KR20140077793A - Nauta operational transconductance amplifier for uwb filter in uwb based indoor positioning system - Google Patents

Abstract

Disclosed is an operational transconductance amplifier (OTA) for an ultra-wideband (UWB) filter which is a core block enabling processing of UWB signals among RF blocks constituting a coherent UWB receiver corresponding with IEEE 802.15.4a UWB standards. A Nauta-operational transconductance amplifier for the UWB filter in a UWB based indoor positioning system, according to the present invention comprises: two first g_m cells which receive an input positive voltage and an input negative voltage, respectively; and two second g_m cells, each of which has an input terminal connected to an output terminal of any one among the first g_m cells and an output terminal connected to an output terminal of the other among the first g_m cells. The Nauta-operational transconductance amplifier for the UWB filter outputs a first output current and a second output current having a current difference proportional to a voltage difference between the input positive voltage and the input negative voltage.

Description

[0001] NUTA OPERATIONAL TRANSCONDUCTANCE AMPLIFIER FOR UWB FILTER IN UWB BASED INDOOR POSITIONING SYSTEM [0002] FIELD OF THE INVENTION [0003] The present invention relates to a Nauta operational transconductance amplifier for an UWB-

The present invention relates to a Q-factor control method of a Nauta operational transconductance amplifier for an UWB-based indoor location recognition system, and more particularly, to a Nauta operational mutual conductance amplifier for implementing a high frequency Gm-C filter integrated circuit (IC) Gt; amplifier. ≪ / RTI >

The location-based service (LBS), which has been the mainstay of telematics navigators, has been attracting attention as a killer application in the mobile internet business due to the smartphone trend that has been waged since 2009. This is because the location information has its own independent value and also has the attribute of creating new value added when combining with existing information / contents, leading to creation of new industry. These trends have been centered on the electronic space where human beings and computers are connected to the information communication networks such as the Internet, but the basic axis of future informationization is the ubiquitous space center where human, computer and objects are connected to wireless information communication network It is consistent with the conventional view that it will move. In ubiquitous society, it is necessary to construct a communication network in which people and surrounding objects exchange mutual information at anytime and anywhere.

Along with this trend, LBS is creating a new market by forming a new service through merging with Augmented Reality (AR) and SNS which are showing rapid growth. LBS has expanded from conventional vehicle location and travel route services, and now a variety of location-based services for pedestrians are emerging in the marketplace, providing information, advertising, and games in the fields of logistics, finance, media, medical, education, entertainment, , SNS, and other services are provided in connection with location information, and their application fields are continuously diversified.

Various technology alternatives are being actively researched in order to secure the position recognition technology in the indoor space. However, each technology alternative has unique strengths and weaknesses, and a clear solution has not yet emerged. In particular, UWB technology can realize precise position recognition / tracking within a few tens of centimeters in an indoor or a shaded area. Therefore, when such superior time precision is used, it is applied to a high-precision position recognition field in an indoor space having multipath fading Long-term studies on the possibility are needed.

In the UWB system, a transmitter transmits a pulse of several nanoseconds (nsec) corresponding to a bandwidth of 500 MHz at intervals of several hundreds of nanoseconds, and a receiver measures a communication distance by estimating a first arrival time of a signal received through a multipath channel.

The characteristic of UWB technology is that it does not use carrier wave and it uses low transmit power of noise level in communication system with ultra wideband frequency bandwidth, so it does not cause interference to existing communication system. The advantage of the UWB system over the conventional narrowband system is that the distance estimation accuracy is excellent because the UWB signal has a width of several nanoseconds in the time domain and therefore has excellent time resolution at the receiver. Due to these advantages, UWB is attracting attention as a high-precision position recognition system and adopted UWB technology as a standard in IEEE 802.15.4a. Although the IEEE 802.15.4a UWB modulation scheme is divided into a synchronous scheme and an asynchronous scheme, the synchronous UWB receiver has a disadvantage in that the structure of the receiver is complicated, but the distance estimation accuracy is excellent in comparison with the asynchronous scheme.

