. - As thus described, when the strength of the voltage signal S3 is variable in the triode region of the Nch-MOS transistors N1 a and N1 b, a cutoff frequency of the low-
pass filter 4 a is variable. - Accordingly, appropriate setting of the value of the cutoff frequency of the low-
pass filter 4 a allows adjustment of the strengths of the signals S10, S11 to be inputted into theamplifier 2, and it is thus possible to set the frequency characteristic of the equalizer according to characteristics of thetransmission paths - It is to be noted that
FIG. 7 shows another example of the detailed configuration of the low-pass filter 4. The low-pass filter ofFIG. 7 is differentiated by provision of a symbol “4 b”. - In the low-pass filter 4 b, the power supply potential VDD is given, in place of the voltage signal S3, to the gate of the Nch-MOS transistors N1 a, N1 b. Except for this respect, the low-pass filter 4 b has the same configuration as the low-
pass filter 4 a ofFIG. 6 . - Namely, in the low-pass filter 4 b of
FIG. 7 , a cutoff frequency is a fixed value. Therefore, appropriate designing of the gate size, an injection amount of the impurity in the channel region, and the like in terms of the Nch-MOS transistors N1 a, N1 b enables configuration of a control-free low-pass filter. -
FIG. 8 shows one example of a detailed configuration of the voltage-currentsignal conversion unit 60 a capable of converting the received voltage signals S1, S2 into the current signals S6, S7 in direct proportion to the voltage signals, and the voltage-currentsignal conversion unit 60 b capable of converting the received voltage signals S4, S5 into the current signals S8, S9 in direct proportion to the voltage signals. - It should be noted that an input/output gain of the voltage-current
signal conversion unit 60 a is constant in a frequency band of a signal component contained in the received voltage signals S1, S2, and the input/output gain of the voltage-currentsignal conversion unit 60 b is constant in a frequency band of a signal component contained in the received voltage signals S4, S5. - Moreover, the voltage-current
signal conversion units signal conversion unit 60 a, whereas the input signals are the voltage signals S4, S5 and the output signals are the current signals S8, S9 in the voltage-currentsignal conversion unit 60 b. Therefore, only the circuit configuration of the voltage-currentsignal conversion unit 60 a is shown inFIG. 8 , with each signal in the case of the circuit configuration of the voltage-currentsignal conversion unit 60 b shown in parenthesis. Also in the following description, each signal in the case of the voltage-currentsignal conversion unit 60 b is shown in parenthesis, as inFIG. 8 . - The voltage-current
signal conversion unit 60 a (60 b) is configured by six Pch-MOS transistors P1 to P6 and two Nch-MOS transistors N3, N4. - The voltage signal S1 of the positive logic (the voltage signal S4 of the negative logic) is given to a gate of the Pch-MOS transistor P5. Further, the voltage signal S2 of the negative logic (the voltage signal S5 of the positive logic) is given to a gate of the Pch-MOS transistor P6.
- A drain of the Pch-MOS transistor P1 is connected to a source of the Pch-MOS transistor P5. Further, a drain of the Pch-MOS transistor P2 is connected to a source of the Pch-MOS transistor P6. The power supply potential VDD is supplied to both sources of the Pch-MOS transistors P1 and P2. A bias potential “bias” is supplied to both gates of the Pch-MOS transistors P1, P2.
- A drain of the Pch-MOS transistor P3 is connected to a drain of the Pch-MOS transistor P4 and, also, to the source of the Pch-MOS transistor P5. Further, a source of the Pch-MOS transistor P3 is connected to a source of the Pch-MOS transistor P4 and, also, to the source of the Pch-MOS transistor P6.
- A gate of the Pch-MOS transistor P3 is connected to the gate of the Pch-MOS transistor P5, and a gate of the Pch-MOS transistor P4 is connected to a gate of the Pch-MOS transistor P6.
- The drain of the Pch-MOS transistor P5 is connected to a drain of the Nch-MOS transistor N3 and, also, to a gate of the Nch-MOS transistor N3. Further, a drain of the Pch-MOS transistor P6 is connected to a drain of the Nch-MOS transistor N4 and, also, to a gate of the Nch-MOS transistor N4. The ground potential GND is supplied to both sources of the Nch-MOS transistors N3, N4.
