KR100339662B1 - Apparatus for removing direct current deflection in bluetooth system - Google Patents

Abstract

The present invention relates to an apparatus for canceling a DC (DC) residual error in a Bluetooth unit receiving an access code, the apparatus comprising: a demodulation block for receiving and demodulating an access code in a received signal; An access code storage unit storing a plurality of types of access codes; A correlation detection block for detecting a correlation between an access code of a demodulation block and an access code stored in an access code storage unit and providing access code information exceeding a predetermined threshold value; A first means for calculating and storing an average value of an access code of the correlation detection block; Second means for calculating a normalization value by multiplying an average value of the access signal from the demodulation block by a predetermined normalization constant; A first subtracter for subtracting the average value of the first means from the normalized value of the second means to calculate a DC residual deviation value; And a second subtractor for subtracting the DC residual deviation value of the first subtracter from the received signal of the demodulation block to eliminate the DC residual deviation of the received signal.

That is, in the present invention, since the access code is used as a training signal for eliminating the DC residual deviation, it is possible to accurately remove the DC residual deviation in the Bluetooth system.

Description

[0001] APPARATUS FOR REMOVING DIRECT CURRENT DEFLECTION IN BLUETOOTH SYSTEM [0002]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a short range wireless Bluetooth system, and more particularly, to a DC residual deviation elimination apparatus for a short range wireless Bluetooth system for eliminating a DC residual deviation generated during transmission of information.

Bluetooth is a small, affordable, and low-power, wireless connection (2.4GHz ISM Open Band) within a small area (10m-100m) between mobile devices such as cell phones, laptops, network access points and other peripherals. Is a technical specification specification.

FIG. 1 shows a structure of a baseband equivalent model of a GFSK (Gaussian Frequency Shifting Keying) transmission apparatus used in a conventional Bluetooth system. As shown, the packet generator 2 in the Bluetooth transmitter generates the packet g (t) under the control of the Bluetooth link controller 1 and the packet g (t) is the Gaussian low- Lt; / RTI > The output signal b (t) of the Gaussian low-pass filter 3 is integrated in the integrator 4 and then modulated in the modulator 5.

In the Bluetooth system, a binary GFSK modulation scheme with a modulation index (h) of 0.28 to 0.35 is adopted to simplify the structure of the transceiver

In the transmitter shown in Fig. 1, the transmission signal p (t) is expressed by Equation (1).

Where is the energy per symbol, Is the frequency of the carrier signal, Is a phase function including information to be transmitted as defined in Equation (2).

Here, h denotes a modulation index, g (t) denotes an NRZ data signal, and (b (T)) denotes a Gaussian low- (3).

The NRZ data signal g (t) can be expressed by Equation (3).

Here, b (t) can be expressed by Equation (4).

In Equation (4), Bb represents the 3-dB bandwidth of the Gaussian low-pass filter 3, and the GFSK modulation method in the Bluetooth system satisfies B _b T = 0.5. In Equation (3), v (t) is expressed by Equation (5).

here Represents an error function.

On the other hand, when the Bluetooth system is operated in an office, a home, or an indoor space such as an airport, statistical modeling of a Bluetooth system is performed by modeling a channel of the transmitted signal into a multi- And arrive at a Poisson process with a certain ratio. Also in that group, another series of rays at a fixed rate are also modeled as arriving at the Poisson process. The complex channel response function of the baseband is modeled by Equation (6).

here, Is the arrival time of the k-th ray in the ℓ-th group, and β _kl and θ _kl denote the k-th gain and phase in the ℓ-th group, respectively. In the channel modeling, the probability density function for the arrival time of the group and the ray is the Poisson function, the probability density function for the path gain (β _kl ) is the Rayleigh function, and the phase (θ _kl ) As a distribution function.

In the measurement data, the group and the ray have valid values within 200 ns, and the rms delay spread value of the gain is assumed to be 70 ns. Since the Bluetooth system uses a GFSK modulation scheme having a data rate of 1 Mbit / s and a slot size of 625 microseconds, the channel modeling of the Bluetooth system is a channel having one tap in Equation (6) Frequency non-selective channels. 2 is an equivalent circuit of a channel made up of additive white Gaussian noise in the present invention.