Generally, a low-noise amplification (low-noise amplification) and a downconverting process are performed in a RF transceiver in a wireless communication transceiver, and an analog signal generated by converting a high frequency signal into a low frequency signal is supplied to an undesired frequency The block is a filter that removes the frequency band and transfers only the clean signal to the digital signal processing block. In the case of the Gm-C filter, the basic block of the circuit configuration is the operational mutual conductance amplifier (OTA). The operational transconductance amplifier is a functional block that receives a voltage signal as an input and outputs a current signal as an output. When the input signal is Vin and the output signal is Iout, the following relationship holds between the input / output signals.

[Equation 1]

Where g _m is usually a conductance gain or a proportional coefficient, simply referred to as conductance. LC filters typically require large inductors and capacitors. However, it is difficult to implement an inductor on an integrated circuit in view of the required area and Q-factor characteristics. Therefore, an active filter is used as an alternative. In an active filter, a transconductance element is used together with a capacitor to form a gyrator, thereby realizing the impedance of the inductor. Using a gyrator consisting of two operational mutual conductance amplifiers, you can implement a function block called an impedance inverter that makes the capacitor appear as an inductor. For high frequency applications in the VHF band, the Gm-C technique is preferred. In particular, in order to implement a Gm-C filter operating at a high frequency as in the UWB system, it is desirable to design a circuit to prevent the occurrence of parasitic pole points inside the transconductor, which is the obstacle to high frequency operation. There are two root causes that cause the parasitic pole. One is the use of a current mirror for the phase reversal of the signal (which is basically a function of the active filter implementation) and the other is the use of the Cascoding technique to increase the output resistance of the transconductor. The use of a current mirror can be avoided by employing a fully differential transconductor (balanced design). It is possible to avoid using the Cascoding technique by employing a technique that reduces the common mode output resistance of the transconductor and maximizes only the differential mode output resistance without introducing additional nodes. A representative example of this approach is the Operational Transconductance Amplifier ("Nauta-OTA") proposed by Nauta. In general, Nauta-OTA uses the single-input single-output inverter as a basic unit block and constructs it as shown in Fig. 1 is a circuit diagram of a conventional Nauta-OTA. In the operation characteristics, only the differential mode is effectively amplified while suppressing the common mode among the signal components. Since the inverter is used as a basic unit block, there is no other node inside the power supply and ground node, which is a very advantageous structure for high-frequency operation. And includes only two transistors in series between the power supply and the ground node, making it suitable for low-voltage operation. 2 is a circuit diagram of a conventional Nauta-Gyrator. However, in the case of the conventional gyrotron structure using the Nauta-OTA, the common mode component of the input signal has a constant gain and appears on the output side, and it is not easy to raise the quality factor thereof. Therefore, there is a need for a new OTA circuit structure that can effectively improve the quality factor by independently controlling the quality factor without affecting the transconductance value required for the frequency characteristic of the filter in the conventional Nauta-OTA circuit structure It is urgent.

The object of the present invention is to effectively improve the quality factor by independently controlling the quality factor without affecting the transconductance value required for the frequency characteristic of the filter in the conventional Nauta-OTA circuit structure using the Nauta-Transconductor To provide a new OTA circuit structure.

In order to achieve the above object, a Nauta operational transconductance amplifier for UWB-based indoor location recognition system for UWB filters comprises two first g _m cells receiving input constant voltage and input voltage, respectively; And each comprise the first cells 1 g _m any one of the two cells to which 2 g _m input terminal is connected and an output terminal connected to the other output terminal of said first _m 1 g of the cells to the output terminal.

At this time, the Nauta operational transconductance amplifier for ultra-wideband filter outputs a first output current and a second output current having a current difference proportional to the voltage difference between the input constant voltage and the input voltage.

According to the present invention, in the conventional Nauta-OTA circuit structure, a means for effectively improving the quality factor by independently controlling the quality factor without affecting the transconductance value required for the frequency characteristic of the filter is provided .

Further, the present invention can be implemented by using a smaller number of basic unit block cells in the conventional Nauta-OTA circuit structure.

1 is a circuit diagram of a conventional Nauta-OTA.
2 is a circuit diagram of an improved OTA (Modified-Nauta OTA) according to the present invention.
3 is a diagram showing the vector rotation and the coordinate system rotation on a plane.
4 is a diagram showing a new coordinate axis (Principal Axis) in which a matrix A is described as a diagonal matrix and unit vectors on each coordinate axis.
5 is a structure diagram of a transceiver of an IEEE 802.15.4a UWB system.

