KR20090049728A - Mb-ofdm receiver and dc-offset estimation and compensation method thereof - Google Patents

Abstract

A signal receiving method of a multiband orthogonal frequency division multiplexing system according to the present invention comprises the steps of: (a) detecting signal power for received symbols; (b) performing automatic gain control on the received symbols in response to detecting the signal power; And (c) estimating a DC offset included in the received symbols before the gain adjusted through the automatic gain control converges.

According to the above-described method for estimating and compensating DC offset, the time for operations such as fine symbol timing synchronization and frequency offset estimation in an ultra wideband (UWB) wireless communication having a short preamble, particularly in a receiver of a multiband orthogonal frequency division multiplexing system It can be secured.

Description

Receiver of Multiband Orthogonal Frequency Division Multiplexing System and Its DC Offset Estimation and Compensation Method {MB-OFDM RECEIVER AND DC-OFFSET ESTIMATION AND COMPENSATION METHOD THEREOF}

The present invention relates to a communication system, and more particularly to a receiver of a multi-band orthogonal frequency division multiplexing system (MB-OFDM) and a method for estimating and compensating its direct current offset.

Ultra-Wide Band (UWB) wireless communication technology is a revolutionary next-generation wireless transmission technology that transmits data in the ultra-wide frequency band using very low power. In the field of wireless communication, the demand for frequency is increasing rapidly compared to limited frequency resources. One way to solve this problem is the interest of ultra-wideband (UWB) wireless communication technology, which uses frequency resources more efficiently by sharing the frequency spectrum of existing communication systems.

Ultra-wideband (UWB) wireless communication technology can build a high-speed, high-performance wireless network at very low power, providing a reliable system. Based on these characteristics, ultra-wideband (UWB) wireless communication technology is used in particularly high-reliability applications such as collision avoidance equipment for aircraft, altimeters that measure altitude from the ground in aircraft and other aeronautics, and location tracking. come. In addition, ultra-wideband (UWB) wireless communication technology is recognized as an important technology that can have a significant impact on the medical field, such as checking the patient's condition, maternal fetal health examination, human physical examination. However, because UWB wireless communication technology uses the ultra wide band, it may cause interference with radio frequencies used in global positioning systems (GPS) and mobile communication networks. Therefore, the Federal Communications Commission (FCC) has banned the commercial use of ultra-wideband (UWB), but has recently allowed commercial use of this technology.

The UWB wireless communication system defined by the Federal Communications Commission (FCC) refers to a communication method having a bandwidth of 20% or more of the center frequency or a frequency bandwidth of 500 MHz or more. The Federal Communications Commission (FCC) currently sets limits on transmit signal power for the 3.1-10.6 GHz frequency band for communications purposes. Ultra-wideband (UWB) signals have a very low power spectrum density in the frequency domain because they can use a wide frequency band. Since the power spectral density is low, signals of an ultra wideband (UWB) wireless communication system provide little interference even when superimposed on a frequency at which other communication signals exist. Initially, the UWB-based signals obtained wide frequency bands by using short pulses, but are now multi-band orthogonal frequency division multiplexing systems (MB-OFDM) and direct-sequence ultra-wideband (DS-UWB). ) Is becoming commonplace. Multi-band Orthogonal Frequency Division Multiplexing System (MB-OFDM) technology meets the transmit signal power regulations set forth by the Federal Communications Commission (FCC) and provides multiple Simultaneous Operating Piconet (TF) while minimizing power consumption. Frequency) jump pattern is used. The multi-band orthogonal frequency division multiplexing system (MB-OFDM) is different from the conventional orthogonal frequency division multiplexing system (OFDM) in that the frequency must be changed according to a time frequency (TF) leap pattern for each OFDM symbol. Except for this, the multi-band orthogonal frequency division multiplexing system (MB-OFDM) transmits data in parallel to sub-carriers having respective orthogonalities as in the transmission scheme of the orthogonal frequency division multiplexing system (OFDM). It is a communication method.

In a multi-band orthogonal frequency division multiplexing system (MB-OFDM) receiver, the signal is demodulated by a direct conversion method using no intermediate frequency. In the multi-band orthogonal frequency division multiplexing system (MB-OFDM), the advantage of using intermediate frequencies is not available. Therefore, the biggest problem is the problem of DC-Offset. DC offsets have a major impact on devices such as analog-to-digital converters (ADCs). The DC offset varies the level of the transmitted signal, increasing the bit error in the demodulated data. Therefore, the DC offset should be removed from the received signal because it causes the performance degradation of the receiver.

1 is a block diagram schematically illustrating a receiving end of a general multi-band orthogonal frequency division multiplexing system (MB-OFDM) system. Referring to FIG. 1, the received signal r (t) received through the antenna 10 is provided to the low noise amplifier 20. The received signal R (t) amplified through the low noise amplifier 20 is mixed by the local oscillation signal A (t) provided from the local oscillator 30 and the mixer 40 to be converted into a baseband signal. do.

