JP4840131B2 - Reception device, reception method, learning device, learning method, and program - Google Patents

Description

  The present invention relates to a receiving device, a receiving method, a learning device, a learning method, and a program, and in particular, a receiving device, a receiving method, a learning device, and a decoding device that can accurately decode a received signal having distortion due to multipath. The present invention relates to a learning method and a program.

  For example, when wireless communication using radio waves (electromagnetic waves) is performed in a casing of a predetermined device, the receiver reflects and diffracts the transmitted radio waves by obstacles in the casing, and receives the same radio waves from multiple paths. Resulting in. In this way, the transmission of the transmitted radio wave is reflected and diffracted by an obstacle in the housing, and the radio wave from a plurality of paths is received is called multipath. Note that multipath occurs not only inside the housing, but also occurs outside the housing using, for example, buildings such as buildings or terrain as obstacles.

  In the multipath, the receiver receives radio waves of a plurality of routes having different route distances, that is, different transmission times, and thus the received waveform is distorted. Due to the distortion of the received waveform due to the multipath (hereinafter referred to as multipath fading), the code may not be decoded. In other words, there is a problem that the bit error rate increases due to a DC offset or the like generated by multipath.

  FIG. 1 shows a received waveform when radio waves radiated inside and outside a casing of a certain apparatus are received. The modulation method is an ASK (Amplitude Shift Keying) modulation method.

  In FIG. 1, the four waveforms on the right side indicate reception waveforms due to communication within the housing, and the four waveforms on the left side indicate reception waveforms due to communication outside the housing, with transmission rates of 250 kbps, 500 kbps, and 1 Mbps, respectively. , And 4 types at 2 Mbps.

  For example, when the transmission rate shown at the bottom is 2 Mbps, which shows the best features, the received waveform outside the housing shows the “0” and “1” sections clearly. On the other hand, the received waveform in the housing is likely to be judged as “1” because the reflected wave covers the section that should be “0” and the waveform collapses. In addition, the distortion of the received waveform in the housing increases as the transmission speed increases. That is, it can be seen that the reception effect varies greatly depending on the speed of the transmission signal, and the communication quality is greatly degraded by reflection.

  Therefore, there are problems that the channel capacity cannot be increased due to the influence of multipath fading, and that it is difficult to maintain arbitrary signal quality with simple signal processing.

  Thus, for example, as a means for solving multipath fading, it has been proposed to suppress reflection of electromagnetic waves on the wall surface of the housing in the housing by pasting a radio wave absorber on the entire surface of the housing (for example, Patent Documents). 1). However, the radio wave absorber is expensive and it is difficult to attach the radio wave absorber to the entire surface of the housing from the viewpoint of exhaust heat.

In addition, as a conventional multipath countermeasure by signal processing of general wireless communication,
1. 1. Use OFDM (Orthogonal Frequency Division Multiplexing) as a modulation method. 2. Use spread spectrum and rake reception. 3. Receive with multi-antenna It is conceivable to convolutionally encode the interleaved signal and perform Viterbi decoding and deinterleaving.

JP 2004-220264 A

  However, when the OFDM modulation method is used, it is necessary to perform Fourier transform at high speed, and there is a concern that high heat is generated. When spread spectrum is used, signal processing at a speed higher than that of the transmission signal is required, and it is difficult to improve the performance of the receiver.

  When a multi-antenna is used, in order to obtain the multi-antenna effect, the distance between the antennas needs to be sufficiently separated according to the wavelength, so that it is difficult to apply to communication in a narrow space. When convolutionally encoding interleaved signals and performing Viterbi decoding and deinterleaving, a large memory element is required at the time of decoding, and it is difficult to apply to communications such as video signals that require large capacity and real-time characteristics. There is.

  FIG. 2 shows a frequency distribution of the amplitude values of the baseband signal generated from the received signal when the received signal in the housing shown in FIG. 1 is received for a predetermined long time. The horizontal axis in FIG. 2 represents the amplitude value of the baseband signal, and the vertical axis represents the occurrence probability of the baseband signal taking each amplitude value.

  The distribution shown in FIG. 2 is close to the distribution that expresses multipath fading with large intersymbol interference, but is definitely different from the M distribution that statistically represents the frequency distribution of multipath amplitude values. It turns out that it is. Looking at this distribution, it is difficult to decode a received signal having distortion due to multipath generated in the housing as shown in FIG. 1 by a decoding method corresponding to normal multipath.

  The present invention has been made in view of such a situation, and makes it possible to accurately decode a received signal having distortion due to multipath.

  The receiving apparatus according to the first aspect of the present invention is a receiving apparatus that receives and decodes a signal having multipath stationarity, and recently receives the signal and divides the signal into bit time units corresponding to 1 bit. The signal is classified based on a sampling value sampled from the signal at the received current bit time and a sampling value sampled from the signal at a past bit time before the current bit time, and the classification result Classification means for determining the classification code, bit data storage means for storing a bit data table in which the classification code and bit data are associated, the bit data table, and the classification code determined by the classification means And bit data determining means for determining the bit data corresponding to the received signal.

  The bit data determined by the bit data determining means may include data of the past bit time.

  Classification code storage means for storing a classification code table in which a statistical value of a signal having multipath continuity and the classification code are associated is further provided, wherein the classification means includes the current bit time and the past bit time. The classification code can be determined by comparing the sampling value and the statistical value.

  The classification means calculates a range of a signal having multipath continuity from the statistical value, and determines whether the sampling values of the current bit time and the past bit time are included in the range. Thus, the classification code can be determined.

  Sampling means for obtaining a plurality of sampling values from the signal of one bit time can be further provided, and the classification means can determine the classification code using the plurality of sampling values at least for the current bit time. .

  The sampling means has clock signal conversion means for shifting the phase around the clock signal supplied thereto for a predetermined time, and based on the clock signal and the clock signal output from the clock signal conversion means Thus, the sampling value can be obtained.

  A receiving method according to a first aspect of the present invention is a receiving method of a receiving apparatus that receives and decodes a signal having multipath continuity, and the signal is divided into bit time units corresponding to 1 bit. Classifying the signal on the basis of a sampling value sampled from the signal of the most recently received current bit time and a sampling value sampled from the signal of a past bit time prior to the current bit time; Determining a classification code as a classification result, and determining the bit data corresponding to the received signal from the bit data table in which the classification code and bit data are associated with each other and the determined classification code including.

  The program according to the first aspect of the present invention is a program for causing a computer to execute a process of receiving and decoding a signal having multipath continuity, and dividing the signal into bit time units corresponding to 1 bit. And classifying the signal based on a sampling value sampled from the signal at the most recently received current bit time and a sampling value sampled from the signal at a past bit time prior to the current bit time. The classification code as the classification result is determined, and the bit data corresponding to the received signal is determined from the bit data table in which the classification code and bit data are associated with each other and the determined classification code Including the steps of:

  In the first aspect of the present invention, when a signal is divided into bit time units corresponding to 1 bit, a sampling value sampled from a recently received signal of the current bit time and a past before the current bit time are obtained. Based on the sampling value sampled from the bit time signal, the signal is classified and a classification code is determined. Then, the bit data corresponding to the received signal is determined from the bit data table in which the classification code and the bit data are associated with each other and the determined classification code.

  A learning device according to a second aspect of the present invention is a learning device that receives data having multipath continuity and learns data used in a receiving device that decodes the signal, and the signal corresponding to predetermined bit data Receiving means for receiving the signal, storage means for storing the signal for each predetermined classification code, statistical value calculation means for calculating the statistical value of the signal for each classification code, and the classification code corresponding to the signal And correspondence table creating means for creating a transmission / reception data correspondence table in which the bit data and the statistical value are associated with each other.

  The statistical value of the signal may be an average value and a variance value of the signal.

  A learning method according to a second aspect of the present invention corresponds to predetermined bit data in a learning method for a learning device that receives a signal having multipath stationarity and learns data used in a receiving device that decodes the signal. Receiving the signal, storing the signal for each predetermined classification code, calculating a statistical value of the signal for each classification code, the classification code corresponding to the signal, the bit data, and the A step of creating a transmission / reception data correspondence table in which the statistical values are associated with each other.