The present invention will now be described in detail with reference to the accompanying drawings. Hereinafter, a repeated description, a known function that may obscure the gist of the present invention, and a detailed description of the configuration will be omitted. Embodiments of the present invention are provided to more fully describe the present invention to those skilled in the art. Accordingly, the shapes and sizes of the elements in the drawings and the like can be exaggerated for clarity.

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings.

First, an analysis of the conventional Nauta-OTA will be described with reference to FIG.

&Quot; (2) "

&Quot; (3) "

As usual, if (g = g _m1 = g _{m2a m2b)} = g _m, and (g _o1 = g = g _{o2a o2b)} = g _o Assuming,

&Quot; (4) "

&Quot; (5) "

From here,

&Quot; (6) "

Because of

&Quot; (7) "

&Quot; (8) "

In order to summarize this,

&Quot; (9) "

If you multiply y by both sides

&Quot; (10) "

. This is the basic input / output equation of the conventional Nauta-OTA.

Using a capacitive load as the load admittance and performing the Laplace transform of Equation (10), the following Equation (11) can be obtained.

&Quot; (11) "

By dividing both sides by C,

&Quot; (12) "

&Quot; (13) "

Considering the case where homogeneous (input V _i1 = V _i2 = 0), this equation becomes as shown in the following equations (14) and (15).

&Quot; (14) "

&Quot; (15) "

That is, in quantum mechanics,

It can be seen that the problem is caused by the eigenvalue / eigenvector problem. Let us first find the eigenvalues / eigenvectors of matrix A to find the transformation matrix for diagonalizing matrix A. The eigenvalue can be obtained by setting the determinant of the left side coefficient matrix of Equation (14) to zero. However, when we look at the form of the matrix A, A ⁺ = A, it is Hermitian (or Self-Adjoint). Therefore, from the general nature of Hermitian, the eigenvalues are expected to have real values and the eigenvectors to be orthogonal. If the eigenvalue is actually obtained, it can be seen that it is a real number as shown in the following equation (16).

&Quot; (16) "

Since g _o = 1 / r _out , s ₁ and s ₂ have the dimension of the reciprocal of the time constant in the RC-circuit.

The eigenvectors corresponding to the respective vectors can be obtained by substituting s ₁ , s ₂ in Equation (14). Normalizing the size to 1 results in the following equations (17) and (18).

&Quot; (17) "

&Quot; (18) "

Where | r _i > is a column vector, and <r _j | is a row vector. For these two unit vectors

As shown above, it can be confirmed that it is orthogonal.

Now, a transformation matrix for diagonalizing the matrix A can be constructed as shown in the following equation (19).

&Quot; (19) &quot;

When the matrix A is diagonalized by using this transformation matrix, a diagonal matrix

.

&Quot; (20) &quot;

Here, R ⁺ is a transpose matrix of < R's Complex-Conjugate matrix &gt;. If the equation (13) is written in a matrix form, the following equation (21) is obtained

&Quot; (21) &quot;

When the linear transformation of the following expression (22) is performed in the expression (21), the following expression (23) is obtained.

&Quot; (22) &quot;

&Quot; (23) &quot;

By multiplying the left sides of the above Equation 23 by R ^-1 = R ⁺ , the following equations (24), (25) and (26) are obtained.

&Quot; (24) &quot;

&Quot; (25) &quot;

&Quot; (26) &quot;

The following equation (27) is obtained by expressing the component elements of the matrix.

&Quot; (27) &quot;

It can be seen here that the form of the transformation matrix R exactly coincides with the matrix describing the rotational movement of the coordinate system on the plane (rotation angle = 45 degrees). Hereinafter, the relevance of the transformation matrix R to the rotational motion transformation will be analyzed in more detail. To solve this problem, the problem of rotation of a coordinate system on a general plane will be described with reference to FIG.

3 is a diagram showing the vector rotation and the coordinate system rotation on a plane.

Matrix A is a vector

A new vector

As shown in FIG.

&Quot; (28) &quot;

Now we can display the given vector

from

Rotation of the coordinate axis to convert

Is applied.

