JP2011228890A - Radio communication equipment and reception signal processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve performance in packet detection on a reception side of an OFDM signal and in synchronization processing with a symbol by reducing power consumption without increasing a circuit scale.SOLUTION: In a stage of a processing of a synchronization processing portion 17 before an equalization processing is performed by an equalizer 14, a propagation characteristic estimating portion 18 obtains a characteristic when a packet is received before. A pattern signal correcting portion 20 corrects a training pattern signal which is used in a synchronization processing and is previously decided by using the propagation characteristic used for the equalization processing to absorb fluctuation of the propagation characteristic. In the stage of the synchronization processing, effect equal to the equalization processing of the reception signal is obtained, and the performance of the synchronization processing is improved to improve effective transmission performance as a result. The pattern signal correcting portion 20 is started only when the propagation characteristic is obtained to perform a correction processing without newly adding a circuit. Thus, a circuit scale is not increased and power consumption can be reduced.

Description

  The present invention relates to a communication device that performs data transmission wirelessly, for example, used in a wireless LAN, and more specifically, an orthogonal frequency division multiplexing (OFDM) packet formed by combining symbol units of a predetermined length. The present invention relates to a radio communication apparatus and a received signal processing method provided with means for improving performance in packet detection and symbol synchronization processing performed on the receiving side when transmitting and receiving with a Multiplexing signal.

In recent years, various information devices used in offices and homes have been provided with a mobile function by enabling wireless devices to exchange information using wireless communication technology. In this type of wireless communication technology, for example, wireless LAN (Local Area Network) communication based on IEEE802.11 standard (IEEE802.11a, 802.11b, 802.11g, etc.), wireless USB (Universal Serial Bus) standard, etc. Based on the wireless PAN (Personal Area Network) communication and the like, various products are put on the market.
Since wireless communication uses a limited radio wave resource (frequency band), it is necessary to operate the communication by a communication method conforming to the Radio Law or a communication standard. As a wireless standard, IEEE802.11 standard communication terminals have recently been widely used.

In addition, IEEE802.11 standard and other recent wireless communication systems use a multicarrier transmission technique called OFDM, and high-speed data transmission can be realized in wireless communication.
In addition, most of recent wireless data communication is based on a packet transmission method, and a training signal is added to a preamble at the head of a packet to be transmitted for transmission. In wireless communication using the OFDM system, various factors such as multipath fading due to a radio wave propagation environment, frequency offset, phase noise, quadrature error, transmission power amount, and nonlinear distortion affect reception performance.
The receiver needs to be designed with these points in mind, and wireless communication is established corresponding to the communication environment by using static or dynamic correction and equalization (equalization) technology. Is configured to do.

For example, in OFDM wireless communication, a receiver uses a training symbol (pattern signal) added to a preamble at the head of a transmission data packet in order to correct propagation characteristics that vary depending on a radio wave propagation path. For each packet, frequency offset correction and propagation path characteristic correction by a propagation characteristic estimation unit (equalizer) are performed from the training pattern signal to improve reception performance.
Taking an OFDM wireless LAN as an example, in general, a synchronization processing unit that performs packet detection and symbol synchronization for each packet, a Fourier transform processing unit that performs Fourier transform processing for each symbol period in the synchronized packet, A propagation characteristic estimation unit that estimates a different propagation characteristic for each packet using a training signal of an OFDM signal that has been subjected to synchronization processing, and a frequency domain after Fourier transform processing using a propagation characteristic estimation result by the propagation characteristic estimation unit What is provided with the equalizer which performs an equalization process with respect to this signal is known. Note that the equalizer corrects the amplitude and phase of the signal in the frequency domain. This corrects the frequency characteristics due to the influence of frequency selective fading and the like. Note that the equalized signal is demodulated by a downstream modulator, and restored to transmission source data by performing demapping, deinterleaving, and error correction.

It is said that communication performance and multipath fading environment determine reception performance in the case of wireless communication. Since the communication distance corresponds to the received power, reception becomes inevitably difficult as the distance increases. On the other hand, when affected by multipath fading, not only the received power but also its amplitude and phase fluctuate drastically.
In particular, in the case of multicarrier transmission such as the OFDM method, the influence is different for each subcarrier due to selective fading. For this reason, the equalizer according to the prior art performs equalization processing using the propagation characteristic estimated by the propagation characteristic estimation unit, and corrects the variation in the amplitude and phase of the OFDM signal due to fading.

However, after the synchronization is established by the synchronization processing unit, the propagation characteristic estimation unit uses a known preamble signal of the OFDM signal (a long training symbol in the case of the IEEE802.11a standard) to depend on the propagation path. Since the characteristics are estimated, propagation path correction cannot be performed before synchronization is established, that is, equalization processing cannot be performed before the synchronization processing.
The synchronization processing unit performs detection and synchronization processing of each packet by cross-correlation calculation using a known pattern signal provided in the preamble of each packet. When the correlation power value becomes smaller, for example, when it becomes smaller than a threshold value set for performing packet detection, there is a problem that packets cannot be detected. This results in packet loss. In such a case, for example, in the case of a wireless LAN, the packet may be recovered by retransmission of the packet. However, if the packet loss increases, the retransmission frequently occurs, resulting in a decrease in effective transmission performance.

As a prior art proposed for solving the above problem, an OFDM demodulator described in Patent Document 1 can be shown. In this prior art, the pre-equalization is performed before the OFDM signal input from the A / D converter is input to the synchronization processing unit using the average transfer function obtained by averaging the transfer function previously stopped by the propagation path estimation unit. Pre-equalization is performed by the apparatus. That is, a pre-equalization device is provided separately from the equalizer provided on the downstream side of the synchronization processing unit, and the provision of the pre-equalization device increases the signal level of the OFDM signal before the synchronization processing unit. It is said that this problem can be solved by increasing the success probability of signal detection by.
Note that the pre-equalization apparatus used in this conventional technique is adapted to convert the frequency domain propagation path characteristic normally obtained by the OFDM system propagation characteristic estimation unit in the time domain by inverse FFT (Fast Fourier Transform) or the like. Preliminary equalization processing is realized by applying a transversal filter process using the transfer function as a coefficient to the received OFDM signal subjected to A / D conversion, using a transfer path transfer function.