  A program according to a second aspect of the present invention is a program for causing a computer to execute a learning process for learning data used in a receiving apparatus that receives and decodes a signal having multipath continuity. Receiving the signal corresponding to the data, storing the signal for each predetermined classification code, calculating the statistical value of the signal for each classification code, the classification code corresponding to the signal, and the bit data And a step of creating a transmission / reception data correspondence table in which the statistical values are associated with each other.

  In the second aspect of the present invention, a signal corresponding to predetermined bit data is stored for each predetermined classification code, a statistical value of the signal is calculated for each classification code, a classification code corresponding to the signal, and a bit A transmission / reception data correspondence table in which data and statistical values are associated is created.

  According to the first aspect of the present invention, it is possible to accurately decode a received signal having distortion due to multipath.

  According to the second aspect of the present invention, it is possible to learn data for accurately decoding a received signal having distortion due to multipath.

  Embodiments of the present invention will be described below. Correspondences between the constituent elements of the present invention and the embodiments described in the specification or the drawings are exemplified as follows. This description is intended to confirm that the embodiments supporting the present invention are described in the specification or the drawings. Therefore, even if there is an embodiment which is described in the specification or the drawings but is not described here as an embodiment corresponding to the constituent elements of the present invention, that is not the case. It does not mean that the form does not correspond to the constituent requirements. Conversely, even if an embodiment is described here as corresponding to a configuration requirement, that means that the embodiment does not correspond to a configuration requirement other than the configuration requirement. Not something to do.

  The receiving apparatus according to the first aspect of the present invention is a receiving apparatus (for example, LSI 100 in FIG. 3) that receives and decodes a signal having multipath stationarity, and the signal is a bit time unit corresponding to 1 bit. Based on a sampling value sampled from the recently received signal at the current bit time and a sampling value sampled from the signal at a past bit time prior to the current bit time, And classifying means (for example, the classification code determining unit 361 in FIG. 11) for determining the classification code as the classification result, and bit data storing a bit data table in which the classification code and the bit data are associated with each other Storage means (for example, bit value RAM 382 in FIG. 18), the bit data table, and the classification code determined by the classification means From and a bit data determining means for determining the bit data corresponding to said signals received (e.g., received data determination unit 381 of FIG. 18).

  Further provided is a classification code storage means (for example, a distributed value RAM 362 in FIG. 11) for storing a classification code table in which a statistical value of a signal having multipath continuity and the classification code are associated with each other. The classification code can be determined by comparing the statistical values with the sampling values of the current bit time and the past bit time.

  Sampling means (for example, AD converter 301 in FIG. 7) for obtaining a plurality of sampling values from the signal of 1 bit time is further provided, and the classification means uses a plurality of sampling values at least for the current bit time. Thus, the classification code can be determined.

  The receiving method or program according to the first aspect of the present invention is a receiving method of a receiving apparatus for receiving and decoding a signal having multipath continuity or a program for causing a computer to execute the decoding process. Sampled from the signal of the most recently received current bit time, and from the signal of the past bit time prior to the current bit time, when divided into bit time units corresponding to 1 bit. Based on the sampling value, the signal is classified, a classification code as the classification result is determined (for example, step S24 in FIG. 20), a bit data table in which the classification code and bit data are associated with each other, The bit data corresponding to the received signal is determined from the determined classification code ( Eg to include a step S25) step of FIG. 20.

  A learning device according to a second aspect of the present invention is a learning device (for example, LSI 100 in FIG. 3) that receives a signal having multipath continuity and learns data used by a receiving device that decodes the signal. Receiving means (for example, the AD conversion unit 301 in FIG. 7) that receives the signal corresponding to the bit data, and storage means (for example, the received value RAM 401 in FIG. 21) that stores the signal for each predetermined classification code; , For each classification code, statistical value calculation means for calculating the statistical value of the signal (for example, average value calculation unit 402 and variance value calculation unit 403 in FIG. 21), the classification code corresponding to the signal, and Correspondence table creation means (for example, correspondence table creation unit 404 in FIG. 21) for creating a transmission / reception data correspondence table in which bit data and the statistical value are associated with each other is provided.

  A learning method or program according to a second aspect of the present invention is a learning method for a learning device that learns data used in a receiving device that receives and decodes a multipath stationary signal, or learns data. In a program that causes a computer to execute learning processing, the signal corresponding to predetermined bit data is received (for example, step S49 in FIG. 23), and the signal is stored for each predetermined classification code (for example, FIG. 23). Step S50), for each classification code, calculate a statistical value of the signal (for example, steps S51 and S52 in FIG. 23), and the classification code corresponding to the signal, the bit data, and the statistical value Is included (for example, step S53 of FIG. 23).

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 3 shows a configuration example of an embodiment of an information processing apparatus to which the present invention is applied.

  The information processing apparatus 1 includes, in the housing 2, substrates 11 to 14 and a base substrate 15 on which the substrates 11 to 14 are mounted. The substrates 11 and 12 are mounted on the base substrate 15 via the substrate fixing jig 16, and the substrates 13 and 14 are mounted directly on the base substrate 15.

  Each of the substrates 11 to 14 includes at least one LSI (Large Scale Integration) having a wireless communication function and at least one antenna. That is, the substrate 11 has an LSI 100 and an antenna 101, and the substrate 12 has an LSI 102 and an antenna 103. The substrate 13 includes an LSI 104 and an antenna 105, and an LSI 106 and an antenna 107, and the substrate 14 includes an LSI 108 and an antenna 109. The LSIs 100, 102, 104, 106, or 108 have functions as a transmission device and a reception device that use radio waves (electromagnetic waves) as transmission media via the antennas 101, 103, 105, 107, or 109, respectively.

  Note that the arrangement, installation method, and number of substrates shown in FIG. 3 are merely examples, and are not limited thereto.

  The LSI 100 transmits predetermined data such as images and sounds to other LSIs (LSIs 102, 104, 106, or 108) in the information processing apparatus 1 through the antenna 101 by wireless communication using radio waves as a transmission medium. Or data transmitted from another LSI is received via the antenna 101. The LSI 100 can communicate both burst signals and intermittent packet signals that require real-time characteristics.

  Similarly to the LSI 100, the LSIs 102, 104, 106, and 108 also perform wireless communication with other LSIs in the information processing apparatus 1. Hereinafter, a case where the LSI 100 transmits or receives data to other LSIs will be described, and a description of transmission and reception of data of other LSIs will be omitted.

  Signals received on the receiving side due to white noise (thermal noise), colored noise (noise emitted from the LSI) present in the housing 2, and signals reflected and diffracted by the wall surface or substrate in the housing 2 Waveform distortion occurs. That is, multipath distortion occurs in the waveform of the signal received by the LSI 100 in the information processing apparatus 1.

  FIG. 4 shows a transmission waveform corresponding to transmission data transmitted from another LSI (hereinafter referred to as LSI 102, for example), and a reception waveform when the LSI 100 receives the transmission waveform.

  In FIG. 4, the horizontal axis represents time, and the vertical axis represents the amplitude value of the signal waveform. The gray line represents the transmitted or received signal waveform itself (data), and the black line represents the envelope waveform (env) of the transmitted or received signal waveform. The signal modulation method is the ASK modulation method.

  As can be seen with reference to FIG. 4, when the LSI 100 receives a signal waveform corresponding to the transmission data transmitted from the LSI 102, the waveform is distorted.

  FIG. 5 shows the envelopes of the 14 transmission waveforms and the reception waveforms shown in FIG. The gray line in FIG. 5 represents each waveform (data) of 14 waves, and the black line represents a waveform (ave) obtained by averaging the waveforms of the 14 waves.

  As can be seen from FIG. 5, the degradation of the received waveform due to multipath occurring in the housing 2 is substantially constant with respect to time, and the time that the multipath wave affects the received waveform is short. Therefore, in a multipath environment in the housing 2, it can be said that the received waveform received by the LSI 100 is affected in a short time and can be regarded as constant with respect to time. As described above, a reception signal that is affected by being received in a multipath environment in a short time and can be regarded as being constant with respect to time is referred to as a stationary multipath signal. In addition, the influence that can be regarded as constant with respect to time is referred to as continuity.