&Quot; (29) &quot;

Left

In the new coordinate system

And the right side

In the new coordinate system

, It can be seen that the linear transformation expressed in the spherical coordinate system by the matrix A is represented by the matrix A '= (BAB ^-1 ) in the new coordinate system. That is,

The role of A in vector space

Means that A '= (BAB ^-1 ) is made (this transformation is called Similarity Transformation). Then, when B is rotated counterclockwise

The matrix B is described by the following equation (30). &Quot; (30) &quot;

&Quot; (30) &quot;

Here, the element of the matrix B

Is the direction cosine.

&Quot; (31) &quot;

Now we will look at the relevance of the transformation matrix R to the coordinate system rotation motion transformation.

vector

A reference coordinate system

As shown in Fig.

(32)

Now,

Coordinate axes

Expression in

silver

.

&Quot; (33) &quot;

Consider the form of Equation (25) in relation to the form of Equation (29). In Equation 33,

Can be understood as an expression in the new coordinate system after the coordinate axis rotation movement of any fixed vector. Then, we take a viewpoint that accepts matrix A as an Operator. In this case, the vector in the spherical coordinate system

In the new coordinate system,

The matrix A '= (BAB ^-1 ) is performed. Accordingly, the equation (25) is the original equation (21)

As a reference. Let's look at what kind of matrix B the coordinate axis describes as the rotation movement.

From the above equation (22)

When this equation is compared with the equation (33), it can be seen that the matrix B describing the rotational movement of the coordinate axes is B = R ^-1 . Thus, the matrix R ⁺ = R ^T = R ^{- 1} can be seen as a matrix B describing the rotation of the coordinate axes on the plane (then the relation R = B ^-1 holds). Namely, as shown in FIG. 4, when a matrix B describing the rotational movement of the coordinate axes and specifically selecting the inverse matrix R ^-1 of the matrix R consisting of the eigenvectors, the expression A '= (BAB -1) is expressed as a diagonalized form as shown in Equation (27). 4 is a diagram showing a new coordinate axis (Principal Axis) in which a matrix A is described as a diagonal matrix and unit vectors on each coordinate axis.

&Quot; (34) &quot;

Equation (34) consists of two row vectors <r ₁ | and <r ₂ |, which are the two direction cosines associated with the two coordinate axes of x 'and y' relative to the original coordinate system do. When Equation (34) is set as shown in Equation (30), the angle of the coordinate axis rotation movement is

= -45 degrees.

Note that the matrix R ⁺ = R ^T is a coordinate axis that is rotated by a new coordinate axis described by a diagonal matrix. This new coordinate axis is represented by two eigenvectors | r ₁ >, | r ₂ > . These two eigenvectors are the unit vectors on a new coordinate axis (referred to as "Principal Axis") where the matrix A is described as a diagonal matrix

,

).

Now we apply the following linear transformation using the transformation matrix R derived above for the input and output signal pairs.

(For simplicity, we use CM _k instead of DM _k and V _k2 ^' instead of V _k1 ^' .)

&Quot; (35) &quot;

From this, a relation of the following expression (36) is obtained.

&Quot; (36) &quot;

From this, the following equations (37) and (38) are obtained.

&Quot; (37) &quot;

&Quot; (38) &quot;

This is substituted into Equation (13) to obtain Equation (39).

[Equation 39]

The physical meaning of the above equations will be described as follows.

Equation (39) represents the forward transfer characteristic relationship in which DM _o and CM _o are detected as outputs in Port o when DM _i and CM _i are applied as inputs in Port i. The characteristic observed in Equation (39) is that as the form of the transfer function matrix is a diagonalized matrix, only the CM _i , which is the common mode among the input signals, and the CM _o , which is the common mode among the output signals, The fact that the differential mode of the signal, DM _i, and the output signal are mutually influenced only between the common mode CM _o , is that the decoupling between the common mode and the differential mode is completely achieved. This property is desirable in view of the fact that the signal to be processed is not included in the common mode at all. However, in the case of the conventional Nauta-OTA circuit structure, it is pointed out that it is not easy to increase the quality factor as shown in the above two equations. In fact, when high-frequency operation is required, as in the case of a UWB transceiver, when the operating frequency of the filter exceeds several hundreds of MHz, the g _m value is very large and the quality factor is considerably lowered. Nevertheless, there is no other realistic means of independently controlling the quality factor. In order to improve the operational characteristics, an improvement of the conventional Nauta-OTA circuit structure according to the present invention will be described below.