However, in the OFDM demodulator described in Patent Document 1 shown as the prior art, a pre-equalizer is provided separately from the equalizer provided on the rear stage side of the synchronization processing unit, so that the received OFDM signal can be processed. Perform preliminary equalization. That is, the baseband signal of the OFDM signal must be A / D converted and the obtained sample signal must be filtered. This processing requires a circuit of multipliers and adders for the number of filter TAPs, and the circuit scale increases by providing a preliminary equalization device. Moreover, in order to perform preliminary equalization processing on the received OFDM signal, the preliminary equalization apparatus must be operated at all times, resulting in an increase in power consumption.
The present invention has been made in view of the above-described problems of the prior art in an OFDM communication apparatus that wirelessly transmits data, and its object is to reduce power consumption without increasing the circuit scale. The object is to improve the performance in packet detection and symbol synchronization processing performed on the receiving side of the OFDM signal.

The present invention is a wireless communication apparatus configured by combining a symbol unit of a predetermined length and transmitting / receiving a packet having a training pattern signal in a preamble using an orthogonal frequency division multiplexing signal, and a preamble signal of the received packet and training A synchronization processing means that detects a packet and generates a synchronization timing signal with a symbol, and a propagation characteristic that estimates a propagation characteristic of a radio wave for each packet based on a preamble reception signal A wireless communication apparatus having an estimation means, and a training pattern storage means for storing a predetermined pattern signal common to a pattern signal used for a preamble as a training pattern signal used for processing by the synchronization processing means; Same as the received signal of the current preamble When the synchronization is established with respect to the preamble reception signal, the propagation characteristic estimation result obtained by the propagation characteristic estimation means when the synchronization is established previously is stored in the training pattern storage means. Pattern signal correcting means for correcting the pattern signal, wherein the synchronization processing means correlates the received signal of the current preamble and the training pattern signal corrected by the pattern signal correcting means. .
The present invention relates to a preamble signal of a received packet and a training pattern in wireless communication in which a packet having a training pattern signal in a preamble is transmitted / received by orthogonal frequency division multiplexing signals and configured by combining symbol units of a predetermined length. Receives data that correlates with the signal, detects the packet, performs synchronization processing that generates a synchronization timing signal with the symbol, and performs propagation characteristic estimation that estimates the propagation characteristic of the radio wave for each packet based on the received signal of the preamble In the signal processing method, as a training pattern signal used for processing in the synchronization processing, a training pattern for storing a predetermined pattern signal common to a pattern signal used for a preamble is stored, and the received signal of the current preamble is stored. If synchronization is not established In addition, the propagation characteristic estimation result obtained by the propagation characteristic estimation when synchronization is established with respect to the received signal of the preamble previously, a predetermined common signal with the pattern signal used for the preamble stored as the training pattern Pattern signal correction is performed to correct the pattern signal, and the synchronization processing correlates the received signal of the current preamble and the training pattern signal corrected by the pattern signal correction.

  According to the present invention, it is possible to improve performance in packet detection and symbol synchronization processing performed on the receiving side of an OFDM signal without significantly reducing power consumption and increasing the circuit scale as compared with the prior art. it can.

It is a figure which shows the structure of the reception process part of the radio | wireless communication apparatus used as the application object of this invention. It is a figure which shows the basic composition of the reception process part of the radio | wireless communication apparatus which concerns on embodiment of this invention. It is a figure which shows the internal structure of the pattern signal correction | amendment part of the reception process part shown in FIG. It is a figure which shows the waveform of the input-output signal of each part of the pattern signal correction | amendment part shown in FIG. It is a figure which shows the structure which added the means corresponding to a communication mode in the reception process part (FIG. 2) of the radio | wireless communication apparatus which concerns on embodiment of this invention. It is a figure which shows the internal structure of the pattern signal correction | amendment part of the reception process part shown in FIG. It is a figure which shows the flow of the process which concerns on the pattern signal correction | amendment in the reception process part shown in FIG. FIG. 7 is a diagram showing a configuration in which means for managing a correction expiration date of a pattern signal is added to the pattern signal correction unit shown in FIG. 6. It is a figure which shows the flow of the process which concerns on the pattern signal correction | amendment in the reception process part shown in FIG. FIG. 9 is a diagram illustrating a configuration in which a propagation characteristic averaging processing unit used for pattern signal correction is added to the pattern signal correction unit illustrated in FIG. 8. It is a figure which shows the flow of the process which concerns on the pattern signal correction | amendment in the reception process part shown in FIG.

Embodiments of a wireless communication apparatus according to the present invention will be described with reference to the accompanying drawings.
[Assumed reception processor configuration]
First, the configuration of the wireless communication apparatus assumed by the present invention will be described based on the example shown in FIG. FIG. 1 shows a configuration of a reception processing unit of a wireless communication apparatus before adding means peculiar to the present invention for solving the problems of the invention (see [Problems to be Solved by the Invention] above). The same as the configuration of In addition, FIG. 1 mainly describes the reception processing unit, and therefore, for example, other known techniques for configuring a wireless communication device such as a device configuration on the transmission side and an RF (Radio Frequency) analog block on the reception side are adopted. Illustrations of the elements to be obtained are omitted.
In the following embodiments of the present invention, an example based on IEEE802.11a, which is a wireless LAN related standard, will be described.

In the reception processing unit of the wireless communication apparatus of FIG. 1, a received signal converted into a baseband signal by an RF (Radio Frequency) unit (not shown) is received as an ADC (analog / digital converter). 11 is converted to a digital signal. The digitized received signal is input to an AGC (Automatic Gain Control) unit 12, a synchronization processing unit 17, and a propagation characteristic estimation unit 18, respectively.
When the packet signal is input, the synchronization processing unit 17 performs a cross-correlation process with the training pattern signal 19 on the signal of the short preamble (the long preamble will be described later) arranged at the head of the packet. The correlation power value is calculated by performing the above process, and the packet is detected by determining whether or not the calculated value exceeds a predetermined threshold, and the process of the AGC unit 12 is performed based on the detection result. Get the control amount you need.
The training pattern signal 19 is prepared as a time domain pattern signal predetermined for training. In addition, in the short preamble of the received signal, the pattern signal for training 10 times is inserted in the field for 2 symbols at the transmission source based on the common training pattern signal.