- Respective currents In, Ip in the drains of the Nch-MOS transistors N3, N4 are current signals S6 (S8), S7 (S9) as output signals from the voltage-current
signal conversion unit 60 a (60 b). -
FIG. 9 is a graph showing the relation between the voltage signal S1 (S4) to be inputted into an input end PI and each of the currents In, Ip, i.e., the current signals S6 (S8), S7 (S9), in the voltage-currentsignal conversion unit 60 a (60 b). - With the circuit configuration of
FIG. 8 adopted, the gate size, an injection amount of the impurity in the channel region, and the like, in terms of the Pch-MOS transistors P1 to P6 and Nch-MOS transistors N3, N4, are appropriately designed so as to obtain the currents In, Ip, i.e., the current signals S6 (S8), S7 (S9), in direct proportion to the voltage signal S1 (S4), as shown inFIG. 9 . - It is to be noted that the relation between the voltage signal S2 (S5) to be inputted into an input end NI and each of the currents In, Ip, i.e., the current signals S6 (S8), S7 (S9), is the same as the relation shown in
FIG. 9 . -
FIG. 10 is a graph showing the relation between a frequency component contained in the voltage signals S1 (S4), S2 (S5), and an input/output gain between the voltage signals S1 (S4), S2 (S5) and the current signals S6 (S8), S7 (S9), in the voltage-currentsignal conversion unit 60 a (60 b). - With the circuit configuration of
FIG. 8 adopted, the gate size, an injection amount of the impurity in the channel region, and the like, in terms of the Pch-MOS transistors P1 to P6 and the Nch-MOS transistors N3, N4, are appropriately designed so as to keep the input/output gain of the voltage-currentsignal conversion unit 60 a (60 b) constant in a frequency band of a signal component contained in the received voltage signals S1 (S4), S2 (S5), as shown inFIG. 10 . - As thus described, the voltage-current
signal conversion unit 60 a (60 b) converts the voltage signals S1 (S4), S2 (S5) to be inputted, in a state where the input/output gain is constant in a frequency band of a signal component contained in the voltage signals S1 (S4), S2 (S5), into the current signals S6 (S8), S7 (S9), and then outputs the converted current signals S6 (S8), S7 (S9). - Accordingly, even when the operating power supply voltage of the equalizer is small, a reception signal can be reduced to a signal suitable for the operating power supply voltage, and thus the reception signal can be converted into a signal suitable for signal processing in the equalizer. Alternatively, even when the amplitude of a reception signal is small and thus monotonous, the signal can be magnified to a signal with a large amplitude, and thus can be converted into a signal suitable for signal processing in the equalizer.
-
FIG. 11 shows the detailed configuration of the amplifierpre-stage unit 50. It is to be noted that the amplifier pre-stage unit ofFIG. 11 is differentiated by provision of a symbol “50 a”. - The amplifier
pre-stage unit 50 a is configured by two groups each including two Pch-MOS transistors P7 a, P7 b and four Nch-MOS transistors N5 a, N6 a, N5 b, N6 b. Among them, two groups each including the Pch-MOS transistors P7 a, P7 b and the Nch-MOS transistors N5 a, N5 b, configure theresistors adders - It should be noted that the
adder 51 a adds the current signal S6 to the current signal S8, and reconverts the addition result into the voltage signal S10, to be outputted. Further, theadder 51 b adds the current signal S7 to the current signal S9, and reconverts the addition result into the voltage signal S11, to be outputted. - Further, the
resistors adders resistor 52 a and theadder 51 a, whereas the input signals are the current signals S7, S9 and the output signal is the voltage signal S11 in theresistor 52 b and theadder 51 b. Therefore, only the circuit configurations of theresistor 52 a and theadder 51 a are shown inFIG. 11 , with each signal in the case of the circuit configuration of theresistor 52 b and theadder 51 b shown in parenthesis. Also in the following description, respective signals in the case of theresistor 52 b and theadder 51 b are shown in parenthesis, as inFIG. 11 . - The current signal S6 of the positive logic (current signal S7 of the negative logic) is given to a gate of the Nch-MOS transistor N6 a. Further, the current signal S8 of the negative logic (the current signal S9 of the positive logic) is given to a gate of the Nch-MOS transistor N6 b. The ground potential GND is supplied to both sources of the Nch-MOS transistors N6 a, N6 b.
- A drain of the Nch-MOS transistor N6 a is connected to a source of the Nch-MOS transistor N5 a while a source of the Nch-MOS transistor N5 b is connected to a drain of the Nch-MOS transistor N6 b. Further, a drain of the Pch-MOS transistor P7 a is connected to a drain of the Nch-MOS transistor N5 a while a drain of the Pch-MOS transistor N7 b is connected to a drain of the Nch-MOS transistor N5 b. Moreover, the power supply potential VDD is supplied to both sources of the Pch-MOS transistors P7 a, P7 b.
- Drains of the Pch-MOS transistors P7 a, P7 b are also connected to each other, and a voltage at this node is the voltage signal S10 (S11) as an output signal from the amplifier
pre-stage unit 50 a to theamplifier 2. - The voltage signal S12 a (S12 b) is given from the outside to gates of the Pch-MOS transistors P7 a and P7 b. Moreover, a voltage signal /S12 a (/S12 b) as a reversed signal of the voltage signal S12 a (S12 b) is also given from the outside to gates of the Nch-MOS transistors N5 a and N5 b. The strengths of the voltage signals S12 a (S12 b), /S12 a (/S12 b) are variable in the triode region (linear region) of the Pch-MOS transistors P7 a and P7 b as well as the Nch-MOS transistors N5 a and N5 b. It is to be noted that the voltage signal S12 is composed of the voltage signals S12 a and S12 b.
- According to the above-mentioned connection configuration, when the strengths of the voltage signal S12 a (S12 b), /S12 a (/S12 b) are variable in the triode region of the Pch-MOS transistors P7 a and P7 b as well as the Nch-MOS transistors N5 a and N5 b, resistance values of the
resistors - Accordingly, appropriate setting of the resistance values of the
variable resistors amplifier 2, and it is thus possible to set the frequency characteristic of the equalizer according to characteristics of thetransmission paths - It is to be noted that
FIG. 12 shows another example of the detailed configuration of the amplifierpre-stage unit 50. The amplifier pre-stage unit ofFIG. 12 is differentiated by provision of a symbol “50 b”. - In the amplifier
pre-stage unit 50 b, the Nch-MOS transistors N5 a and N5 b in the amplifierpre-stage unit 50 a ofFIG. 11 are omitted. The drain of the Nch-MOS transistor N6 a is directly connected with the drain of the Pch-MOS transistor P7 a, and the drain of the Nch-MOS transistor N6 b is directly connected with the drain of the Pch-MOS transistor P7 b. - Further, the ground potential GND is supplied, in place of the voltage signal S12 a, to the gates of the Pch-MOS transistors P7 a and P7 b. The Pch-MOS transistors P7 a and P7 b configure a resistor 52 c (52 d), and the Nch-MOS transistors N6 a and N6 b configure an adder 51 c (51 d).