  The LSI 100 shapes the waveform distortion of the received signal and converts it into a digital signal using the above-described continuity.

  FIG. 6 is a block diagram illustrating a configuration example of the LSI 100.

  The LSI 100 includes an input / output I / F (Interface) 201, an algorithm processing unit 202, an RF (Radio Frequency) signal processing unit 203, and a BB (Base band) signal processing unit 204.

  The input / output I / F 201 mediates data between the algorithm processing unit 202 and another block (not shown) on the substrate 11 based on the control signal and the reference clock signal. The signal that the input / output I / F 201 exchanges with other blocks on the substrate 11 is a signal that complies with a standard such as LVDS (Low Voltage Differential Signal). The control signal and the reference clock signal are supplied to each unit in the LSI 100 as necessary.

  The algorithm processing unit 202 performs predetermined signal processing on the signal supplied from the input / output I / F 201 or the BB signal processing unit 204. For example, when the digital signal supplied from the BB signal processing unit 204 is an image signal, the algorithm processing unit 202 performs linear interpolation processing or DRC (Digital Reality Creation) for performing high image quality processing. be able to. Further, when the digital signal supplied from the BB signal processing unit 204 is an audio signal, the algorithm processing unit 202 can perform processing for separating the supplied audio signal into a surround signal.

  Note that the algorithm processing unit 202 can perform multiplication and division processing on data corresponding to the input signal at a higher speed than the reference clock represented by the reference clock signal. Therefore, the algorithm processing unit 202 can perform calculation even if the signal rate of the signal supplied from the BB signal processing unit 204 is faster than the reference clock. Of course, it is possible to cope with the case where the signal rate of the signal supplied from the BB signal processing unit 204 is slower than the reference clock.

  The algorithm processing unit 202 supplies the processed digital signal to the RF signal processing unit 203 after performing predetermined signal processing on the signal supplied from the input / output I / F 201. The algorithm processing unit 202 supplies the processed digital signal to the input / output I / F 201 after performing predetermined signal processing on the signal supplied from the BB signal processing unit 204.

  The RF signal processing unit 203 performs modulation processing such as ASK modulation on the digital signal supplied from the algorithm processing unit 202, and further multiplies the modulated signal by the carrier signal of the carrier frequency fc to generate RF A signal is generated and supplied to the antenna 101.

  Further, the RF signal processing unit 203 synchronously detects an ASK-modulated modulated wave as an RF signal supplied from the antenna 101, converts it into a baseband signal, and supplies the baseband signal to the BB signal processing unit 204.

  In the present embodiment, the RF signal processing unit 203 transmits and receives an ASK-modulated analog signal. In addition, a PSK (Phase Shift Keying) -modulated analog signal, an analog signal by a spread spectrum method, and the like. May be transmitted and received.

  The BB signal processing unit 204 converts the analog baseband signal supplied from the RF signal processing unit 203 into a digital signal and supplies the digital signal to the algorithm processing unit 202. The baseband signal supplied from the RF signal processing unit 203 is a stationary multipath signal and has a waveform distorted by multipath. The BB signal processing unit 204 changes the communication rate with the algorithm processing unit 202 based on the rate control information from the algorithm processing unit 202.

  FIG. 7 is a block diagram illustrating a detailed configuration example of the BB signal processing unit 204.

  The BB signal processing unit 204 includes an AD (Analog / Digital) conversion unit 301, a data holding unit 302, a classification unit 303, a decoding unit 304, a data holding unit 305, a learning unit 306, and a known signal database 307.

  The BB signal processing unit 204 includes a reception mode for performing reception processing for receiving an analog baseband signal and converting it to a digital signal, and a learning mode for performing learning processing for learning various correspondence tables used in the reception processing. There are two modes. The reception process is performed by the AD conversion unit 301, the data holding unit 302, the classification unit 303, the decoding unit 304, and the data holding unit 305, and the learning process is performed by the AD conversion unit 301, the data holding unit 302, the classification unit 303, and the decoding unit. This is performed by the unit 304, the learning unit 306, and the known signal database 307. Therefore, when the BB signal processing unit 204 is in the learning mode, the LSI 100 functions as a learning device, and when the BB signal processing unit 204 is in the receiving mode, the LSI 100 functions as a receiving device.

  The analog baseband signal output from the RF signal processing unit 203 is supplied to the AD conversion unit 301 of the BB signal processing unit 204. The AD conversion unit 301 samples the baseband signal at a plurality of points for each time representing one bit of the received data of the baseband signal (hereinafter referred to as 1 bit time), and obtains a sampling value obtained as a result in the data holding unit 302 Supply. In this embodiment, as will be described later with reference to FIGS. 8 and 9, the AD conversion unit 301 performs sampling at 3 points per bit time. Or four or more points may be sufficient.

  The data holding unit 302 includes a flip-flop or the like, and holds the sampling value supplied from the AD conversion unit 301 for a predetermined time. After holding the sampling value for a predetermined time, the data holding unit 302 supplies the data to the classification unit 303 when the BB signal processing unit 204 is in the reception mode, and when the BB signal processing unit 204 is in the learning mode. To the learning unit 306.

  In the learning mode, the classification unit 303 obtains, from the learning unit 306, a classification code table (FIG. 12) in which statistical values of received data (sampling values) before waveform shaping having multipath continuity and classification codes are associated with each other. get. Then, the classification unit 303 classifies the reception data in the reception mode based on the classification code table acquired in the learning mode. More specifically, the classification unit 303 stores sampling values for the past two bit times, and the baseband during the most recent one bit time (hereinafter referred to as the current bit time) from the data holding unit 302. The received data is classified by determining the classification code that matches the statistical value of the received data before waveform shaping from the sampled value sampled from the signal and the sampled value sampled in the past 2 bit times, and the classification The resulting classification code is supplied to the decoding unit 304.

  In the learning mode, the decoding unit 304 acquires a bit data table (FIG. 19) in which the classification code learned by the learning unit 306 and digital transmission data (hereinafter referred to as bit data) are associated with each other. The decrypting unit 304 can also obtain a bit data table from an external storage device (not shown) such as an HDD (Hard Disk Drive) or a memory card. In the reception mode, the decoding unit 304 refers to the bit data table and determines bit data based on the classification code supplied from the classification unit 303.

  Note that the classification unit 303 determines the classification code based on the received data before waveform shaping for 3 bits including the current bit time, and the bit data table shows the current bit time corresponding to the classification code. And 3-bit bit data for the past 2-bit time are associated with each other. Therefore, the decoding unit 304 determines 3 bits of bit data and supplies the data to the data holding unit 305.

  The data holding unit 305 is configured by DRAM (Dynamic Random Access Memory) or the like, holds bit data supplied from the decoding unit 304 for a predetermined time so as to have a communication rate instructed by the rate control information, and then This is supplied to the algorithm processing unit 202.

  The learning unit 306 learns multipath continuity in the received signal in the learning mode.

  More specifically, the learning unit 306 corresponds to the transmission / reception data in which the transmission data transmitted by the transmission side, which is the same as the above-described bit data, the classification code, and the statistical value of the reception data before waveform shaping are associated with each other. A table (FIG. 22) is supplied from the known signal database 307. However, since the process of obtaining the statistical value of the received data before waveform shaping is the learning process, the statistical value of the received data before waveform shaping is the initial value in the transmission / reception data correspondence table supplied at the beginning of the learning process. Zero is assigned.

The learning unit 306 accumulates the reception data before waveform shaping supplied from the data holding unit 302 for each classification code. In the learning mode, since the order in which the transmission data is transmitted from other LSIs is determined in advance, the learning unit 306 performs the data holding unit 302 for each classification code corresponding to the transmission data. The received data before waveform shaping supplied from is stored. The learning unit 306 calculates the statistical value of the reception data before waveform shaping from the accumulated reception data before waveform shaping for each classification code, and updates the transmission / reception data correspondence table. The statistical value of the received data is, for example, the average value μ and the variance σ ² of the sampling values. As a result, the learning unit 306 creates a transmission / reception data correspondence table in which the classification code is associated with the transmission data (bit data) and the statistical value of the reception data before waveform shaping.