The AGC unit 12 adjusts the received signal to a desired power level. The processing of the AGC unit 12 may be power-controlled by digital processing, or may be feedback-controlled to a VGA (variable gain amplifier) or the like of the RF analog unit for analog gain control.
The synchronization processing unit 17 operates in parallel with the processing of the AGC unit 12 and generates a symbol synchronization timing signal simultaneously with detection of a packet by pattern signal cross-correlation processing. Thereafter, the generated symbol synchronization timing signal is In synchronism with this signal, it is used to execute subsequent processing after the FFT unit 13.
The FFT unit 13 performs a Fourier transform process on the received signal adjusted to a desired power level for each symbol period to convert a time domain signal into a frequency domain signal. Note that the processing of the FFT unit 13 is a known technique and is not directly related to the problem of this embodiment, and thus detailed description thereof is omitted.

Further, based on a long training pattern signal (not shown) for a long preamble signal arranged next to the short preamble, the propagation characteristic estimation unit 18 estimates the frequency offset and propagation characteristic of each subcarrier.
The obtained frequency offset and propagation characteristic estimation results of each subcarrier are used to correct each subcarrier by the equalizer 14 for the signal after FFT processing in symbol units, and equalized to a predetermined signal level. Processing is performed.
Next, demodulation processing such as demapping, deinterleaving, and error correction processing is performed by the post-stage demodulator 15, and the demodulated data is output to a MAC (Media Access Control) 16.
The MAC 16 performs transmission control according to access control by CSMA / CA (Carrier Sense Multiple Access with Collision Avoidance) based on a wireless LAN standard (here, IEEE802.11a).

[Basic configuration for pattern signal correction]
The reception processing unit described above with reference to FIG. 1 is a technique premised on the present invention and does not perform an equalization process before the synchronization process. Therefore, the conventional technique described in [Background Art] above. In this state, the effective transmission performance is low.
In addition, Patent Document 1 proposed to deal with the problem of a decrease in effective transmission performance, as previously described in [Problems to be Solved by the Invention], increases the circuit scale by providing a preliminary equalization device. Increase and power consumption increase.

Therefore, in this embodiment, as a means for enabling reduction in power consumption without causing an increase in circuit scale, when synchronization is not established with respect to a received signal of the current preamble, synchronization processing is performed. A method is used in which a predetermined pattern signal is corrected by a propagation characteristic obtained as an estimation result when synchronization is established with respect to a preamble reception signal before.
In other words, at the stage of the synchronization process before the equalization process is performed, the propagation characteristics used for the equalization process are corrected by applying correction to the pattern signal used for the synchronization process. As a result, the synchronization processing performance is improved, and as a result, the effective transmission performance is improved.

FIG. 2 is a diagram illustrating a basic configuration of the reception processing unit of the wireless communication apparatus according to the embodiment of the present invention.
The reception processing unit of the wireless communication apparatus in FIG. 2 differs from the reception processing unit in FIG. 1 in that a pattern signal correction unit 20 is added as an additional configuration. That is, in the reception processing unit of FIG. 1, the pattern signal 19 prepared for training is used as a fixed value as it is, but in this embodiment, a pattern signal correction unit 20 is provided, and a propagation path is added to the prepared training pattern signal. The correlation power value is calculated by the matched filter processing of the synchronization processing unit 17 based on the corrected pattern signal. In the reception processing unit of FIG. 2, the processing units other than the added pattern signal correction unit 20 are basically the same as those of the reception processing unit of FIG. The description of these processing units is omitted here with reference to the above description.

FIG. 3 is a diagram showing an internal configuration of the pattern signal correction unit 20 of the reception processing unit shown in FIG. As shown in the figure, the pattern signal correction unit 20 is provided with a pattern signal 205 in a predetermined frequency domain for training, and this pattern signal is obtained from the propagation characteristic signal in the frequency domain obtained from the received signal of the preamble. The correction processing unit 201 to be corrected and the conversion unit 202 that converts the training pattern signal in the frequency domain corrected by the correction processing unit 201 into the training pattern signal in the time domain are used as components.
FIG. 4 is a diagram showing the waveforms of input / output signals of each part of the pattern signal correction unit 20.
Referring to FIG. 4, the training pattern signal correction process will be described in more detail. The pattern signal correction unit 20 transmits the propagation characteristic signal in the frequency domain (FIG. 4B) used in the equalizer 14 for the equalization process. ) And reference), and the correction processing unit 201 corrects the training pattern signal 205 (see FIG. 4A) in the initially set frequency domain with the propagation characteristic signal.

When the radio wave propagates normally, the propagation characteristic signal in the frequency domain is originally the characteristic shown in FIG. 4A in which the power of each subcarrier on the frequency axis is constant and the envelope is flat. As shown by a broken line L in FIG. 4B, the power of a part of the subcarrier is reduced and does not become flat. In spite of such a situation, if the synchronization process is performed using a training pattern signal having a flat characteristic on the assumption that radio waves propagate normally, the performance deteriorates.
For this reason, the propagation characteristic signal in the frequency domain used for the equalization processing by the equalizer 14 is regarded as a propagation characteristic signal indicating the actual state of the propagation path, and this signal is used to initialize the frequency characteristic in the frequency domain that is initially set. The training pattern signal 205 is corrected.
The propagation characteristic signal input to the pattern signal correction unit 20 is obtained for each packet, and is a propagation characteristic signal in the frequency domain used for equalization processing by the equalizer 14. The propagation characteristics obtained and stored in the reception process are acquired from the equalizer 14.

The training pattern signal 205 to be corrected is a signal having a flat envelope characteristic shown in FIG. 4A, and is a pattern signal common to a pattern signal previously determined for training and inserted into the short preamble. have. However, even if the pattern signal is originally the same, the pattern signal used for the correction processing by the pattern signal correction unit 20 is the training pattern signal 205 in the frequency domain, and the time domain pattern signal inserted in the short preamble of the received signal. And the area is different.
Therefore, the pattern signal used for the correction processing in the pattern signal correction unit 20 is a signal obtained by converting the time domain pattern signal into the frequency domain by FFT conversion or the like, and storing the data in the frequency domain. This is used as a training pattern signal 205.