- Except for these respects, the amplifier
pre-stage unit 50 b ofFIG. 12 has the same configuration as the amplifierpre-stage unit 50 a ofFIG. 11 . - Namely, in the amplifier
pre-stage unit 50 b ofFIG. 12 , resistance values of theresistors 52 c, 52 d are fixed values. Therefore, appropriate designing of the gate size, an injection amount of the impurity in the channel region, and the like in terms of the Pch-MOS transistors P7 a and P7 b enables configuration of a control-free resistor. - A second embodiment is a modification of the equalizer according to the first embodiment, obtained by omitting the low-
pass filter 4 in the first embodiment and instead providing a low-pass filter function to the voltage-currentsignal conversion unit 60 b. - As shown in
FIG. 10 , appropriate designing of the gate size, an injection amount of the impurity in the channel region, and the like in terms of the Pch-MOS transistors P1 to P6 and Nch-MOS transistors N3, N4 which configure the voltage-currentsignal conversion unit 60 b can keep the input/output gain of the voltage-currentsignal conversion unit 60 a (60 b) constant in a frequency band of a signal component contained in the received voltage signals S1 (S4), S2 (S5). - In this embodiment, the gate size, an injection amount of the impurity in the channel region, and the like in terms of each MOS transistor of the voltage-current
signal conversion unit 60 b are designed such that the input/output gain of the voltage-currentsignal conversion unit 60 b is constant in a region not higher than a predetermined frequency in the frequency band of the signal component contained in the received voltage signals S4, S5, and gradually decreases in a region exceeding the predetermined frequency. - Descriptions of the other respects are omitted since those respects are the same as in the case of the equalizer according to the first embodiment.
- According to this embodiment, since the voltage-current
signal conversion unit 60 b serves an equivalent function to the low-pass filter 4, it is possible to suppress a decrease in S/N ratio of the reception signal, as in the case of the equalizer according to the first embodiment. - Further, as in the case of the equalizer according to the first embodiment, there occurs no delay in the signal, which passes through the voltage-current
signal conversion unit 60 b to reach thesubtraction unit 5, due to theamplifier 2, thereby making it easy to maintain the simultaneity between the reception signal and the output signal from the voltage-currentsignal conversion unit 60 b which are computed in thesubtraction unit 5. The equalizer of this embodiment is thus excellent in reproducibility of a transmission signal. - Further, as in the case of the equalizer of
FIG. 4 , the equalizer according to this embodiment can exert the same effect as the equalizer ofFIG. 2 with a simpler circuit configuration than the equalizer ofFIG. 2 . Moreover, the equalizer can convert the signal into a signal suitable for signal processing of the equalizer by means of the voltage-currentsignal conversion unit 60 b. - A third embodiment is also a modification of the equalizer according to the first embodiment, where voltage-voltage signal conversion units are adopted in place of the voltage-current
signal conversion units -
FIG. 13 is a circuit diagram showing the equalizer according to this embodiment which functions as areception circuit 301 b. This equalizer also includesadders 51 e, 51 f similar to theadders resistors pass filter 4, and theamplifier 2. It should be noted that theadders 51 e, 51 f and theresistors pre-stage unit 50 c. - However, the equalizer of
FIG. 13 includes voltage-voltagesignal conversion units signal conversion units - The voltage-voltage
signal conversion unit 61 a converts the received voltage signals S1, S2 into voltage signals S6 a, S7 a in direct proportion to the received voltage signals, and outputs the converted signals respectively to theadders 51 e, 51 f as thesubtraction unit 5. - Further, the voltage-voltage
signal conversion unit 61 b converts the output voltage signals S4, S5 from the low-pass filter 4 into voltage signals S8 a, S9 a in direct proportion to the output voltage signals, and outputs the converted signals S8 a, S9 a respectively to theadders 51 e, 51 f as thesubtraction unit 5. - In the adder 51 e, a voltage signal S6 a of the positive logic from the voltage-voltage
signal conversion unit 61 a is added to a voltage signal S8 a from the voltage-voltagesignal conversion unit 61 b. In theadder 51 f, while a voltage signal S7 a of the negative logic from the voltage-voltagesignal conversion unit 61 a is added to a voltage signal S9 a from the voltage-voltagesignal conversion unit 61 b. - Voltage signals S10 a, S11 a as the addition results are outputted from the
adders 51 e, 51 f. The voltage signals S10 a, S11 a are respectively inputted into the input ends of theamplifier 2. The output signals S13, S14 from theamplifier 2 are respectively given to the output ends 3 a, 3 b. - It is to be noted that one ends of the
resistors amplifier 2, and the power supply potential VDD is supplied to the other ends of theresistors -
FIG. 14 shows one example of a detailed configuration of the voltage-voltagesignal conversion unit 61 a capable of converting the received voltage signals S1, S2 into the voltage signals S6 a, S7 a in direct proportion to the received voltage signals, and the voltage-voltagesignal conversion unit 61 b capable of converting the received voltage signals S4, S5 into the voltage signals S8 a, S9 a in direct proportion to the received voltage signals. - It should be noted that an input/output gain of the voltage-voltage
signal conversion unit 61 a is constant in a frequency band of a signal component contained in the received voltage signals S1, S2, and the input/output gain of the voltage-voltagesignal conversion unit 61 b is constant in a frequency band of a signal component contained in the received voltage signals S4, S5. - Moreover, the voltage-voltage
signal conversion units signal conversion unit 61 a, whereas the input signals are the voltage signals S4, S5 and the output signals are the voltage signals S8 a, S9 a in the voltage-voltagesignal conversion unit 61 b. Therefore, only the circuit configuration of the voltage-voltagesignal conversion unit 61 a is shown inFIG. 14 , with each signal in the case of the circuit configuration of the voltage-voltagesignal conversion unit 61 b shown in parenthesis. Also in the following description, each signal in the case of the voltage-voltagesignal conversion unit 61 b is shown in parenthesis, as inFIG. 14 . - The voltage-voltage
signal conversion unit 61 a (61 b) is configured by six Pch-MOS transistors P1 to P6 and two Nch-MOS transistors N3, N4. - The voltage signal S1 of the positive logic (voltage signal S4 of the negative logic) is given to a gate of the Pch-MOS transistor P5. Further, the voltage signal S2 of the negative logic (the voltage signal S5 of the positive logic) is given to a gate of the Pch-MOS transistor P6.