However, the amplified received signal R (t) leaks into the local oscillator 30 and the mixing of the same received signal component occurs by the received signal R '(t) mixed in the mixer 40. In addition, the local oscillation signal A (t) output from the local oscillator 30 for tuning is mixed with the local oscillation signal A '(t) reflected by the antenna 10. The action of mixing the same signal components by leakage is called self-mixing. The self-mixing increases the DC component in the baseband signal, and the increase in the DC component means an increase in the DC offset. The DC offset saturates the amplitude component of the baseband signal, resulting in nonlinear distortion, which increases the bit error of the analog-to-digital converter (ADC), which in turn reduces the signal-to-noise ratio (SNR) of the receiver.

2 is a simplified block diagram of a transmitter 100 of a multi-band orthogonal frequency division multiplexing system (MB-OFDM). Referring to FIG. 2, a payload sequence provided from a media access control (MAC) layer is provided as input data. Input data corresponding to the payload sequence is encrypted by the decryptor 105 and provided with security.

The header generator 110 generates a header, which is control information provided from a physical layer (PHY, not shown) of the multi-band orthogonal frequency division multiplexing system (MB-OFDM). The header is combined with the input data secreted by the multiplexer 115 to form a transmission data sequence. The transmission data sequence is then channel coded by a convolutional encoder 120. The interleaver 125 performs an interleaving operation to prepare for a burst error for the channel-coded transmission data sequence. The interpolated transmission data sequence is combined by the preamble provided by the preamble generator 130 and the multiplexer 135 to form one packet.

One packet includes a preamble, a header, and payload sequences. The preamble is a data sequence for providing timing synchronization for transmission data to the receiver. In general, the preamble consists of a packet / frame synchronization sequence and a channel estimation sequence. The packet is mapped to the signal space by the modulation mapper 140 by a specified specific modulation scheme. This mapping is called digital modulation. For example, a packet may be modulated in signal space using a modulation scheme such as binary phase shift keying (BPSK), quadrature phase shift keying (QPSK), or quadrature amplitude modulation (QAM). Is mapped to. Digitally modulated packets, which are time domain data, are converted into frequency domain data by an Inverse Fast Fourier Transformer (145).

The inverse transform data into the frequency domain is delivered to the guard interval mixer 150, and a guard interval (GI) is added to avoid inter-symbol interference (ISI) caused by the multipath channel. The inverse transform data of the frequency region to which the guard period GI is added is transmitted to the symbol waveform shaping unit 155. The symbol waveform shaping unit 155 shapes the symbol pulse waveform to maximize the channel capacity in the limited channel bandwidth of the transmission channel. In general, a rectangular shape occupies a minimum bandwidth in the frequency domain, and a rising cosine method may be used to generate a pulse occupying this frequency domain. The shaped data is then converted by the digital-to-analog converter 160 into an analog signal for transmission to the channel at radio frequency. The analog signal is mixed with a band number provided according to a hopping pattern (where the band number means a subcarrier frequency signal for selecting a band when tuning) and the transmission signal S (t) is hopped and transmitted in multiple bands. Generated by).

According to the transmitter 100 of the multi-band orthogonal frequency division multiplexing system (MB-OFDM), the transmitted signal includes a preamble for time synchronization. The waveform of the transmitted signal hopping of the bands according to the hopping pattern will be described in detail later with reference to FIG. 3.

3 is a waveform diagram schematically illustrating a transmission signal S (t) transmitted through a multiband. Referring to FIG. 3, the transmission signal S (t) exemplarily shows a schematic waveform of 3-OFDM in which a symbol is transmitted through three bands. The portion where the signal is concentrated in each band corresponds to a symbol of one orthogonal frequency division multiplexing system (OFDM). The symbols from symbol 1 to symbol 24 are sequences used for synchronization of packets or frames of the receiver. Subsequently, six symbols corresponding to the channel estimation sequence and a header corresponding to the header and the payload sequence containing the transmitted data information are followed. In a multi-band orthogonal frequency division multiplexing system (MB-OFDM), the length of each symbol is variable depending on the communication scheme. Symbol 1 is transmitted on band 1, symbol 2 is transmitted on band 2, and symbol 3 is transmitted on band 3. The hopping pattern of these symbols may be generally provided as a pseudo-noise code (PN code). However, for the sake of simplicity, the band number leap is assumed to be a regular case such as (1-2-3-1-2-3-…).

4 is a table briefly showing various examples of a hopping pattern of a band in which each symbol hops. Referring to FIG. 4, the security of the system may be provided by providing various jump patterns with the number of various cases.

Pattern 1 is identical to the way in which the sequence of hopping band numbers is described in FIG. That is, the hopping pattern assigned to the symbols has a band number sequence of (1-2-3-1-2-3 -...). In pattern 2, the hopping pattern assigned to the symbols is assigned to the multiband in a band number sequence of (1-3-2-1-3-2 -...). Pattern 3 is a band number sequence (1-1-2-2-3-3-…), and pattern 4 is each of the multiple bands according to the band number sequence (1-1-3-3-2-2-…). The symbols are allocated and transmitted. And, pattern 5 is assigned to use only band 1 without jumping to other symbols, pattern 6 to use band 2 only, and pattern 7 to use band 3 only. Here, although the hopping patterns for the three bands have been described by way of example, it is obvious to those skilled in the art that the examples of the described patterns are only a part of the implementable patterns.