In the correction processing of the training pattern signal 205 in the frequency domain, the training pattern signal 205 to be corrected is P (F), and the propagation characteristic signal used for the equalization processing by the equalizer 14 is H (F). The correction pattern signal RP (H) in the frequency domain is calculated based on the following equation (1).
RP (H) = P (F) × H (F) (1)
The correction processing unit 201 calculates a correction pattern signal RP (H) in the frequency domain based on the above equation (1). According to the waveform example in FIG. 4, when the training pattern signal 205 to be corrected in FIG. 4A is corrected by the propagation characteristic signal in FIG. 4B, the correction pattern signal in the frequency domain in FIG. RP (H) is obtained.
Also, the corrected training pattern signal in the frequency domain in FIG. 4C is converted by the time domain conversion unit 202, and the training pattern in the time domain used by the synchronization processing unit 17 in FIG. Output as a signal. The training pattern signal in the time domain of FIG. 4 (d ₀ ) in FIG. 4 is a pattern signal when correction is not performed (corresponding to the training pattern signal 205 in the frequency domain), and will be described later. However, this output is used depending on the communication mode.

According to the reception processing unit of the present embodiment described with reference to FIGS. 2 to 4, the training pattern signal 205 in the initially set frequency domain is corrected by the propagation characteristics obtained by the packet reception performed previously. The pattern signal correction unit is activated only after receiving a packet, and a training pattern signal used for synchronization processing for a standby packet may be generated. For the OFDM signal received in Patent Document 1 shown above, always, a spare, etc. Compared with the method of operating the digitizing device, the power consumption is overwhelmingly small.
In addition, since the pattern signal correction unit 20 of the present embodiment can share the FFT circuit, the equalizer, and the like that are already included in the reception processing unit, a new circuit is provided as in the preliminary equalization device of Patent Document 1. It is possible to suppress the increase in the circuit scale and improve the performance in packet detection and symbol synchronization processing.

[Support for communication mode]
The reception processing unit described above with reference to FIGS. 2 to 4 has a basic configuration related to pattern signal correction, and inconvenience may occur depending on the communication mode.
Here, an embodiment having means for avoiding inconveniences that may occur due to the communication mode will be described by taking communication in the infrastructure mode of the wireless LAN as an example.
In the wireless LAN infrastructure mode, the connection partner of the local station is an access point (AP), but radio waves from other stations (ST) connected to the AP can also be received. When the ST of another station transmits data to the AP, it is also necessary for the local station to receive the signal, demodulate it, and deliver the data to MAC (Media Access Control).
This is not a method of making transmission / reception timing by time division such as TDMA (Time Division Multiple Access), but when a wireless LAN of CSMA / CA (Carrier Sense Multiple Access with Collision Avoidance) method is used, packets are transmitted to itself. This is because it is necessary to keep the receiver on standby at all times. Even in the power save mode, it is necessary to always keep the receiver on standby for a certain period.

In the wireless LAN standard, the data output from the physical layer is processed in the physical layer of the modem unit without determining whether the packet is addressed to the local station. For this reason, it is necessary to receive a packet that is not addressed to the own station, and to determine whether or not the packet is addressed to the own station on the subsequent MAC side.
When a packet that is not addressed to the own station is received, the propagation characteristic of the propagation path is estimated by reception processing, but the propagation characteristic of the propagation path obtained there is not the propagation characteristic of the propagation path of the AP and the own station. Therefore, if the pattern signal is corrected by the pattern signal correction unit 20 as in the above-described embodiment (FIGS. 2 to 4) using the propagation characteristics of the propagation path, packets that should be received from the expected connection partner can be lost. Inconvenience that the performance increases.
Therefore, after judging whether the received packet is addressed to the own station by MAC, based on the judgment result, the propagation characteristic estimation unit obtains the propagation characteristic of the propagation path used in the equalizer by the pattern signal correction unit. It is determined whether the pattern signal is used for correction, and correction is performed only when addressed to the own station, so that correction is not performed with the propagation characteristics of the wrong propagation path.

FIG. 5 is a diagram showing the configuration of the reception processing unit of the wireless communication apparatus according to this embodiment.
The reception processing unit of the wireless communication apparatus in FIG. 5 is different from the reception processing unit in FIG. 2 in that the correction operation in the pattern signal correction unit is performed by the local station packet determination signal from the MAC (the received packet is a signal addressed to the local station). This is different in that a configuration to be controlled by a signal) is added. That is, in this embodiment, the local station packet determination signal is input to the pattern signal correction unit 21 as a control signal from the MAC 16 that is not included in the reception processing unit of FIG. The configuration for controlling the correction operation by the local station packet determination signal is provided (details will be described later in FIGS. 6 and 7). In the reception processing unit of FIG. 5, processing units other than the above-described configuration of the added pattern signal correction unit 21 are basically the same as those of the reception processing unit of FIGS. 1 and 2, and the same elements are denoted by the same reference numerals. Therefore, the description of these processing units is omitted here with reference to the above description.

  FIG. 6 is a diagram showing an internal configuration of the pattern signal correction unit 21 of the reception processing unit shown in FIG. As shown in the figure, the pattern signal correction unit 21 prepares a pattern signal 215 in a predetermined frequency domain for training, and this pattern signal is estimated from the received signal of the preamble in the frequency domain. Alternatively, the correction processing unit 211 that corrects by the propagation characteristic signal 216 based on the Dirac (Δ) function, and the conversion unit that converts the training pattern signal in the frequency domain corrected by the correction processing unit 211 into the training pattern signal in the time domain. 212. Whether the correction processing unit 211 uses the propagation characteristic signal in the frequency domain estimated from the received signal of the preamble used for the correction process or the propagation characteristic signal 216 based on the Dirac (Δ) function is used to determine the packet from the MAC 16 Since the selection is based on the signal, a selector 213 is provided that selects and outputs one of these propagation characteristic signals using the local packet determination signal as a control signal.