The GFSK modulated signal s (t) transmitted through the channel shown in FIG. 2 can be expressed by Equation (7).

here Additive complex white Gaussian noise, M (t) is a complex envelope of the baseband equivalent signal of the GFSK modulated signal p (t), defined as Equation (8).

Thus, equation (8) is defined by equation (9).

Here, A (t) is expressed by Equation (10) and B (t) is represented by Equation (11).

The GFSK demodulation scheme can be implemented in synchronous and asynchronous manner, and uses an FM disk limiter in consideration of the characteristics of a low-power, small-size Bluetooth system.

3 shows a configuration of a GFSK demodulator using an FM disc limiter. In the received signal s (t) in FIG. 3, phase information is included, and the hard limiter 31 compensates the amplitude magnitude of the received signal s (t) by a constant value. The FM disk limiter 32 extracts the phase of the received signal s (t) using Equations (12) and (13) and extracts necessary information included in the phase.

However, as can be seen from Equation (13), it can be seen that the channel, the filter of the receiver, the noise, etc. distort the demodulated GFSK signal and may also appear in the form of DC residuals. The frequency shift of the local oscillator in the GFSK demodulator also appears as the DC residual deviation at the output of the demodulator. That is, the frequency shift (Δf _c ) of the local oscillator is represented by the DC residual deviation of 2πΔf _c in Equation (13). In a direct conversion receiver, the high-frequency signal reflected from the mixer or low-noise amplifier appears nonlinearly at the demodulator output and is present as a DC residual error. The DC residual error causes the analog-to-digital converter's operating range to reach saturation and, consequently, distort the demodulated signal and act as a major cause of degradation of the system's performance.

A simple DC blocking method or a high-pass filter method has been proposed as a method of removing the DC residual error. However, this method is not suitable for the GFSK modulation method because it distorts the received signal in the low-band. Meanwhile, in the conventional Bluetooth system, four preamble symbols and four trailer symbols constituting the access code are used in order to eliminate the DC residual deviation. As shown in FIG. 4, the access code includes a total of 72 symbols including four preamble symbols, 64 sync words, and 4 trailer symbols.

The 4-bit trailer symbol in the access code exists only when there is a packet header in the structure of the packet or when the packet is FHS. In the conventional Bluetooth system, the DC residual deviation is once estimated using the four preamble symbols, and then the accurate DC residual deviation is estimated using the three MBSs and the trailer of the synchronous word. However, the preamble and the trailer used for the DC residual deviation estimation in the Bluetooth system have insufficient data amount to calculate the accurate DC residual deviation. Moreover, in the Bluetooth system, the start point of the received signal is searched using an uncertainty window allowing an error of ± 10 μs. However, it is very difficult to accurately find the start of the receive slot during the preamble period including four symbols . Therefore, in order to eliminate an accurate DC residual deviation, a training signal having a relatively large amount of data and accurate reception synchronization is required rather than a preamble or a trailer of an access code. However, in the specification of a Bluetooth system, You can not do it, so you have to find a new way.

SUMMARY OF THE INVENTION It is an object of the present invention to provide an apparatus for removing a DC residual error by using a synchronous word of an access code used in a Bluetooth system as a training signal for DC residual error estimation.

In order to achieve the above object, the present invention provides an apparatus for removing DC residual error in a Bluetooth unit that receives an access code, the apparatus comprising: a demodulation block for receiving and demodulating the access code in a received signal; An access code storage unit storing a plurality of types of access codes; A correlation detecting block for detecting a correlation between an access code of the demodulation block and an access code stored in the access code storage and providing access code information exceeding a predetermined threshold; A first means for calculating and storing an average value of an access code of the correlation detection block; Second means for calculating a normalization value by multiplying an average value of the access signal from the demodulation block by a predetermined normalization constant; A first subtracter for calculating a DC residual deviation value by subtracting an average value of the first means from the normalized value of the second means; And a second subtracter for subtracting a DC residual deviation value of the first subtracter from a reception signal of the demodulation block to eliminate a DC residual deviation of the reception signal.