Considering the transmission signal symbol structure of the multi-band orthogonal frequency division multiplexing system (MB-OFDM), the estimation and compensation of the DC offset is generally performed after the automatic gain control unit AGC has sufficiently converged. However, in the multi-band orthogonal frequency division multiplexing system (MB-OFDM) for the ultra wideband (UWB) communication scheme, it has a relatively short preamble. In addition, it is necessary to secure time to perform functions such as fine symbol timing or frequency offset estimation that must be additionally provided. Therefore, there is a need for a technique for quickly estimating a DC offset before the output of the automatic gain regulator AGC sufficiently converges. In addition, there is an urgent need for a technique of increasing the accuracy of DC offset estimation through independent DC offset estimation and compensation for each of multiple bands.

The present invention provides a multi-band Orthogonal Frequency Division Multiplexing System (MB-OFDM) receiver capable of estimating and compensating DC offsets at high speed and a DC offset estimation and compensation method thereof.

A signal receiving method of a multiband orthogonal frequency division multiplexing system of the present invention for achieving the above object comprises: (a) detecting signal power for received symbols; (b) performing automatic gain control on the received symbols in response to detecting the signal power; And (c) estimating a DC offset included in the received symbols before the gain adjusted through the automatic gain control converges. Through the above-described steps, estimation of the DC offset is started before the automatic gain control operation is completed, thereby enabling fast DC offset estimation and compensation.

A receiver of a multiband orthogonal frequency division multiplexing system of the present invention for achieving the above object comprises: a mixer for selecting received symbols corresponding to each of a plurality of bands from a received signal according to a band number; An automatic gain control loop for adjusting signal levels of received symbols output from the mixer; An analog-to-digital converter for converting received symbols amplified by the automatic gain control loop into a digital signal; The signal power of the received symbols converted into the digital signal is detected to provide the band number, and the DC offset included in the received symbols converted into the digital signal before the gain of the automatic gain control loop converges is estimated. A signal detection and offset estimating unit for estimating; And an adder for removing a DC offset included in received symbols output from the analog-digital converter using the estimated DC offset. The receiver of the ultra-wideband multi-band orthogonal frequency division multiplexing system of the present invention having the above configuration performs fast estimation and removal of a DC offset in a short preamble. Therefore, it is possible to provide time margin for performing operations such as fine timing adjustment and frequency offset adjustment.

A multi-band orthogonal frequency division multiplexing system (MB-OFDM) receiver according to the present invention through the above configuration is a time for operation such as fine-tuning or estimation of frequency offset by fast estimation and compensation of the DC offset for each of the multi-band Can provide room.

It is to be understood that both the foregoing general description and the following detailed description are exemplary, and that additional explanations of the claimed invention are provided. Reference numerals are shown in detail in preferred embodiments of the invention, examples of which are shown in the reference figures. In any case, like reference numerals are used in the description and the drawings to refer to the same or like parts. DETAILED DESCRIPTION Hereinafter, exemplary embodiments of the present invention will be described with reference to the accompanying drawings so that those skilled in the art may easily implement the technical idea of the present invention.

5 is a block diagram schematically illustrating a receiver structure capable of estimating and compensating a direct DC offset according to the present invention. Referring to FIG. 5, the receiver 200 of the present invention receives a signal r (t) and receives a signal detection and offset estimator for estimating a DC offset before the automatic gain controller 260 converges. 290). In more detail,

The weak received signal r (t) obtained from the antenna 210 is amplified by the low noise amplifier 220. The received power of the received signal r (t) has a very low power level due to the effects of attenuation and noise. Since the received signal r (t) is also a signal that has already been mixed with a lot of noise in the channel, it must be amplified to minimize noise first of all. Therefore, low current consumption should be designed using transistors with low noise figure (NF) or low thermal noise (eg resistance). In general, a band pass filter (BPF) is inserted to select a signal of multiple bands (band 1, band 2, band 3) from the received signal r (t). However, since a filter configuration such as a band pass filter is far from the functional description of the present invention, it will be omitted for simplicity.

The received signal R (t) amplified by the low noise amplifier 220 is passed to the mixer 230 so that any one of the multiple bands is selected. The mixer 230 selects any one band signal according to a hopping pattern among a plurality of bands included in the received signal R (t). The hop pattern is implemented with a regular sequence of band numbers. That is, the mixer 230 selects one of the band signals included in the multi-band according to the band number provided from the signal detection and offset estimator 290. The selected signal is demultiplexed into an in-phase component and a quadrature-phase component. That is, the selected band signal is demultiplexed into orthogonal signals Ia and Qa. Two demultiplexed orthogonal signals Ia and Qa, which are mutually orthogonal, are processed independently of each other, but will be described as one component in the following description. For demultiplexing, the selected signal and carrier signals each with a -90 ° phase are mixed through a multiply modulator. However, the configuration for demultiplexing will be omitted here. However, since the first received signal is before the fine symbol timing is set, the mixer 230 selects the default value (for example, band 1). After the signal power is detected by the signal detection and offset estimator 290 and the fine symbol timing is adjusted, the band will be selected according to the hopping pattern.

The quadrature signals Ia and Qa obtained thereafter are provided to the variable gain amplifiers 240 and 245, respectively. The variable gain amplifiers 240 and 245 amplify the quadrature signals Ia and Qa according to the adjusted gains fed back from the automatic gain control unit 260 and the digital-analog converter 270.