<Flow of propagation characteristic signal selection process (1)>
As a propagation characteristic signal used for correction of the training pattern signal in the correction processing unit 211 of the pattern signal correction unit 21 in FIG. 6, the propagation characteristic signal in the frequency domain estimated from the preamble reception signal or the propagation characteristic by the Dirac (Δ) function A processing procedure for selecting and outputting one of the signals 216 will be described based on the processing flow of FIG.
The processing flow shown in FIG. 7 is started together with the reception processing unit, and executes processing for each packet.
When this processing flow is started together with the reception processing unit, first, it waits for receiving a packet (step S101), and when a packet is received, as described above with reference to FIG. AGC and synchronization processing are performed based on the preamble signal. Furthermore, the process of estimating the propagation characteristics based on the long preamble signal is performed, the equalization process is performed under the obtained equalization conditions, and the demodulated process is performed on the signal after the equalization process. Restoration is performed (step S102).

Further, the propagation characteristic used in the equalization process in the preceding stage of the demodulation process of the received data, that is, the propagation characteristic in the frequency domain estimated from the received signal of the long preamble is determined as the propagation characteristic of the propagation path of the current received packet ( Step S103).
Next, the MAC 16 that receives the demodulated reception data determines whether or not the packet is addressed to the local station from the current reception data by the packet determination unit 161 addressed to the local station in the MAC 16 and branches the process depending on the determination result. (Step S104). At this time, if the mode setting unit 163 in the MAC 16 is a packet addressed to the own station, the mode setting unit 163 generates the own station reception flag signal and outputs the received signal to the pattern signal correction unit 21 to change the own station packet determination signal into the pattern. A control signal used for selection of propagation characteristics performed by the signal correction unit 21 is used.

If it is confirmed in step S104 that the packet is addressed to the local station by the local station packet determination signal (step S104-YES), the pattern signal correction unit 21 obtains the propagation characteristic obtained from the current received packet determined in step S103. The propagation characteristic used for correction of the training pattern signal is selected (step S105).
On the other hand, if it cannot be confirmed that the packet is destined for the own station (step S104-NO), the pattern signal correction unit 21 performs the Dirac (Δ) function (propagation having a flat characteristic in the frequency domain) that is an initialization propagation characteristic. The propagation characteristic signal 216 based on the function that defines the characteristic is selected as the propagation characteristic signal used for correcting the training pattern signal (step S107).
In this embodiment, the selection operation of the propagation characteristic signal is performed by the selector 213 that operates using the local station reception flag signal output from the MAC 16 as a control signal, and the frequency domain propagation characteristic signal estimated from the preamble reception signal. Alternatively, the propagation characteristic signal 216 by the Dirac (Δ) function is selected.
Next, the propagation characteristic signal selected in step S105 or step S107 is output to the correction processing unit 211 as a training pattern signal in the frequency domain for correction (step S106), and the processing flow for this packet is finished. The process returns to step S101 for waiting to receive a packet.

By the above processing, the pattern signal correction unit 21 corrects the pattern signal using the propagation characteristic signal in the frequency domain estimated from the preamble reception signal only when addressed to the own station. In this way, for example, in the operation of the wireless LAN infrastructure mode, only the characteristics of the propagation path addressed to the own station can be reflected, so that the accuracy of the synchronization processing such as the packet detection accuracy can be maintained.
In the wireless LAN standard, there is an ad hoc mode other than the infrastructure mode described above. When this ad hoc mode is used, there are a plurality of connection partner terminals to be communicated. Therefore, the correlation detection is performed with the correction pattern signal obtained by specifying the propagation path in the packet detection waiting state and performing the propagation path correction. I can't. Therefore, when operating in the ad hoc mode, the synchronization process is performed by returning to the propagation characteristic signal based on the Dirac (Δ) function, which is the initialization propagation characteristic.

[Response to changes over time]
The reception processing unit described above with reference to FIGS. 2 to 7 does not take into account fluctuations in propagation characteristics that occur over time in the propagation path. For this reason, even if the pattern signal is corrected based on the propagation characteristics obtained in the reception processing of the immediately preceding packet, a considerable time may elapse while the corrected pattern signal is applied to the next received packet. Depending on the elapsed time, the propagation characteristics may vary, resulting in a disadvantage that the correction effect is lost.
Therefore, in this embodiment, the synchronization probability is determined by the received signal of the packet addressed to the own station immediately before, and the elapsed time from the time when the propagation characteristic of the propagation path used in the equalizer is obtained is measured. The expiration date of the pattern signal is managed.

  In this embodiment, a reception processing unit that performs an operation corresponding to the infrastructure mode of the wireless LAN described above with reference to FIGS. 5 to 7 is taken as an example, and means for avoiding inconveniences that may occur due to changes over time An embodiment having Specifically, the pattern signal correction unit is provided with a timer that receives a local station packet determination signal from the MAC and starts measuring a predetermined elapsed time. When the time counted by the timer is within the predetermined elapsed time, When a packet signal addressed to the station is received, the pattern signal is corrected by the propagation characteristic signal in the frequency domain estimated by the packet reception process, and the signal used for the synchronization process is updated. When the elapsed time elapses, the synchronization process is performed by returning to the propagation characteristic signal by the Dirac (Δ) function which is the initialization propagation characteristic.

FIG. 8 is a diagram showing an internal configuration of the pattern signal correction unit 21 of the reception processing unit according to this embodiment. Note that the overall configuration of the reception processing unit of the wireless communication apparatus according to this embodiment is the same as that of FIG. 5 described above, so the description is omitted here and the description is omitted here.
As shown in FIG. 8, the pattern signal correction unit 21 prepares a pattern signal 215 in a predetermined frequency domain for training, and this pattern signal is estimated from the received signal of the preamble in the frequency domain. Alternatively, the correction processing unit 211 that corrects by the propagation characteristic signal 216 based on the Dirac (Δ) function, and the conversion unit that converts the training pattern signal in the frequency domain corrected by the correction processing unit 211 into the training pattern signal in the time domain. 212. Whether the correction processing unit 211 uses the propagation characteristic signal in the frequency domain estimated from the received signal of the preamble used for the correction process or the propagation characteristic signal 216 based on the Dirac (Δ) function is used to determine the packet of the local station from the MAC 16. Since the selection is based on the signal, a selector 213 is provided that selects and outputs one of these propagation characteristic signals using the local packet determination signal as a control signal.
The pattern signal correction unit 21 is the same as the pattern signal correction unit 21 shown in FIG. However, this embodiment is different in configuration in that it includes a timer 218 for managing the expiration date of the corrected pattern signal. The self-station packet determination signal from the MAC 16 is not directly input to the selector 213, but the passage of a predetermined time set in the timer 218 is measured through the timer 218, and the output timer signal is input to the selector 213 as a control signal. .