- A drain of the Pch-MOS transistor P1 is connected to a source of the Pch-MOS transistor P5. Further, a drain of the Pch-MOS transistor P2 is connected to a source of the Pch-MOS transistor P6. The power supply potential VDD is supplied to both sources of the Pch-MOS transistors P1 and P2. The bias potential “bias” is supplied to both gates of the Pch-MOS transistors P1, P2.
- A drain of the Pch-MOS transistor P3 is connected to a drain of the Pch-MOS transistor P4 and, also, to the source of the Pch-MOS transistor P5. Further, a source of the Pch-MOS transistor P3 is connected to a source of the Pch-MOS transistor P4 and, also, to the source of the Pch-MOS transistor P6.
- A gate of the Pch-MOS transistor P3 is connected to a gate of the Pch-MOS transistor P5, and a gate of the Pch-MOS transistor P4 is connected to a gate of the Pch-MOS transistor P6.
- A drain of the Pch-MOS transistor P5 is connected to a drain of the Nch-MOS transistor N3. The power supply potential VDD is supplied to a gate of the Nch-MOS transistor N3. A drain of the Pch-MOS transistor P6 is connected to a drain of the Nch-MOS transistor N4. The power supply potential VDD is supplied to a gate of the Nch-MOS transistor N4. The ground potential GND is supplied to both sources of the Nch-MOS transistors N3, N4.
- Respective voltages NO, PO in the drains of the Nch-MOS transistors N3, N4 are voltage signals S6 a (S8 a), S7 a (S9 a) as output signals from the voltage-voltage
signal conversion unit 61 a (61 b). -
FIG. 15 is a graph showing the relation between the voltage signal S1 (S4) to be inputted into the input end PI and each of the voltages NO, PO, i.e., the voltage signals S6 a (S8 a), S7 a (S9 a), in the voltage-voltagesignal conversion unit 61 a (61 b). - With the circuit configuration of
FIG. 14 adopted, the gate size, an injection amount of the impurity in the channel region, and the like, in terms of the Nch-MOS transistors N3, N4 and the Pch-MOS transistors P1 to P6, are appropriately designed so as to obtain the voltages NO, PO, i.e., the voltage signals S6 a (S8 a), S7 a (S9 a), in direct proportion to the voltage signal S1 (S4), as shown inFIG. 15 . - It is to be noted that the relation between the voltage signal S2 (S5) to be inputted into the input end NI and each of the voltages NO, PO, i.e., the voltage signals S6 a (S8 a), S7 a (S9 a), is the same as the relation shown in
FIG. 15 . -
FIG. 16 is a graph showing the relation between a frequency component contained in the voltage signals S1 (S4), S2 (S5), and an input/output gain between the voltage signals S1 (S4), S2 (S5) and the voltage signals S6 a (S8 a), S7 a (S9 a), in the voltage-voltagesignal conversion unit 61 a (61 b). - With the circuit configuration of
FIG. 14 adopted, the gate size, an injection amount of the impurity in the channel region, and the like, in terms of the Pch-MOS transistors P1 to P6 and the Nch-MOS transistors N3, N4, are appropriately designed so as to keep the input/output gain of the voltage-voltagesignal conversion unit 61 a (61 b) constant in a frequency band of a signal component contained in the received voltage signals S1 (S4), S2 (S5), as shown inFIG. 16 . - As thus described, the voltage-voltage
signal conversion unit 61 a (61 b) converts the voltage signals S1 (S4), S2 (S5) to be inputted, in a state where the input/output gain is constant in a frequency band of a signal component contained in the voltage signals S1 (S4), S2 (S5), into the voltage signals S6 a (S8 a), S7 a (S9 a), and then outputs the converted voltage signals S6 a (S8 a), S7 a (S9 a). - Accordingly, even in the case of converting a signal into a voltage signal rather than a current signal, a reception signal can be reduced, or magnified to a signal with a large amplitude, and thus can be converted into a signal suitable for signal processing in the equalizer.