The quadrature signals Ia 'and Qa' amplified by the variable gain amplifiers 240 and 245 are converted into quadrature digital signals Id and Qd by the analog-to-digital converters 250 and 255. The quadrature digital signals Id and Qd are first provided to the automatic gain control unit 260. The auto gain control loop configuration of the automatic gain control unit 260, the digital-to-analog converter 270, and the variable gain amplifiers 240 and 245 is based on the level of the input signal of the analog-to-digital converters 250 and 255. Keep it constant. Therefore, it is possible to receive a stable signal against the level variation of the received signal. The automatic gain controller 260 selects an optimum gain by referring to the outputs of the analog-to-digital converters 250 and 255. In particular, while the preamble is received, the automatic gain controller 260 measures the signal power of the digital signals Id and Qd corresponding to the preamble section. Then, the predetermined threshold is compared with the signal power of the preamble to provide the gain of the variable gain amplifiers 240 and 245. The digital-analog converter 270 converts the gain of the variable gain amplifiers 240 and 245 provided in the form of digital data from the automatic gain control unit 260 into an analog signal such as a voltage signal. The gain converted to the analog signal is then provided to the variable gain amplifiers 240 and 245. This gain control of the automatic gain control unit 260 takes place in the time interval at which the preamble is received and converges to the optimum gain. In addition, the converged gain will be fixedly provided in the section in which the channel estimation sequence and the header and data sequence are received.

The adders 280 and 285 output digital signals Id 'and Qd' from which the DC offset is removed by adding or subtracting a DC offset included in the digital signals Id and Qd output from the analog-to-digital converters 250 and 255. do. The digital signals Id 'and Qd' will then be provided to a baseband processor (not shown) and demodulated into the transmitted data.

The signal detection and offset estimator 290 receives the digital signals Id 'and Qd' and the DC offset estimation enable signal DCE_EN to perform two main functions. Here, the DC offset estimation enable signal DCE_EN is provided from a timing controller of a reception modem after a predetermined time (for example, T1 of FIG. 6A) when the power of the reception signal is detected. Note that it is a control signal.

First, the signal detection and offset estimator 290 detects signal power from the digital signals Id 'and Qd'. The signal detection and offset estimator 290 senses the power of the samples of the symbols included in the digital signals Id 'and Qd' to obtain appropriate timing information for estimating and compensating the presence of a received signal and a direct current offset. . Timing synchronization of the received signal is performed according to the timing information, and as a result, a band number on which timing synchronization is performed can be provided.

Second, the signal detection and offset estimator 290 receives the digital signals Id 'and Qd' whose gain is adjusted and measures the DC offset estimates DC_I and DC_Q for each band and feeds them back to the adders 280 and 285. do. The adders 280 and 285 add or subtract DC offset estimates DC_I and DC_Q from the digital signals Id and Qd to remove DC offsets included in the signal. Here, the estimation of the DC offset of the signal detection and offset estimator 290 is performed before the outputs of the variable gain amplifiers 240 and 245 by the automatic gain controller 260 converge. That is, the signal detection and offset estimator 290 estimates the DC offset before the gain control operation of the automatic gain control loops 240, 250, 260, and 270 is completed. Estimation of the DC offset is performed with different parameters for each of before and after gain convergence of the automatic gain control unit 260. Prior to the convergence of the gain, the DC offset for each of the multiple bands is calculated using the recently converged gain as a parameter. After the convergence of the gain, the DC offset is calculated by dividing the number of samples per symbol by the accumulated value ACC of the digital signals Id 'and Qd' (symbol unit) without considering the gain of the automatic gain controller 260. Before the above-mentioned gain convergence and after the gain convergence may be determined by counting the number of inputs of the DC offset estimation enable signal DCE_EN provided from the time synchronization controller (not shown) of the reception modem. However, it is obvious to those skilled in the art that the determination of the switching time point of the DC offset calculation method is not limited to the method of determining the DC offset estimation enable signal DCE_EN by referring to the count number. The detailed configuration and operation of the signal detection and offset estimator 290 will be described later with reference to FIGS. 6A, 6B, and 7.

Through the receiver configuration of the present invention, the DC offset can be estimated before the gain control of the received signal is completed. Therefore, the receiver of the multi-band orthogonal frequency division multiplexing system according to the present invention can sufficiently provide a time margin for performing operations such as fine symbol timing or frequency offset estimation within a short preamble sequence.

6A and 6B are timing diagrams illustrating time distributions of symbols provided through multi-bands and removal of a DC offset performed within a preamble with respect to received signals having different hopping patterns. FIG. 6A shows a received signal whose hop pattern corresponds to a band number sequence (1-1-2-2-3-3…), and FIG. 6B shows a hop pattern showing a band number sequence (1-2-3-1-2). -3 ...) shows an embodiment of a received signal corresponding to "