<Flow of propagation characteristic signal selection process (2)>
The propagation characteristic signal or Dirac (Δ) in the frequency domain estimated from the received signal of the preamble as the propagation characteristic signal used for correcting the training pattern signal in the correction processing part 211 in the frequency domain of the pattern signal correction unit 21 in FIG. A procedure for selecting and outputting one of the propagation characteristic signals 216 by function will be described based on the processing flow of FIG.
The processing flow shown in FIG. 9 is started together with the reception processing unit, and executes processing for each packet.
When this processing flow is started together with the reception processing unit, first, it waits to receive a packet (step S201), and when a packet is received, as described above with reference to FIG. AGC and synchronization processing are performed based on the preamble signal. Furthermore, the process of estimating the propagation characteristics based on the long preamble signal is performed, the equalization process is performed under the obtained equalization conditions, and the demodulated process is performed on the signal after the equalization process. Restoration is performed (step S202).

Further, the propagation characteristic used in the equalization process in the preceding stage of the demodulation process of the received data, that is, the propagation characteristic in the frequency domain estimated from the received signal of the long preamble is determined as the propagation characteristic of the propagation path of the current received packet ( Step S203).
Next, the MAC 16 that receives the demodulated reception data determines whether or not the packet is addressed to the local station from the current reception data by the packet determination unit 161 addressed to the local station in the MAC 16 and branches the process depending on the determination result. (Step S204). At this time, if the mode setting unit 163 in the MAC 16 is a packet addressed to the own station, the mode setting unit 163 generates the own station reception flag signal and outputs the received signal to the pattern signal correction unit 21 to change the own station packet determination signal into the pattern. The signal correction unit 21 is notified.

In step S204, when it is confirmed by the determination of the local station packet that the packet is addressed to the local station (step S204-YES), the pattern signal correction unit 21 receives the local station packet determination signal from the MAC 16, The timer 218 that has set a predetermined time is started and time measurement is started (step S205).
Next, the propagation characteristic obtained from the current received packet determined in step S203 is selected as the propagation characteristic used for correction of the training pattern signal by the correction processing unit 211 in the frequency domain (step S207).
On the other hand, if it is not confirmed in step S204 that the packet is addressed to the local station by the local station packet determination signal (step S204-NO), the predetermined time set in the timer 218 started in step S205, Then, the passage of a predetermined time as the expiration date of the corrected training pattern signal is confirmed (step S206).

If it is determined in step S206 that the predetermined time has not elapsed (step S206-NO), the process returns to step S201 that waits to receive the next packet because it is within the expiration date of the training pattern signal. In this case, the pattern signal correction unit 21 does not perform any processing.
If the predetermined time has passed in step S206 (step S206-YES), the expiration date of the training pattern signal has passed, so the pattern signal correction unit 21 performs Dirac (initialization propagation characteristics). A propagation characteristic signal 216 based on a Δ) function (a function defining a propagation characteristic having a flat characteristic in the frequency domain) is selected as a propagation characteristic signal used for correction of the training pattern signal (step S208).

In this embodiment, the propagation characteristic signal selection operation controls the timer signal that is output when the time set in the timer 218 that starts timing in response to the local packet determination signal output from the MAC 16 has elapsed. A selector 213 operating as a signal selects a propagation characteristic signal 216 based on a frequency domain propagation characteristic signal or a Dirac (Δ) function estimated from a preamble reception signal.
Next, the propagation characteristic signal selected in step S207 or step S208 is output to the correction processing unit 211 as a training pattern signal in the frequency domain for correction (step S209), and the processing flow for this packet is finished. The process returns to step S201 for waiting to receive a packet.

  By performing the above processing, the pattern signal correcting unit 21 corrects the pattern signal using the propagation characteristic signal in the frequency domain estimated from the received signal of the preamble only when addressed to the own station, and further corrects the corrected training signal. Confirm the passage of time determined in advance as the expiration date of the pattern signal, apply the propagation characteristic signal in the frequency domain estimated from the preamble reception signal within the expiration date, and also apply the propagation characteristic signal beyond the expiration date. When a change is assumed, by applying a propagation characteristic signal based on a Dirac (Δ) function, it is possible to avoid a deterioration in the packet detection rate due to a change in propagation characteristic over time.

[Add averaging]
The reception processing unit described above with reference to FIGS. 2 to 9 uses the propagation characteristics obtained in the packet reception process to correct the training pattern signal in consideration of the propagation characteristic fluctuations that occur in the propagation path. The expiration date was managed. However, the propagation characteristic used for correction is the propagation characteristic obtained at the time of reception of the packet addressed to the immediately preceding own station, for example, when receiving the influence of an abnormality that occurred accidentally in the propagation path at the time of reception, If the propagation path returns to normal when the current packet is received, it is not appropriate to correct the current received signal based on propagation characteristics when an abnormality occurs.
Therefore, in this embodiment, the propagation characteristics previously used for correction are stored, and the propagation characteristics obtained at the time of reception of the packet addressed to the local station immediately before are used for correction of the training pattern signal. Instead of using the propagation characteristics as they are, the averaged process with the previously obtained propagation characteristics is performed, and the obtained propagation characteristics are used for correction, thereby reducing the influence of accidental abnormalities.