-
FIG. 17 is an example showing the detailed configuration of the amplifierpre-stage unit 50. It is to be noted that the amplifier pre-stage unit ofFIG. 17 is differentiated by provision of a symbol “50 c”. - The amplifier
pre-stage unit 50 c is configured by two groups each including a Pch-MOS transistors P7 c and two Nch-MOS transistors N6 c, N6 d. Among them, two groups of the Pch-MOS transistor P7 c configure theresistors adders 51 e, 51 f. - It should be noted that the adder 51 e adds the voltage signal S6 a to the voltage signal S8 a, and reconverts the addition result into the voltage signal S10 a, to be outputted. Further, the
adder 51 f adds the voltage signal S7 a to the voltage signal S9 a, and reconverts the addition result into the voltage signal S11 a, to be outputted. - Further, the
resistors adders 51 e, 51 f have an equivalent circuit configuration, except for the following differences. The input signals are the voltage signals S6 a, S8 a and the output signal is the voltage signal S10 a in theresistor 52 e and the adder 51 e, whereas the input signals are the voltage signals S7 a, S9 a and the output signal is the voltage signal S11 a in theresistor 52 f and theadder 51 f. Therefore, only the circuit configurations of theresistor 52 e and the adder 51 e are shown inFIG. 17 , with each signal in the case of the circuit configuration of theresistor 52 f and theadder 51 f shown in parenthesis. Also in the following description, each signal in the case of theresistor 52 f and theadder 51 f is shown in parenthesis, as inFIG. 17 . - The voltage signal S6 a of the positive logic (voltage signal S7 a of the negative logic) is given to a gate of the Nch-MOS transistor N6 c. Further, the voltage signal S8 a of the negative logic (the voltage signal S9 a of the positive logic) is given to a gate of the Nch-MOS transistor N6 d. The ground potential GND is supplied to both sources of the Nch-MOS transistors N6 c, N6 d.
- A drain of the Nch-MOS transistor Plc is connected to both drains of the Nch-MOS transistor N6 c, N6 d. Further, the power supply potential VDD is supplied to a source of the Pch-MOS transistor P7 c.
- A voltage of a drain of the Pch-MOS transistor P7 c is the voltage signal S10 a (S11 a) as an output signal from the amplifier
pre-stage unit 50 c to theamplifier 2. - Namely, in the amplifier
pre-stage unit 50 c ofFIG. 17 , resistance values of theresistors - It should be noted that the signal S12 shown in
FIG. 5 may be given to a gate of the Pch-MOS transistor P7 c so as to make theresistors - Further, this embodiment may be applied to the equalizer according to the second embodiment as well as the equalizer according to the first embodiment.
- A fourth embodiment is also a modification of the equalizer according to the first embodiment, where the voltage signal S12 for determining resistance values of the
variable resistors pass filter 4 a in the first embodiment are automatically generated to be the optimum values. -
FIG. 18 shows an equalizer according to this embodiment. As shown inFIG. 18 , the equalizer according to this embodiment further includes a bit error rate tester (BERT) for measuring bit error rates of the reception signals S1, S2 upon receiving the output signals S13, S14 from theamplifier 2, or aneye pattern detector 70 for detecting eye patterns of the reception signals S1, S2 upon receiving the output signals S13, S14 from theamplifier 2. - The bit error rate tester or the
eye detector 70 transmits a resistance value adjustment signal as the voltage signal S12 for adjusting resistance values of thevariable resistors pass filter 4 a. - When the
block 70 is the bit error rate tester, the output signals S3 and S12 as the resistance value adjustment signal and the cutoff frequency adjustment signal are generated so as to minimize the bit error rates of the reception signals S1, S2. Further, when theblock 70 is the eye pattern detector, the output signals S3 and S12 as the resistance value adjustment signal and the cutoff frequency adjustment signal are generated so as to maximize the areas of the eye patterns of the reception signals S1, S2. - Hence it is possible to optimally set the resistance values of the
variable resistors pass filter 4 a, so as to automatically set the frequency characteristic of the equalizer according to the characteristic of the transmission path for the reception signal. - It is to be noted that, although this embodiment has shown the case where the bit error rate tester or the
eye pattern detector 70 outputs both the resistance value adjustment signal and the cutoff frequency adjustment signal, another configuration example may be to output either of the two signals. - Further, this embodiment may be applied to the equalizer according to the second or third embodiment as well as the equalizer according to the first embodiment.
- A fifth embodiment is a semiconductor device on which an equalizer according to any one of the first to fourth embodiments is mounted.
- An equalizer including an inductor element is shown in each of
FIGS. 3 , 9 and 12 of Japanese Patent Application Laid-Open No. 2003-168944. The case of mounting such an equalizer on a semiconductor device is considered. -
FIG. 19 is a sectional view showing a semiconductor chip as a semiconductor device including a semiconductor substrate on which an equalizer including an inductor element is formed, and a flip chip package containing the semiconductor substrate. This semiconductor chip has a package substrate Sp with a solder ball SB formed on the one-side main face thereof, a package ring PR formed on the periphery of the other-side main surface of the package substrate Sp, and a heat sink RB bonded to the package ring PR. The package substrate Sp, the package ring PR and the heat sink RB configure the flip chip package. - One-side main face of a semiconductor substrate Ss on which the equalizer is formed is bonded to the heat sink RB via a resin RS2. An interlayer insulation film IL is formed on the other-side main face of the semiconductor substrate Ss. A plurality of bumps BP are formed on the surface of the interlayer insulation film IL. The plurality of bumps BP are electrically connected to respective prescribed portions on the package substrate Sp. It is to be noted that a resin RS1 is formed on the peripheries of the plurality of bumps BP and on the surface and in the vicinity of the interlayer insulation film IL, so as to solidify the connection between the plurality of bumps BP and the package substrate Sp.