Referring to FIG. 6A, reception band selection in the mixer 230 (see FIG. 5) assumes that band 1 is set to a default value. When a received signal is provided, the initial power detection (PD) is detected in band 1. That is, signal power is detected at the time of reception of symbol 1 in the figure. When the signal power is detected, it means that the received signal is present, and therefore, the automatic gain control unit 260 starts gain adjustment of the variable gain amplifiers 240 and 245. When a predetermined time T1 elapses from the detection of the signal power, the estimation of the DC offset is performed when the gain of the automatic gain control unit 260 does not converge. In the timing diagram, the estimation section of the DC offset is shown as a time interval T2. In the time interval T2, the signal detection and offset estimator 290 receives the orthogonal digital signals Id 'and Qd' to estimate the DC offset for each band. In FIG. 6A, since the hopping pattern of each band is performed twice for each band, the DC offset estimation enable signal DCE_EN is provided to the signal detection and offset estimation unit 290 three times. However, the DC offset estimation enable signal DCE_EN for each symbol may be performed for every symbol. That is, it is apparent to those who have acquired the general knowledge in this field that six DC offset estimation enable signals DCE_EN may be transmitted to the signal detection and offset estimation unit 290 in the time interval T2. In the time interval T2, the gain adjustment of the automatic gain control unit 260 continues with the DC offset estimation operation. At the end of the time interval T2, the gains of the variable gain amplifiers 240 and 245 by the automatic gain controller 260 converge to a fixed value. Therefore, since the gain is fluctuating in the time interval T2, the signal detection and offset estimator 290 according to the present invention refers to the gains of the variable gain amplifiers 240 and 245 that have converged previously. Estimate the DC offset of band 2, band 3). The signal detection and offset estimator 290 transmits the estimated DC offsets DC_I and DC_Q to the adders 280 and 285 in the time interval T3 when the gain adjustment of the automatic gain controller 260 is completed. The DC offsets of the respective bands included in the digital signals Id and Qd output from the analog-to-digital converters 250 and 255 by the adders 280 and 285 are removed.

Fig. 6B shows the control of the automatic gain and the estimation operation of the DC offset corresponding to the case where the hop pattern is the band number sequence (1-2-3-1-2-3 ...). Referring to FIG. 6B, reception band selection in the mixer 230 (refer to FIG. 5) assumes that band 1 is set to a default value. When a received signal is provided, the initial power detection (PD) is detected in band 1. That is, signal power is detected at the time of reception of symbol 1 in the figure. When the signal power is detected, it means that the received signal is present, so the automatic gain control unit 260 starts adjusting the gain of the variable gain amplifiers 240 and 245. When a predetermined time T1 has elapsed from the detection of the signal power, the signal detection and offset estimator 290 performs estimation of the DC offset at a point in time when the gain by the automatic gain control unit 260 did not converge. . That is, the control of the gain by the automatic gain control unit 260 and the estimation of the DC offset for each band by the signal detection and offset estimation unit 290 are performed in the time interval T2. A detailed method of estimating the DC offset by the signal detection and offset estimator 290 will be described in more detail later with reference to FIG. 7. In the time interval T2, the signal detection and offset estimator 290 receives the orthogonal digital signals Id 'and Qd' to estimate the DC offset for each band. Here, the DC offset estimation enable signal DCE_EN provided in FIG. 6A is performed by the DC offset estimation enable signal DCE_EN at the end of each symbol according to a characteristic of a jump pattern in which successive symbols jump to different bands each time. Should be. Accordingly, the DC offset estimation enable signal DCE_EN may be input six times in the time interval T2. In the time interval T2, the gain adjustment of the automatic gain control unit 260 is parallel with the DC offset estimation operation. At the end of the time interval T2, the gains of the variable gain amplifiers 240 and 245 by the automatic gain control unit 260 converge to a fixed value. Therefore, since the gain is fluctuating in the time interval T2, the signal detection and offset estimator 290 according to the present invention refers to the gains of the variable gain amplifiers 240 and 245 that have converged previously. Estimate the DC offset of band 2, band 3). The signal detection and offset estimator 290 transmits the estimated DC offsets DC_I and DC_Q to the adders 280 and 285 in the time interval T3 when the gain adjustment of the automatic gain controller 260 is completed. The DC offsets of the respective bands included in the digital signals Id and Qd output from the analog-to-digital converters 250 and 255 by the adders 280 and 285 are removed.

According to the timing diagrams of FIGS. 6A and 6B, the receiver according to the present invention estimates the DC offset with reference to the previously converged gains during the time period T2 before the automatic gains converge. The DC offset is removed immediately after the gain adjustment by the automatic gain controller 260 is completed using the estimated DC offset. Therefore, the gain is fixed in the time interval T3, and the estimation and removal of the DC offset will be parallel. In accordance with the DC offset estimation and removal scheme described above, the receiver of the multi-band orthogonal frequency division multiplexing system (MB-OFDM) of the present invention performs the removal of the DC offset quickly to perform operations such as fine time adjustment within a short preamble. It can give you some time to do it.

7 is a block diagram schematically illustrating an internal configuration of the signal detection and offset estimation unit 290 of the present invention. Referring to FIG. 7, the signal detecting and offset estimating unit 290 may include a DC offset estimating unit 296 that starts detecting a DC offset for each band in response to a DC offset estimation enable signal DCE_EN, and an input digital signal. And a signal detector 299 for detecting signal power of signals Id 'and Qd' and generating a band number.

The DC offset estimator 296 receives the digital signals Id 'and Qd' composed of in-phase and quadrature signals and provides DC offsets DC_I and DC_Q for each band. The DC offset estimator 296 accumulates the samples in the symbol unit and calculates an average value of the offset calculator 291 and the offset of each band according to the count of the DC offset estimation enable signal DCE_EN. Average filters 292, 293, 294 to provide continuously.