According to another aspect of the present invention, there is provided an apparatus for canceling DC residual error of a Bluetooth unit that receives an access code, the apparatus comprising: a demodulation block for receiving and demodulating the access code in a received signal; An analog-to-digital converter for digitizing an output signal of the demodulation block; An access code storage unit storing a plurality of types of access codes; A correlation detecting block for detecting a correlation between an access code of the demodulation block and an access code stored in the access code storage and providing access code information exceeding a predetermined threshold; A first means for obtaining a difference value between an access code of the analog-digital converter and an access code stored in the access code storage unit; Second means for obtaining an average value of the first means as a DC residual deviation value; And a first subtractor for subtracting the DC residual deviation value of the second means from the output of the analog-digital converter.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram showing the structure of a general GFSK transmitter,

2 is a modeling of a channel of a wireless Bluetooth system,

3 shows a structure of a general GFSK demodulator,

4 is a diagram showing a structure of an access code used in a Bluetooth system,

FIG. 5 is a block diagram of an apparatus for removing DC residual error in a wireless Bluetooth system according to the present invention.

6 shows an indeterminate area in a receive mode slot in a Bluetooth system,

7 is a diagram showing an output value of a correlation detection block when correct reception time synchronization is performed with a reception access code in a Bluetooth system,

FIG. 8 is a diagram illustrating a result of estimation of a residual deviation by a DC residual deviation eliminating device of a wireless Bluetooth system according to the present invention,

9 is a graph comparing BER performances of a DC residual deviation eliminating device and a conventional DC residual deviation eliminating device of a wireless Bluetooth system according to the present invention.

10 is a view showing another embodiment of an apparatus for removing direct current deviation of a wireless Bluetooth system according to the present invention.

Description of the Related Art

101: Demodulation block 102: Correlation detection block

103: Access code storage unit

104: Gaussian low-pass filter look-up table

105: Totalizer 106:

107: latch 108: integrator

109: latch 110: divider

111: subtracter 112: analog-to-digital converter

Hereinafter, the present invention will be described in detail.

First, the present invention uses an access code used in a Bluetooth system as a training signal for eliminating DC residual deviation. Currently, the synchronous word of the access code in the Bluetooth standard is used to synchronize the reception time with the confirmation of the packet transmitted in the Bluetooth small network. The master device in the Bluetooth network can communicate with a plurality of slave devices. That is, the master device knows the addresses of all the Bluetooth devices (slaves) connected on the network, and the slaves obtain the address of the master device in the process of forming the network.

On the other hand, an access code is used for communication between the master device and the slave devices. As shown in FIG. 4, a total of 72 (N _f ) bits are used for the access code.

The access code indicates the network formed between the master device and the slave devices. In addition, the access code is used to confirm the packet and synchronize the time synchronization between the master device and the slave device. The access code includes a channel access code, a device access code, a general access code, and the like. The CAC includes a low address part (LAP) And 4 bits of an 8-bit Upper Address Part (UAP). Also, GIAC and DAC are used before the network performs connection mode. GIAC is assigned a pre-set code when initial power of network is turned on, and DIAC assigns codes using each unique value of each device. Therefore, the master device and the slave devices in the Bluetooth network can know in advance information on the access codes of all packets transmitted in the network. Therefore, by using the property that the master device and the slave devices know the access code used in the network, the synchronization word of the access code can be used as a sort of training signal for removing the DC residual deviation. As described later, the method using the access code can remove the DC residual deviation using the data that has been correctly received. That is, in the present invention, a correlation value of an access code is detected using a prior access code correlator used in a Bluetooth system, and a synchronous word synchronized from a correlation value exceeding a threshold value is set as a training signal for DC residual deviation elimination use.