  The learning unit 306 extracts the correspondence between the classification code and the statistical value of the received data from the transmission / reception data correspondence table, and supplies the correspondence to the classification unit 303 as the above-described classification code table. Further, the learning unit 306 extracts the correspondence between the classification code and the transmission data from the transmission / reception data correspondence table, and supplies it to the decoding unit 304 as the bit data table described above.

  The known signal database 307 acquires the initial value of the transmission / reception data correspondence table in which the transmission data, its classification code, and the statistical value of the reception data before waveform shaping are associated from an external storage device (not shown) and stores it. At the same time, it is supplied to the learning unit 306 as necessary. When the transmission data and the classification code configuration (number) in the transmission / reception data correspondence table do not change, the known signal database 307 decodes the bit data table that the decoding unit 304 has internally, not from the external storage device. It is also possible to obtain from the unit 304, set the statistical value of the received data before waveform shaping to zero, and supply it to the learning unit 306.

  With reference to FIG. 8 and FIG. 9, sampling of the baseband signal in the reception mode will be described.

  As shown in FIG. 8, the AD conversion unit 301 samples three points for each bit time of the baseband signal of the analog signal and supplies the sampled data to the data holding unit 302. Note that at which time in one bit time sampling is determined in advance.

  Accordingly, three sampling values are sequentially supplied to the data holding unit 302 every bit time, held in the data holding unit 302 for a predetermined time, and then supplied to the classification unit 303.

  As described above, the classification unit 303 stores the sampling values for the past two bit times, but of the three sampling values sampled per bit time, for example, the time in the middle of time Only the sampling value is stored.

Therefore, as shown in FIG. 9, the classification unit 303 supplies the three sampling values a ₁ , a ₂ , and a _{3 of} the current bit time supplied from the data holding unit 302 and the current stored in the inside. The received data is classified from one sampling value a ₄ one bit time before the bit time and one sampling value a ₅ two bit times before the current bit time, and a classification code as the classification result Is supplied to the decoding unit 304. In FIG. 9, the times when the sampling values a _{1 to} a ₅ are obtained are t _{1 to} t ₅ , respectively.

  In the present embodiment, as shown in FIG. 9, the number of sampling values per one bit time used for classification of received data is one, but the number is not particularly limited. Therefore, for example, all three sampling values supplied from the data holding unit 302 are stored, and the number of sampling values per one bit time in the past is the same as the number of sampling values in the current bit time. Also good.

  FIG. 10 is a block diagram illustrating a detailed configuration example of the AD conversion unit 301 and the data holding unit 302.

  The AD conversion unit 301 includes ADCs (Analog / Digital Converter) 331-1 to 331-3 and clock signal conversion units 332-1 and 332-2. The data holding unit 302 includes data holding units 341-1 to 341-3.

  The ADC 331-1 converts the analog baseband signal supplied from the RF signal processing unit 203 (FIG. 6) into a digital signal at the timing of the clock signal supplied from the clock signal conversion unit 332-1, and the data holding unit. 341-1.

  The ADC 331-2 converts the analog baseband signal supplied from the RF signal processing unit 203 into a digital signal at the timing of the internal clock signal supplied thereto, and supplies the digital signal to the data holding unit 341-2.

  The ADC 331-3 converts the analog baseband signal supplied from the RF signal processing unit 203 into a digital signal at the timing of the clock signal supplied from the clock signal conversion unit 332-2, and sends it to the data holding unit 341-3. Supply.

  The clock signal conversion unit 332-1 generates a clock signal whose phase is earlier than the internal clock signal by Tds time, and supplies the clock signal to the ADC 331-1. The clock signal conversion unit 332-2 generates a clock signal whose phase is delayed by Tds time than the internal clock signal, and supplies the clock signal to the ADC 331-3. Here, the Tds time is much shorter than the internal clock signal, and can be generated by using both the wiring length and both edges of the synchronous clock.

  The internal clock signal supplied to the AD converter 301 is either a clock signal generated by operating a clock synchronization circuit from received data, or a clock signal generated in the BB signal processing unit 204 is frequency-synchronized with the received data. This is a clock signal with a small phase noise.

ADC331-1 is, in the example of FIG. 9, and supplies the sampled value a ₁ at time t ₁ to the data holding unit 341-1. Further, ADC331-2 supplies the sampled value a ₂ of time t ₂ to the data holding unit 341-2, ADC331-3 supplies the sampled value a ₃ at time t ₃ in the data holding section 341-3. Accordingly, the AD conversion unit 301 performs sampling faster than the synchronous clock.

  In other words, the AD conversion unit 301 does not satisfy the sampling theorem that it cannot be accurately restored unless the sampling frequency fs is twice or more the clock frequency fp of the internal clock signal, but it is advantageous for decoding received data. High-speed sampling is possible. In addition, since the AD conversion unit 301 does not need oversampling that is normally required when demodulating received data, it is not necessary to perform sampling using an ADC that operates at high speed, thereby simplifying the configuration and reducing power consumption. realizable.

  The AD conversion unit 301 performs the above-described operation in both the reception mode and the learning mode.

  The data holding units 341-1 to 341-3 are configured by, for example, a latch circuit that operates with the clock of the internal clock signal, and delays the sampling values supplied from the ADCs 331-1 to 331-3 by a predetermined time. Thereafter, the data is supplied to the classification unit 303 or the learning unit 306. The data holding units 341-1 to 341-3 supply the delayed sampling values to the classification unit 303 in the reception mode and to the learning unit 306 in the learning mode.

  FIG. 11 is a block diagram illustrating a detailed configuration example of the classification unit 303.

  The classification unit 303 includes a classification code determination unit 361, a distributed value RAM (Random Access Memory) 362, and a past bit RAM 363.

In the reception mode, the classification code determination unit 361 obtains the average value μ and variance σ ² of received data when data transmitted from another LSI is received from the variance value RAM 362. The average value μ and variance σ ² of the received data are determined by a learning process described later. The other LSI represents one of the LSIs 102, 104, 106, or 108 that is the communication partner of the LSI 100. The LSI 100 uses the rate control information from the algorithm processing unit 202 as to which LSI the data is transmitted from. It shall be recognized based on this.

The classification code determination unit 361 calculates the maximum value and the minimum value (range) that the reception data can take in each classification code from the average value μ and the variance σ ² for each classification code of the reception data. For example, in the case of setting the range that the received data can take with a probability of 99%, the range of the received data may be set in the range of μ ± 3σ.

The classification code determination unit 361 includes three sampling values a _{1 to} a _{3 of} the current bit time supplied from the data holding unit 302, and a sampling value a _{4 for} the past two bit times stored in the past bit RAM 363, and each a ₅ is, by sequentially narrow down whether contained in the possible range of the receiving data of a predetermined category code to determine the classification code, and supplies the decoding section 304. In addition, the classification code determination unit 361 determines the sampling value a ₂ at the middle of the sampling values a _{1 to} a ₃ of the current bit time in order to determine the classification code of the next received data after determining the classification code. The past bit RAM 363 is supplied and stored.

The distributed value RAM 362 stores, as shown in FIG. 12, a classification code table in which the average value μ and the distribution σ ² (statistical value) of received data are stored for each classification code, LSIs 102, 104, and 106 as communication partners, and Every 108 is stored. The variance value RAM 362 supplies the average value μ and variance σ ² of received data of other LSIs to the classification code determination unit 361 as necessary.

In FIG. 12, received data is shown as n sampling values, but in this embodiment, n = 5. In FIG. 12, the average value μ and the variance σ ² of the received data are expressed in hexadecimal.

The past bit RAM 363 stores one sampling value a ₄ one bit time before the current bit time and one sampling value a ₅ two bit times before the current bit time, and classifies as necessary. This is supplied to the code determination unit 361. Further, when a new sampling value is supplied from the classification code determination unit 361, the past bit RAM 363 overwrites and stores the sampling value with the older time.