The offset operator 291 accumulates the samples of the provided digital signals Id 'and Qd' in symbol units. The algebraic sum of the samples accumulated in the symbol unit will be referred to as an accumulated value (ACC). The accumulated value ACC is generated as a DC offset by applying different parameters according to the count value of the DC offset estimation enable signal DCE_EN. That is, when there is no input of the DC offset estimation enable signal DCE_EN, when the DC offset estimation enable signal DCE_EN is less than N times, in each case where the DC offset estimation enable signal DCE_EN is N or more times Estimate the DC offset with different parameters. Here, N is the number of counts of the DC offset estimation enable signal DCE_EN input until the gain adjustment of the automatic gain controller 260 is completed. In general, a time average of samples included in an OFDM symbol may be characterized as a random process having zero or a very small value. Accordingly, the accumulated value ACC corresponding to the logarithm sum of the pulses included in the received signal converges to O for a sufficiently large sample. However, for one symbol of a limited sample, the accumulated value (ACC) is zero or very small. Therefore, the DC offset can be obtained by summing all the samples included in the predetermined interval (for example, the symbol interval) and calculating the average thereof. The offset operator 291 may also be implemented as a finite impulse response (FIR) filter that can calculate the arithmetic mean, it is obvious to those skilled in the art. 6A and 6B, the time point at which the DC estimation enable signal DCE_EN is input corresponds to the convergence of the gains of the automatic gain control unit 260 (see FIG. 5). The gain of the automatic gain control unit 260 is not stable during the time interval T2. Therefore, during the time interval T2, the accumulated value ACC may be divided by a value obtained by multiplying the recently estimated previous gain by the number of samples per symbol to compensate for the gain of the unstable automatic gain controller 260. However, the DC offset is estimated by dividing the accumulated value ACC by the number of samples per symbol during the time interval T3 where the gain estimated by the automatic gain controller 260 converges. A detailed operation algorithm of the offset calculator 291 briefly described above will be described in detail later with reference to FIG. 9.

The DC offset calculated by the offset calculator 291 is transferred to the first to third average filters 292, 293, and 294. The first to third average filters 292, 293, and 294 obtain an arithmetic mean value of a plurality of symbols of a DC offset for each of the bands. The physical meaning of the mean filters is to minimize the error by increasing the number of random sample sets. Therefore, the accuracy of the DC offset is increased by the first to third average filters 292, 293, and 294. Here, the first to third average filters 292, 293, and 294 are configurations for obtaining an average value of offsets for each of three bands used in 3-OFDM. Therefore, the number of average filters can be changed according to the number of bands. The DC offsets output from the first to third average filters 292, 293, and 294 are transmitted to the selector 295 which is switched according to the band number, and according to the band number sequence synchronized with the detection of the signal power. Offsets DC_I and DC_Q are sequentially output. The first average filter 292 provides a DC value for the symbols of band 1 provided by the offset calculator 291 as an output value with reference to previously estimated DC offsets. The second average filter 293 provides a DC value for the symbols of band 2 provided by the offset calculator 291 as an output value with reference to previously estimated DC offsets. The third average filter 294 provides a DC value for the symbols of band 3 provided by the offset calculator 291 as an output value with reference to previously estimated DC offsets. Each of the first to third average filters 292, 293, and 294 can provide an average by referring to a DC offset included in a plurality of symbols transmitted in each of the bands, thereby dramatically improving the accuracy of the DC offset estimate. It can increase.

The selector 295 selects an output of each average filter and outputs a DC offset estimate according to the band number to remove the DC offset included in the digital signals provided while hopping a plurality of bands. That is, by continuously outputting a DC offset estimate of any one of the first to third average filters 292, 293, and 294 in synchronization with the band number occupied by the symbol transmitted through the detection of the signal power, DC offset can be eliminated. That is, the DC offset estimates DC_I and DC_Q which are continuously provided are provided to the adders 280 and 285, and are added to and subtracted from the digital signals Id and Qd of in-phase and quadrature phases in the adders 280 and 285. Is removed.

The signal detector 299 provides accurate timing information for the estimation of the DC offset for each band to the DC offset estimator 296 described above. The signal power detector 297 determines whether a signal is received. If it is detected that signal power is present, the signal power detector 297 controls the band number generator 298 to output a band number sequence corresponding to the hop pattern that is already set to synchronize with the time of receipt of the signal. In general, the signal power detector 297 uses a delay and correlation algorithm to detect power of a received signal. Signal detection methods of delay and correlation algorithms are widely used for time synchronization, guard interval detection, and the like. The band number generator 298 generates a band number sequence synchronized with the transmitting side under the control of the detector 297 of signal power.

Through the above configuration, the receiver of the present invention can estimate and compensate the DC offset before the gain of the automatic gain control unit 260 of the received signal is stabilized.

8 is a flowchart schematically illustrating a method of estimating a DC offset according to the present invention described above. Referring to FIG. 8, automatic gain control is performed from the detection of signal power, and estimation of the DC offset is started before the automatic gain control is completed. The DC offset is removed or compensated using the DC offset estimated at the time when the automatic gain control is completed. Briefly described through the drawings as follows.