Figure 5 shows a block diagram of an apparatus according to the invention. As shown in the figure, the demodulation block 101 demodulates the received signal s (t) by GMSK demodulation and outputs the demodulated received signal s (t) to the correlation detection block 102. The correlation detection block 102 receives the access code stored in the access code storage 103, (s (t)) and detects the start point of the received signal. That is, in order to use all the 72-bit access codes in the Bluetooth system, it is necessary to know the starting point (time) of the correct received signal, but it is difficult to accurately detect the starting point and use all the 72-bit access codes. In the Bluetooth system, as shown in FIG. 6, a start point of a received signal is detected using an uncertainty window having a time synchronization error of ± 10 μs, and a correlation method The detection block 102 is used.

The correlation detecting block 102 compares the correct reception time of the received signal s (t) and the access code in the received signal s (t) with Np (t) to detect whether the access code corresponds to CAC, GIAC or DAC Use a pre-access code. That is, the correlation detecting block 424 detects the correlation between the access code (corresponding to each of CAC, GIAC, and DAC) of a predetermined number of bits corresponding to the uncertainty region and the access code A correlation value is detected and monitored for the access code (for the bit corresponding to the uncertainty region). If an access code exceeding the threshold value is detected, the detected access code is synchronized with the detected access code, and the detected access code is stored in the Gaussian low-pass filter look-up table 104 ). 7 shows the correlation value of the GUAC access code using " 0x92833 " and the advance access code when the correct reception time is synchronized. In the example of FIG. 7, Np is 10 bits.

In the present invention, the access code detected in the above correlation detection block 102 is used as a training signal by DC residual deviation elimination. On the other hand, when the access codes (CAC, DAC, GIAC) detected by the correlation detecting block 102 are a'n, the number of the access codes a'n is defined as shown in Equation (14).

In Equation (14), N _f represents the length of the access code. The signal m (t) can be obtained by Equation (15) when the signal after the access code (a ' _n ) passes through the Gaussian low-pass filter 104 in Equations (3) and (5) is m

In the present invention, the access code passes through the Gaussian low-pass filter 104 and uses a value obtained by oversampling? Times the symbol rate as in Equation (16).

The value m (n) of the Gaussian low-pass filter look-up table 104 is provided to an accmulator 105 and the integrator 105 compares the Gaussian low-pass filter look-up table 104, Is accumulated.

As described above, in the present invention, only the data after the trigger signal is generated in the advance access code in the uncertainty region is used in order to use only the correctly received synchronized signal. Thus, the accumulator 105 in the N _f of access code symbol uses the signal and synchronization is reliably N _p +1 th from N _f of the second access code consisting of symbols received from. The value (S _m ) integrated in the accumulator 105 is provided to the divider 106 and an average value is calculated as shown in Equation (18).

The average value M _s passed through the divider 106 is stored in the latch 107, which is a memory element, for use during one slot excluding the access code. This average value M _s is an average value of the access code in which the DC residual deviation does not exist since it is a value assuming an ideal access code and a demodulator.

On the other hand, a signal demodulated through the demodulation block 101 has a DC residual error due to a circuit structure, a frequency deviation of a characteristic local oscillator of the device, and the like. As described above, the integral value (S _r ) is calculated from the signal (? (T)) in which the DC residual deviation exists through the integrator 108 as shown in Equation (19).

=

The integrator 108 also uses the (N _p +1) th through N _fth access code symbols whose synchronization is assured, as in the accumulator 105 of Equation (18). The signal passed through the integrator 108 is stored in the latch 109, which is a memory element, after a time of TN _f after receiving the access code.

The signal stored in the latch 109 is divided by the divider 110 as shown in Equation (20) to obtain an average of the signals passed through the integrator 108.

Here, the constant (rho) is a normalization constant multiplied for comparison with the average value of the equation (18).

The subtracter 111 obtains the difference between the average values of the divider 106 and 110 obtained from the equations (18) and (20), and this value? _DC is an estimate of the DC residual deviation as shown in Equation (22).

The DC residual deviation estimate? _DC obtained by the subtractor 111 is provided to a subtractor 112 and the subtractor 112 subtracts the DC residual deviation estimate? (T) from the signal? (T) in the demodulation block 101 _DC ), the signal provided to the downstream analog-to-digital converter 112 is in a state in which the DC residual deviation is removed.