On the other hand, in the learning mode, the classification code determination unit 361 is supplied from the learning unit 306, and the classification code in which the average value μ, the variance σ ² (statistical value) of the received data in FIG. The table is supplied to the distributed value RAM 362 and stored.

  The determination of the classification code by the classification unit 303 will be described with reference to FIGS.

As described above, the classification code determination unit 361 includes the three sampling values a _{1 to} a ₃ of the current bit time and the sampling values a ₄ and a _{5 for} the past two bit times stored in the past bit RAM 363, respectively. However, the classification code is determined by narrowing down whether it is within the range of the received data of the predetermined classification code. The classification code determination unit 361 may narrow down the classification values of the sampling values a _{1 to} a ₅ in the order of the sampling values a ₅ , a ₄ , a ₃ , a ₂ , a ₁ , which are in order of oldest time. The classification codes may be narrowed down in the order of the sampling values a ₁ , a ₂ , a ₃ , a ₄ , and a ₅ in order, but in the following, the sampling values a ₅ , a ₄ , a ₃ , and a _{2 in the} old order are used. An example of narrowing down the classification codes in the order of, a ₁ will be described. In FIGS. 13 through 16, the sampling values _{a 5, a 4, a 3} , a 2, of the processing performed in the order of a _1, a sampling value of the current bit time a _3, a _2, and a ₁ will be described However, similar processing is performed for the previous sampling values a ₅ and a ₄ .

Now, it is assumed that the sampling values a ₃ , a ₂ , a ₁ of the received data of the current bit time supplied from the data holding unit 302 are 170, 162, 150, respectively. That is, a ₃ = 170, a ₂ = 162, and a ₁ = 150.

FIG. 13 shows the possible range of the sampling value a ₃ for each classification code according to the average value μ and the variance σ ² acquired from the variance value RAM 362. The horizontal axis in FIG. 13 represents the classification code (0 to 31), and the vertical axis represents the amplitude value. In FIG. 13, the maximum value and the minimum value that the sampling value a ₃ corresponding to the classification codes 0 to 15 can take are represented by *, and the maximum value that the sampling value a ₃ corresponding to the classification code 16 to 31 can take and The minimum value is represented by a mark ●, and the possible range of the sampling value a ₃ is between the marks * or ●. In addition, illustration of the range which some classification codes can take is abbreviate | omitted.

When the possible range of the sampling value a ₃ in each of the classification codes 0 to 31 is determined as shown in FIG. 13 by the classification code table, the classification code determination unit 361 uses the sampling supplied from the data holding unit 302. 170 is the value a _3, searches for a classification code including a possible range.

  As shown in FIG. 13, the classification codes that can take the sampling value 170 are 7, 15, 20, and 31. Therefore, the classification code determination unit 361 determines the classification codes 7, 15, 20, and 31 as classification code candidates corresponding to the received data.

Next, the classification code determining unit 361 performs the same processing on the sampling value a ₂ only for the classification codes 7, 15, 20, and 31 remaining as candidates in the search of the classification code for the sampling value a ₃ . .

FIG. 14 shows the possible range of the sampling value a ₂ for each classification code according to the average value μ and the variance σ ² acquired from the variance value RAM 362. In FIG. 14, the maximum value and the minimum value that can be taken by the sampling value a ₂ corresponding to the classification codes 0 to 15 are represented by ▲, and the maximum value that the sampling value a ₂ corresponding to the classification code 16 to 31 can be taken and The minimum value is represented by an x mark, and the possible range of the sampling value a ₂ is between the ▲ mark or the x marks.

Among the classification codes 7, 15, 20, and 31 remaining as candidates in the classification code search for the sampling value a _3, a classification that may take 162 as the sampling value a ₂ supplied from the data holding unit 302. The codes are 15 and 31. Therefore, the classification code determination unit 361 determines the classification codes 15 and 31 as new classification code candidates corresponding to the received data.

Next, the classification code determination unit 361 performs the same process on the sampling value a ₁ only for the classification codes 15 and 31 remaining as candidates in the search of the classification code for the sampling value a ₂ .

FIG. 15 shows the possible range of the sampling value a ₁ for each classification code according to the average value μ and the variance σ ² acquired from the variance value RAM 362. In FIG. 15, the maximum value and the minimum value that can be taken by the sampling value a ₁ corresponding to the classification codes 0 to 15 are represented by ♦, and the maximum value that the sampling value a ₁ corresponding to the classification code 16 to 31 can be taken. The minimum value is represented by the ■ mark, and the possible range of the sampling value a ₁ is between the ◆ mark or the ■ mark.

Among the classification codes 15 and 31 remaining as candidates in the search of the classification code for the sampling value a ₂ , the classification code that may take 150 as the sampling value a ₁ supplied from the data holding unit 302 is 15. is there. Therefore, the classification code determination unit 361 determines the classification code 15 as the final classification code corresponding to the received data.

  The processing in FIGS. 13 to 15 is summarized as shown in FIG.

From the sampling value a ₃ = 170 at time t _3, the classification codes 7, 15, 20, and 31 indicated by ● among the search target classification codes 0 to 31 indicated by ○ are classified as received data. Candidate for code. Next, from the sampling value a ₂ = 162 at time t ₂ , among the classification codes 7, 15, 20, and 31 to be searched indicated by ◯, the classification codes 15 and 31 indicated by ● are received. Listed as data classification code candidates. Finally, from the sampling value a ₁ = 150 at time t _1, the classification code 15 indicated by ● among the classification codes 15 and 31 to be searched indicated by ○ is determined as the classification code of the received data. The

  FIG. 17 is a flowchart of classification code determination processing by the classification unit 303 in the reception mode.

  The LSI 100 recognizes from which LSI the data is transmitted based on the rate control information from the algorithm processing unit 202, and before the classification code determination process, the communication corresponding to each classification code. Assume that the statistical value of the received data for the partner LSI is acquired from the distributed value RAM 362.

_First , in step S <b> 1, the classification code determination unit 361 supplies the three sampling values a _{1 to} a _{3 of} the current bit time supplied from the data holding unit 302 and the past two bit times stored in the past bit RAM 363. Get the sampling values a ₄ and a ₅ for minutes.

In step S2, the classification code determination unit 361 determines a classification code included in a range that can take the first sampling value, for example, the sampling value a ₅ that is the earliest in terms of time, as a classification code candidate.

In step S3, the classification code determining unit 361, among the candidates for the category code, the next sampling value, for example, a classification code including a range capable of having a sampling value a _4, to determine a candidate for a new classification code.

In step S4, it is determined whether or not all sampling values acquired in step S1 have been examined. If it is determined that all sampling values have not been checked, the process returns to step S3, and the processes of steps S3 and S4 are repeated. Thereby, for example, the same processing is performed on the sampling values a ₃ , a ₂ , and a ₁ .

  If it is determined in step S4 that all sampling values have been examined, in step S5, the classification code determination unit 361 determines a final classification code candidate as a classification code corresponding to the received data, and the decoding unit 304. To supply.

  In step S6, the classification code determination unit 361 stores the sampling value of the current bit time in the past bit RAM 363 to determine the classification code of the next received data, and ends the process.

As described above, the classification code determination unit 361 sequentially narrows down whether or not each of the sampling values a _{1 to} a ₅ is within the range that can be received by the reception data of the predetermined classification code. A classification code is uniquely determined.

  In the example described above, the LSI 100 recognizes from which LSI the data is transmitted based on the rate control information from the algorithm processing unit 202, and the statistical value of the received data for the LSI is read from the distributed value RAM 362. If it is assumed that the data has been acquired but it is not known from which LSI the data is transmitted, the statistics and classification of the received data of each of the LSIs 102, 104, 106, and 108 stored in the distributed value RAM 362 The processing shown in FIG. 17 is executed for the code to search for an LSI whose received sampling value falls within the range.

  FIG. 18 is a block diagram illustrating a detailed configuration example of the decoding unit 304.

  The decoding unit 304 includes a reception data determination unit 381 and a bit value RAM 382.

  In the learning mode, the reception data determination unit 381 acquires a bit data table in which the classification code and the bit data of each LSI as a communication partner are associated with each other from the learning unit 306 and stores the bit data table in the bit value RAM 382.