The receiver continuously monitors signal power in order to determine whether a received symbol exists (S10). If signal power is present, this means that there is a received symbol and thus starts automatic gain control. In accordance with automatic gain control, the gain of the variable gain amplifiers 240, 245 will be adjusted. However, the DC offset estimation operation is performed before the automatic gain control operation is completed. In this case, the DC offset is calculated by referring to a gain that has previously converged, not a gain currently being adjusted for DC offset estimation (S20). According to the automatic gain control operation, if the gain converges, the DC offset present in the received symbol is removed according to the estimated DC offset. In addition, after the convergence of the gain, the DC offset will be estimated and removed without considering the gain (S30).

9 is a flowchart schematically illustrating an embodiment of the DC offset estimation operation of FIG. 8 described above. That is, the flowchart of FIG. 9 exemplarily illustrates the operation of the offset calculator 291 of FIG. 7. Referring to FIG. 9, the offset calculator 291 performs a calculation operation of a DC offset according to the number of inputs of the DC estimation enable signal DCE_EN. Initially, when signal power is detected, it means the presence of a received signal, so that an automatic gain control operation is performed. Before the convergence of the gain by the automatic gain control unit 260 is completed, the operation of the overall DC offset is started in response to the DC estimation enable signal DCE_EN provided from the time synchronization control unit (not shown) of the receiving modem. do. The calculation of the DC offset is performed differently according to the number of pulses of the DC estimation enable signal DCE_EN. First, an operation of initializing the number of pulses of the DC estimation enable signal DCE_EN (i = 0) is performed (S110). The offset operator 291 monitors whether the DC estimation enable signal DCE_EN is input (S120). When the DC estimation enable signal DCE_EN is input, the offset calculator 291 counts the number of DC estimation enable signals DCE_EN provided after the detection of the received signal (S140). However, if the DC estimation enable signal DCE_EN is not provided, the DC offset DC_OS maintains the DC offset PRE_DC_OS of each band used when receiving a previous packet or frame (S130).

When it is detected that the DC estimation enable signal DCE_EN is input, the offset calculator 291 counts the number of DC estimation enable signals DCE_EN provided from after the detection of the received signal. In operation S150, it is determined whether the count number i of the DC estimation enable signal DCE_EN is 4 or more. If the count number i of the DC estimation enable signal DCE_EN is less than 4, it means that the gain of the automatic gain control unit 260 is not converged. Therefore, the gain estimated at the reception of the previous packet or frame is considered. Calculate the DC offset. That is, an accumulation value (ACC) obtained by adding samples of the received symbol is divided by a value obtained by multiplying the number of samples per symbol by a previous gain (S160). Accordingly, the DC offset is estimated in consideration of the unconverged unstable gain in a section in which the count number i of the DC estimation enable signal DCE_EN is less than four. If the count number i of the DC estimation enable signal DCE_EN is 4 or more, it means that the gain of the automatic gain control unit 260 has converged to a fixed value. Therefore, the DC offset DC_OS is calculated by dividing the number of samples per symbol by the accumulated value ACC (S170).

The method of calculating the DC offset according to the count number of the DC estimation enable signal DCE_EN is merely a brief implementation example of the algorithm described in Table 1 below. It is apparent to those skilled in the art that the setting of the number of counts of the DC estimation enable signal DCE_EN may vary depending on the symbol hopping pattern scheme.

 if DCE_EN == (1, 2, 3), DC_OS (n) = DC_OS (n) + ACC / (samples per symbol * previous gain); elseif DCE_EN> = 4, DC_OS (n) = DC_OS (n) + ACC / (samples per symbol); else DC_OS (n) = DC_OS (n); end

(Where n is a natural number representing the band number of each of the multiple bands)

The above algorithm describes a method of estimating a DC offset for the case of the jump pattern described in FIG. 6A. However, in the jump pattern as shown in FIG. 6B, the gain of the automatic gain control unit 260 may converge when the count of the DC offset estimation enable signal DCE_EN is greater than or equal to '7'. Therefore, in this case, the number of counts of the DC offset estimation enable signal DCE_EN of the algorithm should be switched from '4' to '7'.

According to the estimation method of the DC offset of the present invention described above, the estimation of the DC offset is started when the gain by the automatic gain control unit 260 does not converge. Therefore, it is possible to secure enough time to perform synchronization operations such as fine time adjustment and frequency offset adjustment in a receiver using a short preamble such as an ultra wideband communication system.

Meanwhile, in the detailed description of the present invention, specific embodiments have been described, but various modifications may be made without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the above-described embodiments, but should be defined by the equivalents of the claims of the present invention as well as the following claims.

1 is a block diagram for briefly showing the generation of a DC offset;

2 is a block diagram schematically showing a configuration of a transmitter of a multiband orthogonal frequency division multiplexing system;

3 is a signal waveform diagram schematically showing a waveform of a multiband signal;

4 is a simplified illustration of examples of hop patterns of a multiband signal;

5 is a block diagram schematically showing a receiver structure of a multiband orthogonal frequency division multiplexing system according to the present invention;

6A is a timing diagram illustrating one embodiment of estimation and compensation of a DC offset in accordance with the present invention;

6B is a timing diagram illustrating another embodiment of the DC offset estimation and compensation of the present invention.