2 is a circuit diagram of an improved OTA (Modified-Nauta OTA) according to the present invention.

Hereinafter, two cases will be described separately.

_{(1) ((g m1 =} g m2b) = g m; (g o1 = g o2b) = g o) is obtained and to the case of g = g _{o2a m2a} = 0 (40) and 41.

[Equation 40]

(41)

Here, since the relationship of the following equation (42) is established, the following equations (43) and (44) are obtained.

(42)

Equation (43)

&Quot; (44) &quot;

If this is summarized and expressed as a determinant, the following equation (45) is obtained.

&Quot; (45) &quot;

By multiplying both sides of the equation (45) by y, the following equation (46) is obtained.

&Quot; (46) &quot;

Equation (46) is a basic input / output equation of Modified-Nauta OTA.

(47) is obtained by using a capacitive load (Capacitor) as the load admittance and performing Laplace transform of the equation (46).

&Quot; (47) &quot;

If both sides of the equation (47) are divided by C, the following equation (48) is obtained.

&Quot; (48) &quot;

Expression 48 can be rewritten as: &quot; (49) &quot;

&Quot; (49) &quot;

Here, considering the case of Homogeneous (input V _i1 = V _i2 = 0 in Equation 49), the following equations 50 and 51 are obtained.

(50)

&Quot; (51) &quot;

That is,

And the eigenvalue / eigenvector problem of the form. Let us first find the eigenvalues / eigenvectors of matrix A to find the transformation matrix for diagonalizing matrix A. The eigenvalue can be obtained by setting the determinant of the left side coefficient matrix of Equation (50) to zero. By the way, the form of the matrix A is A ⁺ = A, so it is Hermitian (or Self-Adjoint). Therefore, from the general nature of Hermitian, the eigenvalues are expected to have real values and the eigenvectors to be orthogonal. If the eigenvalue is actually obtained, it can be understood that it is a real number as shown in the following equation (52).

(52)

Since g _o = 1 / r _out , s ₁ and s ₂ have the dimension of the reciprocal of the time constant in the RC-circuit.

The eigenvectors corresponding to the respective vectors can be obtained by substituting s ₁ and s ₂ in (50). If the size is normalized to be 1, the following equations (53) and (54) are obtained, and the same results as in equations (17) and (18) are obtained. Thus, the transformation matrix R is also given in the same way as in equation (19), where | r _i > is a column vector and < r _j | For these two unit vectors

As shown above, it can be confirmed that it is orthogonal.

&Quot; (53) &quot;

(54)

Equations (35) to (38), which are linear conversion equations using the above-described conversion matrix R for the input signal pair and the output signal pair, are substituted into Equation (49) and expressed as Equation (55).

(55)

That is, it can be seen that g _m intervenes in the denominator of the right side of the equation, which means that it becomes possible to adjust the position of the pole by g _m . In other words, it means that the quality factor can be adjusted by g _m . One problem is that the cut-off frequency is adjusted by the same g _m . Therefore, independent quality-factor adjustment remains to be desired. A method to solve this is shown below.

(2) (

;

)ego

The following equations (56) and (57) are obtained.

&Quot; (56) &quot;

&Quot; (57) &quot;

Here, since the relationship of the following equation (58) is satisfied, the following equations (59) and (60) are obtained.

(58)

(59)

(60)

If this is summarized and expressed as a determinant, the following equation (61) is obtained.

&Quot; (61) &quot;

When both sides of the equation (61) are multiplied by y, the following equation (62) is obtained.

(62)

Equation 62 is the basic input / output equation of Modified-Nauta OTA.

Using the capacitive load as the load admittance and Laplace transform of the equation (62), the following equation (63) is obtained.

Equation (63)

By dividing both sides of the equation (63) by C, the following equation (64) is obtained.

Equation (64)

(64) is again obtained to obtain the following equation (65).

Equation (65)

Here, when Homogeneous (V _i1 = V _i2 = 0 in Equation 65) is considered, the following equations 66 and 67 are obtained.