  The reception data determination unit 381 determines bit data based on the classification code supplied from the classification unit 303 with reference to the bit data table for the communication partner LSI stored in the bit value RAM 382 in the reception mode. To the data holding unit 305.

  As shown in FIG. 19, the bit value RAM 382 stores a bit data table in which a classification code and bit data are associated with each LSI as a communication partner. Note that the decoding unit 304 determines that the classification unit 303 has determined the classification code based on the reception data before waveform shaping for 3 bit times including the current bit time. Since m bit data is determined, m in FIG.

  Next, the reception processing in the reception mode by the BB signal processing unit 204 will be described with reference to the flowchart of FIG.

  First, in step S21, the classification code determination unit 361 of the classification unit 303 determines whether or not to access the distributed value RAM 362. For example, when the communication partner is changed or when the communication partner is unknown and the classification code determination process described with reference to FIG. 17 is performed for a plurality of LSIs one after another, the classification code determination unit 361 In order to acquire a classification code table for a predetermined LSI stored in the distributed value RAM 362, it is determined that the distributed value RAM 362 is to be accessed.

  If it is determined in step S21 that the distributed value RAM 362 is to be accessed, in step S22, the classification code determining unit 361 acquires a predetermined LSI classification code table from the distributed value RAM 362.

  On the other hand, when it is determined in step S21 that the variance RAM 362 is not accessed, and after the processing in step S22, in step S23, the AD conversion unit 301 converts the baseband signal from the RF signal processing unit 203 to A / D. Convert. That is, the AD conversion unit 301 samples three points from the baseband signal for one bit time supplied from the RF signal processing unit 203 and supplies the sampling value obtained as a result to the data holding unit 302. The sampling value supplied to the data holding unit 302 is supplied to the classification unit 303 after being delayed by a predetermined time.

  In step S24, the classification code determination unit 361 of the classification unit 303 determines the classification code of the received data by performing the classification code determination process described with reference to FIG.

  In step S25, the reception data determination unit 381 of the decoding unit 304 refers to the bit data table in which the classification code and the bit data are associated with each other for the communication partner LSI, and determines the classification code supplied from the classification unit 303. Based on this, bit data is determined and supplied to the data holding unit 305.

  In step S26, the data holding unit 305 holds the bit data supplied from the decoding unit 304 for a predetermined time so that the communication rate indicated by the rate control information is obtained.

  After a predetermined time has elapsed, in step S27, the data holding unit 305 outputs the held bit data to the algorithm processing unit 202.

  In step S <b> 28, the BB signal processing unit 204 determines whether or not reception has ended.

  The determination of whether or not the reception in step S28 has been completed is made based on the rate control information from the algorithm processing unit 202 or a control signal input to the LSI 100.

  If it is determined in step S28 that reception has not ended, that is, if the baseband signal is continuously supplied from the RF signal processing unit 203 to the AD conversion unit 301 of the BB signal processing unit 204, the BB signal The processing unit 204 returns the processing to step S23 and repeats the subsequent processing.

  On the other hand, if it is determined in step S28 that the reception has ended, the reception process ends.

  FIG. 21 is a block diagram illustrating a detailed configuration example of the learning unit 306.

  The learning unit 306 includes a reception value RAM 401, an average value calculation unit 402, a variance value calculation unit 403, and a correspondence table creation unit 404, and operates only in the learning mode.

  The received value RAM 401 is supplied with a sampling value from the data holding unit 302 and a classification code from the correspondence table creating unit 404. The reception value RAM 401 stores the sampling value supplied from the data holding unit 302 for each classification code.

  The average value calculation unit 402 uses the sampling value stored in the reception value RAM 401 to calculate the average of the sampling values from the frequency of the classification code, which is the number of times the classification code is assigned, and the sum of the sampling values having the classification code. A value is calculated for each classification code.

  The variance value calculation unit 403 calculates the variance value of the sampling values for each classification code using the sampling value stored in the reception value RAM 401 and the average value calculated by the average value calculation unit 402. Then, the variance value calculation unit 403 supplies the average value from the average value calculation unit 402 and the calculated variance value to the correspondence table creation unit 404.

  The calculation of the average value by the average value calculation unit 402 and the calculation of the variance value by the variance value calculation unit 403 are executed for each LSI with which the LSI 100 communicates.

  Correspondence table creation unit 404 transmits / receives data correspondence table in which transmission data (bit data), its classification code, and statistical values of received data before waveform shaping are associated from known signal database 307 in the initial stage. To get. The initial value of zero is assigned to the statistical value item of the received data in the transmission / reception data correspondence table acquired here.

  In the learning mode, since it is known which sampling data the sampling value supplied from the data holding unit 302 to the received value RAM 401 corresponds to, the correspondence table creating unit 404 receives the received value RAM 401 from the data holding unit 302. The classification code corresponding to the sampling value supplied to the received value RAM 401 is supplied. Then, the correspondence table creation unit 404 updates the transmission / reception data correspondence table with the average value and the variance value supplied from the variance value calculation unit 403 as those of the classification code supplied to the reception value RAM 401.

  The correspondence table creation unit 404 creates the above-described transmission / reception data correspondence table for each communication partner LSI. Then, the statistical value of the reception data before waveform shaping is updated by the sampling value supplied from the data holding unit 302 to the reception value RAM 401 as reception data transmitted from the communication partner LSI, and a transmission / reception data correspondence table is created. Is completed, the correspondence table creation unit 404 extracts the correspondence between the classification code and the statistical value of the received data from the transmission / reception data correspondence table, and supplies it to the classification unit 303 as a classification code table. Also, the correspondence table creation unit 404 extracts the correspondence between the classification code and the transmission data from the transmission / reception data correspondence table, and supplies it to the decoding unit 304 as a bit data table.

  FIG. 22 shows a transmission / reception data correspondence table after the statistical value of the reception data before waveform shaping stored in the internal memory of the correspondence table creation unit 404 is updated.

  As shown in FIG. 22, a transmission / reception data correspondence table in which m-bit transmission data representing bit data, a classification code, and a statistical value of reception data before waveform shaping are associated is provided for each communication partner LSI. It is remembered.

  Next, the learning process in the learning mode by the BB signal processing unit 204 will be described with reference to the flowchart of FIG.

  This learning process is performed when an information processing apparatus 1 is in an initial state such as when the power is turned on for the first time, when a predetermined set time for dealing with a secular change or the like has elapsed, and at any timing or information according to a user instruction This is executed when the configuration in the housing 2 of the processing apparatus 1 changes. When the configuration in the housing 2 of the information processing apparatus 1 changes, when a new module (such as a board or LSI) is additionally mounted on the information processing apparatus 1 or the arrangement of the boards in the information processing apparatus 1 changes. Such as when.

  First, in step S41, the known signal database 307 determines whether a new board is mounted. When it is determined in step S41 that a new board is mounted, in step S42, the known signal database 307 acquires a transmission / reception data correspondence table from an external storage device (not shown) and stores it in the internal memory. It should be noted that the case where the information processing apparatus 1 is in an initial state such as when the power is first turned on includes a case where a new board is mounted.

  On the other hand, if it is determined in step S41 that a new board is not inserted, the known signal database 307 determines in step S43 whether to acquire a new initial value. When it is determined in step S43 that a new initial value is acquired, in step S44, the known signal database 307 acquires a transmission / reception data correspondence table from an external storage device (not shown) and stores it in the internal memory. When it is determined that a new board is not inserted, for example, when a predetermined set time for dealing with secular change or the like has elapsed, when there is an instruction from the user to acquire a transmission / reception data correspondence table, an algorithm When the signal processing in the processing unit 202 is changed.

  The statistical value of the reception data stored in the transmission / reception data correspondence table acquired from the external storage device in steps S42 and S44 may be zero as described above, or other values, for example, used by other information processing devices The value that has been set may be entered.

  When it is determined in step S43 that a new initial value is not acquired, that is, when a statistical value of received data is updated (learned) based on a value currently stored in the BB signal processing unit 204, step S45 is performed. The known signal database 307 acquires the bit data table stored in the decoding unit 304 and substitutes it into the item of received data in the transmission / reception data correspondence table stored in the internal memory.