-
FIG. 20 is a top view seen from the main face (interlayer insulation film IL forming side) of the semiconductor substrate Ss on which the equalizer is formed. Further,FIG. 21 is a sectional view taken along a cutting line XXI-XXI inFIG. 20 . - As shown in
FIGS. 20 , 21, an equalizer EQ is formed on the surface of the semiconductor substrate Ss. The interlayer insulation film IL is formed so as to cover the surface of the semiconductor substrate Ss and the equalizer EQ. Conductive pads PD are formed on the interlayer insulation film IL, and each of the bumps BP is formed on each of the pads PD. It should be noted that the pads PD and the equalizer EQ are electrically connected to each other through wiring WR in the interlayer insulation film IL. - The equalizer EQ includes a ring-like inductor element ID. The inductor element ID generates a magnetic field MF at the time of operation of the circuit. In the case of the equalizer EQ using the inductor element ID, if the flip chip package is adopted, the magnetic field MF generated by inductor element ID is interfered with the bumps BP of the flip chip package, leading to fluctuation in frequency characteristic of the equalizer EQ. Further, in the case of arranging a plurality of equalizers EQ in one package, variations tend to occur in circuit characteristic among the equalizers EQ due to the positional relation between the inductor element ID in each of the equalizers EQ and the bumps BP.
- Since an equalizer according to any one of the first to fourth embodiments is mounted on the semiconductor device according to this embodiment, the equalizer includes no inductor element (see configurations of the low-pass filters in
FIGS. 6 and 7 ). Therefore, the above-mentioned problems that may arise in the case of mounting an equalizer including an inductor element on a semiconductor device will not arise in the semiconductor device according to this embodiment. -
FIG. 22 is a top view of the semiconductor substrate of the semiconductor device according to this embodiment.FIG. 23 is a sectional view taken along a cutting line XXIII-XXIII inFIG. 22 . Symbols used inFIGS. 22 and 23 are respectively the same as those shown inFIGS. 20 and 21 , and descriptions of those symbols are thus omitted. - In
FIGS. 22 and 23 , the inductor element ID is not formed in the equalizer EQ. This equalizer EQ is an equalizer according to any one of the first to fourth embodiments. Namely, the equalizer EQ having no inductor element is formed on the surface of the semiconductor substrate Ss, as in the cases of the low-pass filters 4 a and 4 b inFIGS. 6 and 7 . The flip chip package, configured by the package substrate Sp, the package ring PR and the heat sink RB, contains the semiconductor substrate Ss. - There are problems with the equalizer using the inductor element in that adoption of the flip chip package causes the magnetic field generated by the inductor element to fluctuate the frequency characteristic of the equalizer EQ, and in that variations tend to occur in circuit characteristic among the equalizers due to the positional relation between the inductor element in each of the equalizers and the bumps. However, in the present invention, those problems do not arise since the equalizer EQ includes no inductor element. Further, the problem of increasing a circuit layout area does not arise either, since the equalizer EQ includes no inductor element.
- A sixth embodiment is also a modification of the equalizer according to the first embodiment, where capacitors are provided between the
reception end 1 a, 1 b and the input end of theamplifier 2 in the first embodiment. -
FIG. 24 is a circuit diagram showing an equalizer according to this embodiment, which functions as thereception circuit 301 c. This equalizer also includes theadders resistors signal conversion units pass filter 4, and theamplifier 2, which are shown inFIG. 5 . - Further, the equalizer of
FIG. 24 includes capacitors CPa, CPb respectively on the transmission path for the positive logic signal and the transmission path for the negative logic signal, the respective capacitors having the one electrodes for receiving the reception signals S1, S2, and the other electrodes for receiving the signals S10, S11 to be inputted into theamplifier 2. - With such capacitors CPa, CPb provided, the capacitors CPa, CPb function as bypasses for transmitting a high frequency component of a reception signal to the
amplifier 2, thereby allowing improvement in high frequency characteristic of the equalizer. - Further, this embodiment may be applied to the equalizers according to the second to fifth embodiments as well as the equalizer according to the first embodiment.