The DC residual deviation estimate? _DC obtained above is used to remove the DC residual deviation while receiving the remaining packet symbols, i.e., the packet header and the payload, after receiving the access code. In the above example, it is assumed that the DC residual error removing apparatus of the present invention operates every slot. However, when the rate of change of the actual DC residual error with respect to time is small, the DC residual deviation can be estimated periodically for every M slots. Those of ordinary skill in the art will readily recognize. The DC residual error eliminating apparatus of the present invention operates at the input stage of the analog-to-digital converter 112, thereby preventing saturation of the analog-to-digital converter by DC bias.

8 is a graph illustrating the performance of the DC residual error eliminator through simulation. The channel is modeled as a non-selective fading channel with a group and a ray according to the characteristics of the Bluetooth system in Equation (7). The rms delay spread is assumed to be 70ns and the maximum power delay spread is assumed to be 200ns. In addition, the channel is modeled into a fixed channel with a slow conversion time in a slot of 625 占 of a Bluetooth reception mode. FIG. 8 compares the performance of the proposed Bluetooth residual error eliminator with that of the conventional Bluetooth in terms of the ratio of the maximum size of the demodulated signal. The DC residual error elimination device proposed in FIG. 8 shows excellent performance even in a noisy environment compared with the DC residual deviation eliminator using the existing preamble and trailer.

FIG. 9 shows the results of comparing the performance of the DC residual error eliminator with the BER. The channel environment used in this simulation is the same as the DC residual error eliminator environment, and DC bias assumes that 10% of the maximum amplitude of the demodulated signal is present. Simulation results show that the proposed DC residual error estimator obtains about 0.8dB at BER 10 ^-2 and about 4dB at 10 ^-3 than conventional Bluetooth using preamble and trailer of access code.

The proposed DC residual error estimator described above is structured to operate at the input stage of the analog-to-digital converter 112. It is possible to remove the DC residual deviation from the signal? (T) of the demodulation block 101 which is converted into the digital signal by the analogue digital converter 112, which operates in the digital domain.

10 shows the structure of a DC residual error eliminating device in the digital domain. The DC residual error eliminating device operates after the output terminal of the analog-to-digital converter 202 connected to the downstream end of the demodulation block 201. The DC residual error eliminator of FIG. 10 also uses the synchronization word of the access code which is equal to or greater than the threshold value detected in the correlation detection block 203 as shown in FIG. The access code to be used is determined to be one of the access codes stored in the access code storage unit 204 according to the state of the Bluetooth device and the number of data symbols to be used is N _p +1 To the number of symbols of the N _f -th access code. The reference signal m (k) to be used for the DC residual deviation is expressed by Equations (14) to (17) of the DC residual error elimination device of FIG. In the embodiment of FIG. 10, the reference signal m (k) in Equation 16 is transformed as shown in Equation 23 by using the data sampled at the symbol rate.

The demodulated signal (?) In the demodulation block 201 is sampled at the symbol rate in the analog-to-digital converter 202 and defined as in Equation (24).

The demodulated signal? In the demodulation block 201 can be summarized in relation to the reference signal m (k) as shown in expression (25).

Where p is the normalization constant defined in equation (21) Is the DC residual error, and e is the residual error. To use a simple LS, a vector such as Equations 26 to 29 is defined.

The DC residual error estimation using the LS method derives ξ _DC which minimizes the error exponential function as shown in equation (30).

The DC residual deviation (? _DC ) by the LS criterion is the solution of equation (31).

The DC residual deviation (? _DC ) according to Equation (31) is expressed by Equation (32).

The reference signal m (k) corresponding to the access code selected in the access code storage unit 204 is selected by the Gaussian low-pass filter lookup table 205 and the selected reference signal m (k) (T) demodulated in the demodulation block 201 is sampled at the symbol rate in the analogue to digital converter 202 as shown in equation (24), and the digital signal (K).