  In step S46, the known signal database 307 supplies the transmission / reception data correspondence table stored in the internal memory in the process of step S42, S44, or S45 to the learning unit 306. Thereby, the learning unit 306 creates a transmission / reception data correspondence table.

  In step S47, the correspondence table creation unit 404 of the learning unit 306 sets zero to a variable p that counts the number of repetitions.

  In step S <b> 48, the correspondence table creation unit 404 supplies the received value RAM 401 with a classification code of transmission data transmitted from the communication partner LSI.

  In step S49, the reception value RAM 401 receives data transmitted from the communication partner LSI. More specifically, the reception value RAM 401 receives the sampling value sampled by the AD conversion unit 301 via the data holding unit 302.

  In step S <b> 50, the reception value RAM 401 stores the received sampling value in association with the classification code supplied from the correspondence table creation unit 404.

  In step S51, the average value calculation unit 402 calculates the average value of the sampling values using the sampling values stored in the reception value RAM 401.

  In step S <b> 52, the variance value calculation unit 403 uses the sampling value stored in the reception value RAM 401 and the average value calculated by the average value calculation unit 402 to calculate the variance value of the sampling value. Further, the variance value calculation unit 403 supplies the average value and the variance value to the correspondence table creation unit 404.

  In step S <b> 53, the correspondence table creation unit 404 updates the transmission / reception data correspondence table with the average value and variance value supplied from the variance value calculation unit 403 as those of the classification code supplied to the reception value RAM 401.

  In step S54, the correspondence table creation unit 404 determines whether all transmission data has been learned. Here, the correspondence table creation unit 404 has learned about all transmission data when the processing of steps S48 to S53 described above is performed for all classification codes of all the LSIs with which the LSI 100 communicates. judge.

  If it is determined in step S54 that all the transmission data has not been learned, the process returns to step S48, and the BB signal processing unit 204 performs the processes of steps S48 to S54 described above for the next classification code. repeat. If the processing of steps S48 to S54 has already been performed for all the classification codes of the predetermined LSI, the processing of steps S48 to S54 is performed for the predetermined classification codes of other LSIs that have not yet been executed. The processes of steps S48 to S54 described above are executed.

  On the other hand, if it is determined in step S54 that all transmission data has been learned, in step S55, the correspondence table creation unit 404 determines whether or not the variable p is equal to a preset number of repetitions q.

  If it is determined in step S55 that the variable p is not equal to the preset iteration count q, that is, it is determined that learning has not been performed for the preset iteration count q, in step S56, the correspondence table creation unit 404 The variable p is incremented by 1, and the process returns to step S48. Here, the number of repetitions q is the number of times that data with statistically sufficient reliability can be obtained.

  If it is determined in step S55 that the variable p is equal to the preset repetition count q, in step S57, the correspondence table creation unit 404 determines the correspondence between the classification code and the statistical value of the received data from the transmission / reception data correspondence table. The classification code table obtained by the extraction is supplied to the classification unit 303, and the bit data table obtained by extracting the correspondence between the classification code and the transmission data from the transmission / reception data correspondence table is supplied to the decoding unit 304, and the processing ends. To do.

In the learning process of FIG. 23, receiving data for all LSIs and all their classification codes is repeated q times. However, it is of course possible to repeat q times for each LSI in order for all LSIs. . That is, it is only necessary to obtain the average value μ and the variance value σ ² of the received data with statistically sufficient reliability for all the LSIs that are communication partners.

  As described above, in the learning mode, the statistical value of the reception data having distortion due to multipath corresponding to the transmission data is learned by the learning process. The correspondence between the statistical value of the received data and the classification code is stored in the classification unit 303 as a classification code table, and the correspondence between the classification code and the bit data is stored in the decoding unit 304 as a bit data table.

  Accordingly, since the statistical value of the received data with distortion due to multipath is stored in association with the known transmission data, the data (statistical value) for accurately decoding the received data with distortion due to multipath received in the reception process. Can learn.

  On the other hand, in the reception mode, the AD conversion unit 301 samples the reception data at a timing faster than the synchronous clock, and the classification unit 303 uses three sampling values of the current bit time and two of the past two bit times. From these sampling values, the classification code is determined using the classification code table obtained by the learning process. The decoding unit 304 determines bit data corresponding to the classification code from the classification unit 303.

  Therefore, the BB signal processing unit 204 can accurately decode a received signal having distortion due to multipath. The BB signal processing unit 204 can handle both burst signals and packet signals.

  In addition, since the BB signal processing unit 204 learns in advance the multipath continuity, the BB signal processing unit 204 always transmits a preamble and tracks the change of the preamble as in a conventional receiving apparatus. There is no need to initialize or update.

  Furthermore, in the BB signal processing unit 204, when the board mounted on the information processing apparatus 1 is changed, the transmission / reception data correspondence table for the additionally mounted board is acquired from the external storage device or becomes unnecessary. The transmission / reception data correspondence table of the board can be saved in the external storage device. Also, the user can create or purchase a transmission / reception data correspondence table having predetermined characteristics and transfer it to the known signal database 307. In other words, the user can rewrite the transmission / reception data correspondence table, and can easily upgrade the board (or apparatus). From the perspective of the designer and developer, it is only necessary to finally transfer the upgraded transmission / reception data correspondence table without changing the configuration other than the transmission / reception data correspondence table, thus reducing manufacturing costs and reducing the product price. Can be realized.

  In the embodiment described above, the classification unit 303 determines the classification code based on the reception data before waveform shaping for 3 bits including the current bit time, and the decoding unit 304 corresponds to the classification code. It was designed to determine 3 bits of bit data.

  The data holding unit 305 simply delays the output of the bit data for a predetermined time and outputs the received data of 3 bits as it is. However, if only the output of the current bit time is desired, the data holding unit 305 It is also possible to output after removing the data of 2 bits. According to an experiment, when the output value of the current bit time is compared between the decoding by the receiving process of FIG. 20 and the decoding by a general hard decision process using the intermediate value of the received signal as a threshold, it is about 67. % Bit error rate improvement effect was obtained.

  If the data holding unit 305 holds the output of the decoding unit 304 three times, that is, the data for 9 bit time, 1 bit determined as the bit of the current bit time for a certain bit time. A total of three values are obtained: one piece of data and two pieces of data determined as past bits by one or two bit times from the current bit time. Then, it is possible that these three decoded values do not all match, for example, 0, 1, 0, etc. For example, the data holding unit 305 can take a majority decision of these three decoded values and output the decoded value of the larger number.

In the above-described example, the average value μ and variance σ ² of the received data before waveform shaping are obtained and stored in the variance value RAM 362, and the classification code determination unit 361 uses the received data from the average value μ and variance σ ^2. The maximum value and the minimum value that can be taken are calculated. However, the maximum value and the minimum value of the received data may be stored in the variance value RAM 362 as they are.

  In the above, an example of using “radio waves” as a wireless communication transmission medium has been described. However, as a wireless communication transmission medium, in addition, light, solid waves or surface acoustic waves in solids, liquids, gases, etc. Anything having a wave property may be used. Therefore, for example, the present invention can be applied to fixed microwave communication between buildings or between mountains and radio wave communication between radio towers.

  In addition, multipath stationarity can also occur in cable communication, fixed wired communication that has a branch in the circuit and has impedance mismatching, for example, power line communication or bus-connected wiring.

  Therefore, for example, as shown in FIG. 24, a BB signal processing unit 441 similar to the BB signal processing unit 204 is provided between the input / output I / F 201 and the algorithm processing unit 202, and is output from the input / output I / F 201. By performing the above-described reception processing on a digital signal, a digital signal from which multipath continuity due to impedance mismatching or the like is removed can be generated and supplied to the algorithm processing unit 202. In this case, since the BB signal processing unit 441 operates in synchronization with the reference clock signal, rate control information is not necessary.

  A series of processes such as the reception process described above can be performed between computers arranged in a predetermined closed space instead of between LSIs in the housing 2.