In this embodiment, the reception processing unit that corresponds to the wireless LAN infrastructure mode described above with reference to FIGS. An embodiment having averaging processing means for reducing the influence of an abnormality occurring in the above will be described.
Specifically, an averaging processing unit 219 that is activated by a local station packet determination signal input from the MAC 16 is provided in addition to the timer 218 described above with reference to FIGS. The averaging processing unit 219 receives the input of the propagation characteristic signal in the frequency domain estimated from the preamble reception signal used in the equalizer 14, and at the timing when the timer 218 starts measuring time based on the local packet determination signal, An averaging process is performed based on the propagation characteristic signal obtained in step (b).

FIG. 10 is a diagram showing an internal configuration of the pattern signal correction unit 21 of the reception processing unit according to this embodiment. Note that the overall configuration of the reception processing unit of the wireless communication apparatus according to this embodiment is the same as that of FIG. 5 described above, so the description is omitted here and the description is omitted here.
As shown in FIG. 10, the pattern signal correction unit 21 prepares a pattern signal 215 in a predetermined frequency domain for training, and this pattern signal is a propagation characteristic signal in the frequency domain estimated from the received signal of the preamble. Alternatively, the correction processing unit 211 in the frequency domain corrected by the propagation characteristic signal 216 using the Dirac (Δ) function, and the training pattern signal in the frequency domain corrected by the correction processing unit 211 are used as the training pattern signal in the time domain. A conversion unit 212 for conversion is provided. Whether the correction processing unit 211 uses the propagation characteristic signal in the frequency domain estimated from the received signal of the preamble used for the correction process or the propagation characteristic signal 216 based on the Dirac (Δ) function is used to determine the packet of the local station from the MAC 16. Since the selection is based on the signal, a selector 213 is provided that selects and outputs one of these propagation characteristic signals using the local packet determination signal as a control signal.

In addition, a timer 218 for managing the expiration date of the corrected pattern signal is provided. The local station packet determination signal from the MAC 16 is input as a start signal for the timer 218, and the elapse of a predetermined time set in the timer 218 is measured. The output timer signal is input to the selector 213 as a control signal.
The pattern signal correction unit 21 is the same as the pattern signal correction unit 21 shown in FIG. However, in this embodiment, after the frequency domain propagation characteristic signal estimated from the preamble reception signal used in the equalizer 14 is not directly input to the selector 213, the averaging process is performed through the averaging processing unit 219. Is transmitted as a signal selected by the selector 213.

In addition, the averaging processing unit 219 in this embodiment uses the propagation characteristic signal input from the equalizer 14 at the timing when the timer 218 starts measurement as the propagation characteristic signal obtained this time. Get a signal. In order to perform the averaging process, the previously determined propagation characteristic signal is required, and therefore the averaging processing unit 219 includes storage means (not shown) for storing the signal.
In addition, the averaging process preferably employs a method of averaging back several packets from immediately before, and can be performed by, for example, a moving average process. According to such an averaging process, it is possible to avoid a decrease in accuracy by suppressing the influence of temporal variation.
Furthermore, even when the above-described method of averaging several packets from the immediately preceding time is adopted, the propagation characteristic signal obtained before use in the averaging process was obtained before a predetermined time in consideration of the temporal variation. Higher accuracy can be maintained by adopting an averaging method that excludes propagation path estimation results. For this purpose, it is necessary to manage the storage time by a method such as stamping the time for the propagation characteristic signal stored in the storage means for use in the averaging process. For example, it can be implemented by a management method for deleting a propagation characteristic signal whose storage time exceeds a predetermined time from among the stored propagation characteristic signals.

<Flow of propagation characteristic signal selection process (3)>
The propagation characteristic signal or Dirac (Δ) in the frequency domain estimated from the received signal of the preamble as the propagation characteristic signal used for correcting the training pattern signal in the correction processing part 211 in the frequency domain of the pattern signal correction unit 21 in FIG. A procedure for selecting and outputting one of the propagation characteristic signals 216 by function will be described based on the processing flow of FIG.
The processing flow shown in FIG. 11 is started together with the reception processing unit, and executes processing for each packet.
When this processing flow is started together with the reception processing unit, first, it waits to receive a packet (step S301). When a packet is received, as described above with reference to FIG. AGC and synchronization processing are performed based on the above signal. Furthermore, the process of estimating the propagation characteristics based on the long preamble signal is performed, the equalization process is performed under the obtained equalization conditions, and the demodulated process is performed on the signal after the equalization process. Restoration is performed (step S302).

Further, the propagation characteristic used in the equalization process in the preceding stage of the demodulation process of the received data, that is, the propagation characteristic in the frequency domain estimated from the received signal of the long preamble is determined as the propagation characteristic of the propagation path of the current received packet ( Step S303).
Next, the MAC 16 that receives the demodulated reception data determines whether or not the packet is addressed to the local station from the current reception data by the packet determination unit 161 addressed to the local station in the MAC 16 and branches the process depending on the determination result. (Step S304). At this time, if the mode setting unit 163 in the MAC 16 is a packet addressed to the own station, the mode setting unit 163 generates the own station reception flag signal and outputs the received signal to the pattern signal correction unit 21 to change the own station packet determination signal into the pattern. The signal correction unit 21 is notified.

In step S304, when it is confirmed by the determination of the local station packet that the packet is addressed to the local station (step S304-YES), the pattern signal correction unit 21 receives the local station packet determination signal from the MAC 16, The timer 218 that has set a predetermined time is started and time measurement is started (step S305).
In addition, at the timing when the timer 218 starts measurement, the start timing signal is sent from the timer 218 to the averaging processing unit 219, whereby the processing unit is started and propagation characteristic averaging processing used for correction of the training pattern signal is performed. Is performed (step S306). Here, for example, every time the averaging process is performed, the obtained propagation characteristics are stored, the propagation characteristics of the current packet and the propagation characteristics stored by going back a predetermined number of packets are moving averaged, and the current propagation characteristics are obtained. Averaging processing is performed by obtaining.

Next, the propagation characteristic obtained by performing the averaging process in step S306 is selected as the propagation characteristic used for correction of the training pattern signal by the correction processing unit 211 in the frequency domain (step S308).
On the other hand, if it is not confirmed in step S304 that the packet is addressed to the own station by the own station packet determination signal (step S304-NO), the predetermined time set in the timer 218 previously started in step S305, that is, Then, the passage of a predetermined time as the expiration date of the corrected training pattern signal is confirmed (step S307).