7 is a block diagram schematically illustrating a configuration of a signal detection and offset estimation unit of FIG. 5;

8 is a flowchart briefly showing a method for estimating and compensating a DC offset of the present invention;

9 is a flowchart showing an embodiment of the method for estimating the DC offset of FIG. 8 described above.

* Description of the symbols for the main parts of the drawings *

10, 210: antenna 20, 220: low noise amplifier

30: local oscillator 40, 230: mixer

105: non-firearm 110: header generator

115: multiplexer 120: convolutional encoder

125: interleaver 130: preamble generator

135: multiplexer 140: modulation mapper

145: reverse fast Fourier transformer 150: guard section mixer

155: symbol waveform shaping unit 160: digital-to-analog converter

240, 245: variable gain amplifier 250, 255: analog-to-digital converter

260: automatic gain control unit 270: digital-to-analog converter

280, 285: adder 290: signal detection and offset adjustment unit

291: offset calculator 292: first average filter

293: second average filter 294: third average filter

295: selection unit 296: DC offset estimation unit

297: signal power detector 298: band number generator

299: signal detector

Claims (19)

A signal receiving method of a multiband orthogonal frequency division multiplexing system: (a) detecting signal power for received symbols; (b) performing automatic gain control on the received symbols in response to detecting the signal power; And (c) estimating a DC offset included in the received symbols before the gain adjusted through the automatic gain control converges. The method of claim 1, In response to the detection of the signal power, a hopping pattern for receiving the received symbols is synchronized with the received symbols. The method of claim 2, In the step (c), the direct current offset is estimated for each of the multi-bands. The method of claim 1, In the step (c), the direct current offset is estimated by referring to a previous gain obtained before the automatic gain control operation in consideration of the variable gain. The method of claim 4, wherein In the step (c), the DC offset is estimated by dividing the algebraic sum of the samples included in any one of the received symbols by the product of the number of samples included in the one received symbol times the previous gain. A signal receiving method characterized in that. The method of claim 5, wherein In the step (c), the signal receiving method further comprises the step of obtaining the average value of the plurality of DC offsets estimated from the plurality of received symbols included in the same band as a DC offset estimate. The method of claim 1, And (d) removing the estimated direct current offset from the received symbols when the gain converges in step (c). The method of claim 7, wherein In the step (d), the direct current offset is estimated by dividing the logarithm of the samples included in any one of the received symbols by the number of samples included in any one of the received symbols. The method of claim 1, The multi-band orthogonal frequency division multiplexing system is a signal receiving method, characterized in that implemented in the ultra-wideband (UWB) communication method. In the receiver of a multiband orthogonal frequency division multiplexing system: A mixer for selecting received symbols corresponding to each of the plurality of bands from the received signal according to the band number; An automatic gain control loop for adjusting signal levels of received symbols output from the mixer; An analog-to-digital converter for converting received symbols amplified by the automatic gain control loop into a digital signal; The signal power of the received symbols converted into the digital signal is detected to provide the band number, and the DC offset included in the received symbols converted into the digital signal before the gain of the automatic gain control loop converges is estimated. A signal detection and offset estimating unit for estimating; And And an adder for removing a DC offset included in received symbols output from the analog-digital converter using the estimated DC offset. The method of claim 10, The signal detection and offset estimator, A signal power detector detecting the signal power of the received symbols to detect the presence of the received signal; And And a DC offset estimator for estimating DC offsets of received symbols corresponding to respective bands output from the automatic gain control loop. The method of claim 11, The signal power detector, A signal power detector for detecting the presence of the received signal by detecting signal power of the received symbols; And And a band number generator for generating the band number in synchronization with the symbols when the received signal is present. The method of claim 12, The DC offset estimator, Accumulate the magnitudes of the samples included in each of the received symbols, and the accumulated value is a value obtained by multiplying the number of samples included in each of the received symbols or the number of samples included in each of the received symbols by a previous gain of the automatic gain control loop. An offset operation unit divided by; And And a plurality of average filters that classify the DC offsets output from the offset calculator into each of the plurality of bands and obtain an average value of offsets corresponding to a plurality of received symbols included in each band. The method of claim 13, Before the gain of the automatic gain control loop converges, the offset operator divides the accumulated value by a value obtained by multiplying the number of samples included in the received symbol by a previous gain of the automatic gain control loop. The method of claim 13, And after the gains of the automatic gain control loop converge, the offset operator divides the accumulated value by the number of samples included in each of the received symbols. The method of claim 13, And the gain of the automatic gain control loop is determined by referring to the count of the DC offset estimation enable signal provided from the physical layer after a predetermined time in response to the detection of the signal power. The method of claim 13, And the output of the plurality of average filters is selected by the band number. The method of claim 13, And the offset calculator further includes a finite impulse response filter (FIR filter) that accumulates the sizes of samples included in the received symbols. The method of claim 10, The automatic gain control loop is An automatic gain controller that senses the output of the analog to digital converter and adjusts gain; A digital-analog converter for converting the gain into an analog signal; And a variable gain amplifier adjusting the signal level of the received symbols output from the mixer according to the gain converted into the analog signal and providing the received signal to the analog-to-digital converter.

Images