The subtractor 207 subtracts the normalized signal of the normalization unit 206 from the demodulated signal? (K) of the analog-to-digital converter 202 as shown in equation (33) to calculate the value C _D (k).

The integrator 208 multiplies the value C _D (k) of the subtractor 207 by the following equation (34) And outputs the integrated value (S _D ).

An integrated value (S _D) of the accumulator 208 and the average value is calculated is divided by the divider 209 as shown in Equation 35, the average value of the divider 209 denotes the DC Offset (ξ _DC) .

It can be seen that the DC residual deviation (? _DC ) detected in the equation (35), that is, the divider (209), is equal to the value detected in the equation (32).

The DC Offset (ξ _DC) of the acid 209 after receiving the access code is stored in latch 210, DC Offset estimate (ζ _DC) is stored in latch 210, the remaining packet symbols, that is, packets It is used to remove DC residual deviations while receiving headers and payloads. That is, the subtractor 211 subtracts the DC residual error estimate? _DC of the latch 210 from the signal? (K) from the analog-to-digital converter 202 so that the output of the subtractor 211 is the DC residual error It is removed.

However, if the normal constant p differs from the equation 21 due to the influence of the channel and the receiver, then the normal constant p ) Can be obtained.

From Equation (36), the normalization constant can be obtained as shown in Equation (37).

As described above, in the present invention, the access code is used as a training signal for eliminating the DC residual deviation, so that the DC residual deviation can be accurately removed in the Bluetooth system.

Claims (6)

A DC residual deviation eliminating device of a Bluetooth unit which receives an access code, A demodulation block for receiving and demodulating the access code in the received signal; An access code storage unit storing a plurality of types of access codes; A correlation detecting block for detecting a correlation between an access code of the demodulation block and an access code stored in the access code storage and providing access code information exceeding a predetermined threshold; A first means for calculating and storing an average value of an access code of the correlation detection block; Second means for calculating a normalization value by multiplying an average value of the access signal from the demodulation block by a predetermined normalization constant; A first subtracter for calculating a DC residual deviation value by subtracting an average value of the first means from the normalized value of the second means; And a second subtractor for subtracting the DC residual deviation value of the first subtracter from the received signal of the demodulation block to eliminate the DC residual deviation of the received signal. The method according to claim 1, Wherein the first means comprises: A Gaussian low-pass filter lookup table for Gaussian low-pass filtered access codes designated by the correlation detection block and outputting a sampled reference value at a symbol rate; An accumulator for accumulating a reference value of the Gaussian low-pass filter look-up table; A first divider for calculating an average value of the accumulated reference signals; And a first latch for storing an output of the first divider. 3. The method according to claim 1 or 2, The second means comprises: An integrator for integrating a signal of the demodulation block; A second latch for storing the integrated signal; And a second divider for multiplying the signal of the second latch by the normalization constant and obtaining an average value thereof. A DC residual deviation eliminating device of a Bluetooth unit which receives an access code, A demodulation block for receiving and demodulating the access code in the received signal; An analog-to-digital converter for digitizing an output signal of the demodulation block; An access code storage unit storing a plurality of types of access codes; A correlation detecting block for detecting a correlation between an access code of the demodulation block and an access code stored in the access code storage and providing access code information exceeding a predetermined threshold; A first means for obtaining a difference value between an access code of the analog-digital converter and an access code stored in the access code storage unit; Second means for obtaining an average value of the first means as a DC residual deviation value; And a first subtractor for subtracting the DC residual deviation value of the second means from the output of the analog-digital converter. 5. The method of claim 4, The first means A Gaussian low-pass filter lookup table for Gaussian low-pass filtered access codes designated by the correlation detection block and outputting a sampled reference value at a symbol rate; And a normalization unit that multiplies a reference value of the Gaussian low-pass filter lookup table by a predetermined normalization constant. The method according to claim 4 or 5, The second means comprises: An accumulator for accumulating an output of the first means; A divider for obtaining an average value of the output of the integrator; And a latch for storing the output of the divider and providing the output to the subtractor.