If it is determined in step S307 that the predetermined time has not elapsed (step S307-NO), since the training pattern signal is within the expiration date, the process returns to step S301 that waits to receive the next packet. In this case, the pattern signal correction unit 21 does not perform any processing.
If the predetermined time has passed in step S307 (step S307-YES), since the expiration date of the training pattern signal has passed, the pattern signal correction unit 21 performs Dirac (initialization propagation characteristics). A propagation characteristic signal 216 based on a Δ) function (a function defining a propagation characteristic having a flat characteristic in the frequency domain) is selected as a propagation characteristic signal used for correction of the training pattern signal (step S309).

In this embodiment, the propagation characteristic signal selection operation controls the timer signal that is output when the time set in the timer 218 that starts timing in response to the local packet determination signal output from the MAC 16 has elapsed. The selector 213 operating as a signal receives the propagation characteristic signal in the frequency domain estimated from the preamble reception signal (here, the propagation characteristic signal processed by the averaging processing unit 219) or the propagation characteristic signal 216 based on the Dirac (Δ) function. select. In other words, when there is no local station packet determination signal and the expiration date of the propagation characteristic signal estimated from the received signal of the preamble used for correction obtained by the averaging process has passed, the propagation characteristic by the Dirac (Δ) function Signal 216 is selected.
Next, the propagation characteristic signal selected in step S308 or step S309 is output to the correction processing unit 211 as a training pattern signal in the frequency domain for correction (step S310), and the processing flow for this packet is finished. The process returns to step S301 for waiting to receive a packet.

As described above, the propagation characteristics saved by going back a predetermined number of packets and the current propagation characteristics are averaged, and the resulting propagation characteristics are used for correction. Can reduce the effects of
In addition, if some of the propagation characteristics that are subject to averaging processing are propagation characteristics obtained at the time when a predetermined expiration date has passed, it can be excluded from the averaging processing. Can be maintained.

  11... ADC, 12... AGC section, 13.. FFT section, 14... Equalizer, 15... Demodulator, 16... MAC, 161. Setting unit, 17 ..Synchronization processing unit, 18 ..Propagation characteristic estimation unit, 20, 21 ..Pattern signal correction unit, 19, 205, 215... Initial setting training pattern signal in frequency domain, 201, 211. · Frequency domain correction processing unit 202, 212 · · Time domain conversion unit 216 · · Propagation characteristic signal by Dirac (Δ) function 213 · · Selector 218 · · Timer 219 · · Averaging processing Department.

JP 2004-320143 A

Claims (6)

A wireless communication device configured by combining symbol units of a predetermined length and transmitting / receiving a packet having a training pattern signal in a preamble by an orthogonal frequency division multiplexing signal,
Correlation between the preamble signal of the received packet and the training pattern signal to detect the packet and generate a synchronization timing signal with the symbol, and radio wave propagation based on the preamble reception signal A wireless communication device having propagation characteristic estimation means for estimating characteristics for each packet,
As a training pattern signal used for processing by the synchronization processing means, a training pattern storage means for storing a predetermined pattern signal common to a pattern signal used for a preamble,
When the synchronization is not established with respect to the current preamble reception signal, the training pattern is determined based on the propagation characteristic estimation result obtained by the propagation characteristic estimation means when the synchronization is established with respect to the preamble reception signal. Pattern signal correction means for correcting a predetermined pattern signal stored in the storage means;
The wireless communication apparatus, wherein the synchronization processing means correlates a received signal of a current preamble and a training pattern signal corrected by the pattern signal correcting means. The wireless communication device according to claim 1,
It has a function of performing communication in a mode in which communication is performed with one partner station determined as a communication target and a mode in which communication is performed with a plurality of partner stations as communication targets simultaneously, and transmission control with the partner station is performed in accordance with the communication mode. Media access control means for performing
The media access control means is a case where the communication mode to be used is a case where a communication mode is selected for simultaneously communicating with a plurality of partner stations, and there is one partner station, and the packet received immediately before is the local station. Determine whether the packet is addressed to
The pattern signal correcting means only reads from the packet when the media access control means determines that there is one partner station and that the packet received immediately before is a packet addressed to itself. The wireless communication apparatus, wherein the correction is performed on a predetermined pattern signal stored in the training pattern storage unit based on the propagation characteristic estimation result obtained. The wireless communication device according to claim 1 or 2,
It has a time measuring unit that measures the elapse of a predetermined time since synchronization is established with respect to the preamble reception signal immediately before,
The radio communication apparatus according to claim 1, wherein the pattern signal correction unit performs the correction while the predetermined time has not been measured by the time measuring unit, and does not perform the correction after the predetermined time has elapsed. The wireless communication apparatus according to any one of claims 1 to 3,
The pattern signal correcting means corrects the training pattern signal using an average propagation path estimation result obtained by averaging the propagation characteristic estimation result after synchronization is established with respect to the correction target packet by going back several packets from immediately before. A wireless communication device characterized by the above. The wireless communication device according to claim 4, wherein
The radio communication apparatus according to claim 1, wherein the pattern signal correcting means performs averaging by excluding a propagation path estimation result obtained before a predetermined time when calculating the average propagation path estimation result. Correlation between a preamble signal of a received packet and a training pattern signal in wireless communication in which a packet having a training pattern signal in a preamble is transmitted / received by an orthogonal frequency division multiplexing signal and configured by combining symbol units of a predetermined length Receive signal processing method that detects a packet and generates a synchronization timing signal with a symbol, and a propagation characteristic estimation that estimates a radio wave propagation characteristic for each packet based on a preamble reception signal There,
As a training pattern signal used for processing in the synchronization processing, a training pattern storage for storing a predetermined pattern signal common to a pattern signal used for a preamble,
When synchronization is not established with respect to the received signal of the current preamble, a training pattern is obtained based on the propagation characteristic estimation result obtained by the propagation characteristic estimation when synchronization is established with respect to the received signal of the preamble. Pattern signal correction is performed to correct a predetermined pattern signal common to the pattern signal used for the preamble stored as
The received signal processing method, wherein the synchronization processing correlates a received signal of a current preamble with a training pattern signal corrected by the pattern signal correction.

Images