- In the first to sixth embodiments, the description has been given based upon the equalizer of
FIG. 4 . However, it is possible to configure the equalizer ofFIG. 2 in the same manner as in the cases of the first to sixth embodiments. - Namely, in the case of applying the first embodiment to the equalizer of
FIG. 2 , other adders (these adders are not shown, and correspond to the addition unit 6) are provided respectively on the post-stage of theadders resistors signal conversion unit 60 a are given to those other adders as well as to theadders subtraction unit 5. Subsequently, in the other adders as theaddition unit 6, the output signals from theadders subtraction unit 5 are added to the output signal S6, S7 from the voltage-currentsignal conversion unit 60 a, and the addition results are inputted into theamplifier 2. - Further, in the case of applying the second embodiment to the equalizer of
FIG. 2 , the low-pass filter 4 may be deleted from the foregoing configuration obtained by applying the first embodiment to the equalizer ofFIG. 2 . As in the case of the second embodiment, the gate size, an injection amount of the impurity in the channel region, and the like, in terms of the transistors in the voltage-currentsignal conversion unit 60 b, are designed such that the input/output gain of the voltage-currentsignal conversion unit 60 b is constant in a region not higher than a predetermined frequency in a frequency band of a signal component included in the received voltage signals S1 (S4), S2 (S5) and gradually decreases in a region exceeding the predetermined frequency. - Further, in the case of applying the third embodiment to the equalizer of
FIG. 2 , in the foregoing configuration where the first or second embodiment is applied to the equalizer ofFIG. 2 , the voltage-currentsignal conversion units signal conversion units - Further, in the case of applying the fourth embodiment to the equalizer of
FIG. 2 , the bit error rate tester or theeye pattern detector 70 described in the fourth embodiment may be added to the foregoing configuration where the first or second embodiment is applied to the equalizer ofFIG. 2 . - Further, in the case of applying the fifth embodiment to the equalizer of
FIG. 2 , a semiconductor device having the flip chip package described in the fifth embodiment may be adopted in the foregoing configuration where the first or second embodiment is applied to the equalizer ofFIG. 2 . - Further, in the case of applying the sixth embodiment to the equalizer of
FIG. 2 , the capacitors CPa, CPb described in the sixth embodiment may be added to the foregoing configuration where the first or second embodiment is applied to the equalizer ofFIG. 2 . - With the equalizer of
FIG. 2 configured in the above-described manners, it is possible to obtain the same effect as the effects of the respective equalizers in the first to sixth embodiments. - While the invention has been shown and described in detail, the foregoing description is in all aspects illustrative and not restrictive. It is therefore understood that numerous modifications and variations can be devised without departing from the scope of the invention.
Claims (7)
1. An equalizer comprising:
a low-pass filter for receiving a reception signal;
a subtraction unit for subtracting from said reception signal an output signal from said low-pass filter;
an addition unit for adding said reception signal to an output signal from said subtraction unit; and
an amplifier for amplifying an output signal from said addition unit,
wherein said low-pass filter includes,
a receiving node receiving said reception signal,
an output node outputting said output signal,
a first transistor coupled between said receiving node and said output node, and
a second transistor having a gate connected to said output node, a source being supplied a ground potential, and a drain being supplied said ground potential.
2. The equalizer according to claim 1 , further comprising:
a first signal conversion unit for converting the output signal from said low-pass filter into a signal in direct proportion to the output signal from said low-pass filter to output the converted signal to said subtraction unit; and
a second signal conversion unit for converting said reception signal into a signal in direct proportion to the reception signal to output the converted signal to said subtraction unit and said addition unit, wherein
an input/output gain of each of said first and second signal conversion units is constant in a frequency band of a signal component contained in said reception signal.
3. The equalizer according to claim 1 , further comprising:
a resistor, wherein
one end of said resistor is connected to an input end of said amplifier, and
the other end of said resistor is supplied with a power supply potential.
4. The equalizer according to claim 3 , wherein
a resistance value of said resistor is variable.
5. The equalizer according to claim 1 , wherein
a cutoff frequency of said low-pass filter is variable.
6. The equalizer according to claim 1 , further comprising:
a capacitor having one electrode for receiving said reception signal and the other electrode for receiving a signal inputted into said amplifier.
7. A semiconductor circuit device comprising:
a semiconductor substrate on which the equalizer according to claim 1 is formed; and
a flip chip package containing said semiconductor substrate, wherein
said equalizer does not include an inductor element.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US12/730,061 US20100177814A1 (en) | 2004-11-29 | 2010-03-23 | Equalizer and semiconductor device |
Applications Claiming Priority (6)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2004-344000 | 2004-11-29 | ||
JP2004344000 | 2004-11-29 | ||
JP2005274904A JP2006180459A (en) | 2004-11-29 | 2005-09-22 | Equalizer and semiconductor apparatus |
JP2005-274904 | 2005-09-22 | ||
US11/282,647 US7764731B2 (en) | 2004-11-29 | 2005-11-21 | Equalizer and semiconductor device |
US12/730,061 US20100177814A1 (en) | 2004-11-29 | 2010-03-23 | Equalizer and semiconductor device |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US11/282,647 Continuation US7764731B2 (en) | 2004-11-29 | 2005-11-21 | Equalizer and semiconductor device |
Publications (1)