[Equation 66]

Equation (67)

That is,

It can be seen that the problem is caused by the eigenvalue / eigenvector problem. Let us first find the eigenvalues / eigenvectors of matrix A to find the transformation matrix for diagonalizing matrix A. The eigenvalue can be obtained by setting the determinant of the left side coefficient matrix of Equation (66) to zero. However, when we look at the form of the matrix A, A ⁺ = A, it is Hermitian (or Self-Adjoint). Therefore, from the general nature of Hermitian, the eigenvalues are expected to have real values and the eigenvectors to be orthogonal. If the eigenvalue is actually obtained, it can be seen that it is a real number as shown in the following equation (68).

Equation (68)

Since g _o = 1 / r _out , s ₁ and s ₂ have the dimension of the reciprocal of the time constant in the RC-circuit.

The eigenvectors corresponding to each can be obtained by substituting s ₁ and s ₂ in (66). If the size is normalized to be 1, the following equations (69) and (70) are obtained.

[Equation 69]

[Equation 70]

This is the same result as in equations (17) and (18). Thus, the transformation matrix R is also given by Equation (19), where | r _j > is a column vector and < r _j | For these two unit vectors

As shown above, it can be confirmed that it is orthogonal.

Equations (35) to (38), which are linear conversion expressions using the above-described conversion matrix R for the input signal pair and the output signal pair, are substituted into Equation (65), and the following Equation (71) is obtained.

&Quot; (71) &quot;

This relationship, while maintaining the Quality-factor adjustment possibilities discussed in the case of (1), shows that g _m1 to the Quality-factor adjustment possibilities the cut-off frequency coordination was pointed out as a problem and can be realized as distinct g _m2.

By simplifying the existing Nauta transconductor by re-analyzing it from the new method and perspective, it is analyzed and simplified that there is no surplus element among the components, and a simpler and more efficient circuit structure is achieved through the method of sharing the roles of the remaining inverters. It is possible to achieve the same operation characteristics as those of the conventional Nauta-Transconductor by using the scheme of the present invention with a smaller number of inverters than those of the conventional Nauta-Transconductor, and furthermore, the transconductance value The filter characteristic can be effectively improved by newly providing a means for independently controlling only the quality factor.

5 is a structure diagram of a transceiver of an IEEE 802.15.4a UWB system.

The present invention relates to a method of designing an ultra-wideband filter, which is a key block enabling ultra-wideband signal processing among RF blocks constituting a synchronous UWB receiver conforming to the IEEE 802.15.4a UWB standard, And a method of implementing a gyrator used in a Gm-C filter implemented by an integrated circuit (IC) process using an operational transconductance amplifier (OTA) and a capacitor.

The transceiver structure of the IEEE 802.15.4a UWB system is shown in Fig. In the transmitter, a signal is transmitted through an RS encoder, a convolutional encoder, a symbol mapper, a preamble insertion, a pulse shaper, an RF antenna, and an antenna. In the receiver, a signal received through an RF stage is transmitted to a pulse shaper, a synchronizer, Decoder, and an RS decoder. The RF block includes a transmitter including a filter, an Up mixer, and a power amplifier, a receiver including a low noise amplifier (LNA), a down mixer, a filter, and a variable gain amplifier (VGA) And a frequency synthesizer that provides a local oscillator (LO) to the mixer.

As described above, the Nauta operational transconductance amplifier for a UWB based indoor location recognition system according to the present invention is not limited to the configuration and method of the embodiments described above, All or some of the embodiments may be selectively combined.

11, 12: First g _m cell of Nauta-OTA
13, 14: 2nd g _m cell of Nauta-OTA
15, 16: Third g _m cell of Nauta-OTA
21, 22: a first g _m cell of an improved OTA (Modified-Nauta OTA)
23, 24: a second g _m cell of an improved OTA (Modified-Nauta OTA)

Claims (1)

Two first g _m cells receiving input constant voltage and input voltage, respectively; And
Respectively, to the input terminal is connected to one of output terminals of the first cells 1 g _m and comprising two claim 2 g _m cell which is connected to the output terminal to the other of the output terminals of the first 1 g _m cells,
And a first output current and a second output current having a current difference proportional to a voltage difference between the input constant voltage and the input voltage are output. Turn amplifiers.

Images