  FIG. 25 is a block diagram showing a hardware configuration of a computer that executes the above-described series of processing by a program. A CPU (Central Processing Unit) 501 executes various processes in addition to the functions described above according to a program stored in a ROM (Read Only Memory) 502 or a storage unit 508. A RAM (Random Access Memory) 503 appropriately stores programs executed by the CPU 501 and data. The CPU 501, ROM 502, and RAM 503 are connected to each other by a bus 504.

  An input / output interface 505 is also connected to the CPU 501 via the bus 504. The input / output interface 505 is connected to an input unit 506 made up of a keyboard, mouse, microphone, etc., and an output unit 507 made up of a display, speaker, etc. The CPU 501 executes various processes in response to commands input from the input unit 506. Then, the CPU 501 outputs the processing result to the output unit 507.

  A storage unit 508 connected to the input / output interface 505 includes, for example, a hard disk, and stores programs executed by the CPU 501 and various data. The communication unit 509 communicates wirelessly with an external device using radio waves as a medium. The communication unit 509 includes a router, a modem, and the like as necessary, and performs wired communication.

  A drive 510 connected to the input / output interface 505 drives a removable medium 511 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, and drives programs and data recorded there. Get etc. The acquired program and data are transferred to and stored in the storage unit 508 as necessary.

  As shown in FIG. 25, a program recording medium that stores a program that is installed in a computer and can be executed by the computer is a magnetic disk (including a flexible disk), an optical disk (CD-ROM (Compact Disc-Read Only). Memory), DVD (Digital Versatile Disc), a magneto-optical disk, a removable medium 511 which is a package medium composed of a semiconductor memory, or the like, a ROM 502 in which a program is temporarily or permanently stored, or a storage unit 508 It is comprised by the hard disk etc. which comprise. The storage of the program in the program recording medium is performed via a communication unit 509 using a wired or wireless communication medium such as a local area network, the Internet, or digital satellite broadcasting.

  In the present specification, the steps described in the flowcharts are executed in parallel or individually even if they are not necessarily processed in time series, as well as processes performed in time series in the described order. It also includes processing.

  The embodiments of the present invention are not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the present invention.

It is a figure explaining multipath fading. It is a figure which shows frequency distribution of the amplitude value of the baseband signal produced | generated from the received signal in the housing | casing shown in FIG. It is a perspective view which shows the structural example of one Embodiment of the information processing apparatus to which this invention is applied. It is a figure which shows the received waveform when the transmission waveform transmitted from other LSI and the transmission waveform are received. It is a figure which shows the waveform which averaged the transmission waveform and reception waveform of FIG. FIG. 4 is a block diagram illustrating a configuration example of the LSI of FIG. 3. It is a block diagram which shows the detailed structural example of a BB signal processing part. It is a figure explaining sampling of a baseband signal. It is a figure explaining sampling of a baseband signal. It is a block diagram which shows the detailed structural example of an AD conversion part and a data holding part. It is a block diagram which shows the detailed structural example of a classification | category part. It is a figure which shows the example of the classification code table memorize | stored in distributed value RAM. It is a figure explaining determination of the classification code by a classification part. It is a figure explaining determination of the classification code by a classification part. It is a figure explaining determination of the classification code by a classification part. It is a figure explaining determination of the classification code by a classification part. It is a flowchart explaining the classification code determination process by a classification | category part. It is a block diagram which shows the detailed structural example of a decoding part. It is a figure which shows the example of the bit data table memorize | stored in bit value RAM. It is a flowchart explaining the reception process at the time of the reception mode by a BB signal processing part. It is a block diagram which shows the detailed structural example of a learning part. It is a figure which shows the example of a transmission / reception data correspondence table. It is a flowchart explaining the learning process at the time of the learning mode by a BB signal processing part. FIG. 11 is a block diagram illustrating another configuration example of an LSI. It is a block diagram which shows the structural example of one Embodiment of the computer to which this invention is applied.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 Information processing apparatus, 2 Case, 11 thru | or 14 board | substrate, 100, 102, 104, 106, 108 LSI, 204 BB signal processing part, 301 AD conversion part, 303 Classification part, 304 Decoding part, 306 Learning part, 307 Known Signal database, 361 classification code determining unit, 362 distributed value RAM, 381 received data determining unit, 382 bit value RAM, 401 received value RAM, 402 average value calculating unit, 403 distributed value calculating unit, 404 correspondence table creating unit

Claims (12)

In a receiving apparatus that receives and decodes a signal having multipath stationarity,
When the signal is divided into bit time units corresponding to 1 bit, the sampling value sampled from the signal at the most recently received current bit time and the signal at the past bit time before the current bit time are sampled. Classifying means for classifying the signal based on the sampled value and determining a classification code as a classification result;
Bit data storage means for storing a bit data table in which the classification code and bit data are associated with each other;
A receiving apparatus comprising: the bit data table; and bit data determining means for determining the bit data corresponding to the received signal from the bit data table and the classification code determined by the classification means. The receiving apparatus according to claim 1, wherein the bit data determined by the bit data determining unit includes data of the past bit time. Classification code storage means for storing a classification code table in which a statistical value of a signal having multipath continuity and the classification code are associated;
The receiving apparatus according to claim 1, wherein the classification unit determines the classification code by comparing the sampling values of the current bit time and the past bit time with the statistical value. The classification means calculates a range of a signal having multipath continuity from the statistical value, and determines whether the sampling values of the current bit time and the past bit time are included in the range. The receiving apparatus according to claim 3, wherein the classification code is determined. Sampling means for obtaining a plurality of sampling values from the signal of one bit time;
The receiving apparatus according to claim 1, wherein the classification unit determines the classification code using a plurality of the sampling values at least for the current bit time. The sampling means has clock signal conversion means for shifting the phase around a predetermined time with respect to the clock signal supplied thereto, and the sampling based on the clock signal and the clock signal output from the clock signal conversion means The receiving device according to claim 5, wherein a value is obtained. In a receiving method of a receiving apparatus that receives and decodes a signal having multipath stationarity,
When the signal is divided into bit time units corresponding to 1 bit, the sampling value sampled from the signal at the most recently received current bit time and the signal at the past bit time before the current bit time are sampled. And classifying the signal based on the sampled value, and determining a classification code as the classification result;
A receiving method including a step of determining the bit data corresponding to the received signal from a bit data table in which the classification code and bit data are associated with each other and the determined classification code. In a program for causing a computer to execute a process of receiving and decoding a signal having multipath stationarity,
When the signal is divided into bit time units corresponding to 1 bit, the sampling value sampled from the signal at the most recently received current bit time and the signal at the past bit time before the current bit time are sampled. And classifying the signal based on the sampled value, and determining a classification code as the classification result;
A program including a step of determining the bit data corresponding to the received signal from a bit data table in which the classification code and bit data are associated with each other and the determined classification code. In a learning device that receives data having multipath stationarity and learns data used in a receiving device that decodes the signal,
Receiving means for receiving the signal corresponding to predetermined bit data;
Storage means for storing the signal for each predetermined classification code;
Statistical value calculating means for calculating a statistical value of the signal for each classification code;
A learning apparatus comprising: a correspondence table creation unit that creates a transmission / reception data correspondence table in which the classification code corresponding to the signal is associated with the bit data and the statistical value. The learning device according to claim 9, wherein the statistical value of the signal is an average value and a variance value of the signal. In a learning method for a learning device that receives a signal having multipath stationarity and learns data used by a receiving device that decodes the signal,
Receiving the signal corresponding to predetermined bit data;
Storing the signal for each predetermined classification code;
For each classification code, calculate the statistical value of the signal,
A learning method including a step of creating a transmission / reception data correspondence table in which the classification code corresponding to the signal is associated with the bit data and the statistical value. In a program for causing a computer to execute a learning process for receiving a signal having multipath stationarity and learning data used by a receiving device for decoding,
Receiving the signal corresponding to predetermined bit data;
Storing the signal for each predetermined classification code;
For each classification code, calculate the statistical value of the signal,
A program including a step of creating a transmission / reception data correspondence table in which the classification code corresponding to the signal is associated with the bit data and the statistical value.

Images