Publication Number | Publication Date |
---|---|
US20100177814A1 true US20100177814A1 (en) | 2010-07-15 |
Family
ID=36567356
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US11/282,647 Expired - Fee Related US7764731B2 (en) | 2004-11-29 | 2005-11-21 | Equalizer and semiconductor device |
US12/730,061 Abandoned US20100177814A1 (en) | 2004-11-29 | 2010-03-23 | Equalizer and semiconductor device |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US11/282,647 Expired - Fee Related US7764731B2 (en) | 2004-11-29 | 2005-11-21 | Equalizer and semiconductor device |
Country Status (2)
Country | Link |
---|---|
US (2) | US7764731B2 (en) |
JP (1) | JP2006180459A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100159899A1 (en) * | 2008-12-23 | 2010-06-24 | Qualcomm Incorporated | In-band provisioning for a closed subscriber group |
US9425997B2 (en) * | 2006-12-05 | 2016-08-23 | Rambus Inc. | Methods and circuits for asymmetric distribution of channel equalization between devices |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070223571A1 (en) * | 2006-03-27 | 2007-09-27 | Viss Marlin E | Decision-feedback equalizer simulator |
JP4936128B2 (en) * | 2007-06-07 | 2012-05-23 | 横河電機株式会社 | Loss compensation circuit |
US8018992B2 (en) * | 2007-10-31 | 2011-09-13 | Intel Corporation | Performing adaptive external equalization |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5486787A (en) * | 1993-01-08 | 1996-01-23 | Sony Corporation | Monolithic microwave integrated circuit apparatus |
US5714918A (en) * | 1995-08-12 | 1998-02-03 | Deutsche Itt Industries, Gmbh | Equalizer for digitized signals |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH06105184A (en) * | 1992-09-17 | 1994-04-15 | Olympus Optical Co Ltd | Picture signal processing unit |
JP2000057687A (en) * | 1998-08-07 | 2000-02-25 | Fujitsu Ltd | Optimum parameters detection for storage device |
JP2003168944A (en) | 2001-11-29 | 2003-06-13 | Nec Corp | Equalizer circuit |
JP2003204291A (en) * | 2002-01-07 | 2003-07-18 | Nec Corp | Communication system |
JP2004120468A (en) * | 2002-09-27 | 2004-04-15 | Kawasaki Microelectronics Kk | Input equalizer |
JP4364598B2 (en) * | 2003-10-22 | 2009-11-18 | 株式会社神戸製鋼所 | Filter processing apparatus, filter processing method and program thereof |
-
2005
- 2005-09-22 JP JP2005274904A patent/JP2006180459A/en active Pending
- 2005-11-21 US US11/282,647 patent/US7764731B2/en not_active Expired - Fee Related
-
2010
- 2010-03-23 US US12/730,061 patent/US20100177814A1/en not_active Abandoned
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5486787A (en) * | 1993-01-08 | 1996-01-23 | Sony Corporation | Monolithic microwave integrated circuit apparatus |
US5714918A (en) * | 1995-08-12 | 1998-02-03 | Deutsche Itt Industries, Gmbh | Equalizer for digitized signals |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9425997B2 (en) * | 2006-12-05 | 2016-08-23 | Rambus Inc. | Methods and circuits for asymmetric distribution of channel equalization between devices |
US9900189B2 (en) | 2006-12-05 | 2018-02-20 | Rambus Inc. | Methods and circuits for asymmetric distribution of channel equalization between devices |
US10135647B2 (en) | 2006-12-05 | 2018-11-20 | Rambus Inc. | Methods and circuits for asymmetric distribution of channel equalization between devices |
US10686632B2 (en) | 2006-12-05 | 2020-06-16 | Rambus Inc. | Methods and circuits for asymmetric distribution of channel equalization between devices |
US11115247B2 (en) | 2006-12-05 | 2021-09-07 | Rambus Inc. | Methods and circuits for asymmetric distribution of channel equalization between devices |
US11539556B2 (en) | 2006-12-05 | 2022-12-27 | Rambus Inc. | Methods and circuits for asymmetric distribution of channel equalization between devices |
US20100159899A1 (en) * | 2008-12-23 | 2010-06-24 | Qualcomm Incorporated | In-band provisioning for a closed subscriber group |
Also Published As
Publication number | Publication date |
---|---|
US7764731B2 (en) | 2010-07-27 |
US20060114980A1 (en) | 2006-06-01 |
JP2006180459A (en) | 2006-07-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US8867668B2 (en) | Coupling system for data receivers | |
US8324982B2 (en) | Integrated front-end passive equalizer and method thereof | |
US20100177814A1 (en) | Equalizer and semiconductor device | |
US8605397B2 (en) | Configurable, power supply voltage referenced single-ended signaling with ESD protection | |
US10840861B1 (en) | Receiver intermediate variable gain stage for isolator products | |
US11233482B2 (en) | Receiver front end for digital isolators | |
US8346096B2 (en) | Amplifier, optical receiver circuit, optical module and data exchange system | |
JP4676646B2 (en) | Impedance adjustment circuit and semiconductor device | |
US11005688B2 (en) | Common-mode control for AC-coupled receivers | |
US6696890B2 (en) | Differential signal transfer circuit | |
US8884691B2 (en) | Demodulators | |
JP2006254303A (en) | Signal transmission circuit, ic package, mounting substrate and ic chip | |
JP4237402B2 (en) | Output buffer for symmetric transmission line drive | |
US11522505B2 (en) | Semiconductor integrated circuit and receiver device | |
Gerfers et al. | A 0.2–2 Gb/s 6x OSR receiver using a digitally self-adaptive equalizer | |
JP2004096351A (en) | Terminating circuit for differential signal transmission line | |
Delshadpour et al. | A BMC Analog and Digital PHY for USB Type-C Power Delivery | |
US6922086B2 (en) | Method and apparatus for generating an input switching reference | |
US7199657B2 (en) | Amplification gain stages having replica stages for DC bias control | |
TWI599163B (en) | A semiconductor chip | |
JP2008124660A (en) | Filter circuit and semiconductor integrated device | |
JP2004289384A (en) | Amplifier circuit and semiconductor integrated circuit employing the same |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: RENESAS ELECTRONICS CORPORATION, JAPAN Free format text: CHANGE OF NAME;ASSIGNOR:NEC ELECTRONICS CORPORATION;REEL/FRAME:024915/0526 Effective date: 20100401 Owner name: NEC ELECTRONICS CORPORATION, JAPAN Free format text: MERGER;ASSIGNOR:RENESAS TECHNOLOGY CORP.;REEL/FRAME:024915/0556 Effective date: 20100401 |
|
STCB | Information on status: application discontinuation |
Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION |