JP4142758B2

Method and circuit for the synchronization of an OFDM receiver

Abstract

This record has no abstract on file.

Term

Term ended

Expired 21 October 2017, 8.9 years ago.

7 claims: 6 independent, 1 dependent

1
And transmitted through the communication pathsymbolPeriod TSIn a data stream withsymbolThe symbol has an effective section and a guard section, and the guard section is a duplicate of a part of the effective section. The method is at least a copy of the stream. OnesymbolPeriod TSToSteps to sample over the included interval andData streamIn the above effective sectionConsists of sample data of a pair of stream parts that are only separated from each otherFirst comparison blockaboutThe step of measuring the first variance and the first comparison blockConsists of sample data of the pair of stream parts and a pair of stream parts separated by a predetermined section.Second comparison blockaboutThe step of measuring the second variance, the first variance and the second variance.About the size comparison ofA method comprising the step of calculating a statistic and the step of synchronizing the stream based on the statistic and restoring the signal by a fast Fourier transform. 通信路を通って伝送されかつシンボル期間TSを有するデータストリーム中のシンボルの境界を決定する方法であって、前記記号は有効区間とガード区間とを有し、前記ガード区間は前記有効区間の一部の複製であることを特徴とし、前記方法は、前記ストリームの少なくとも1つのシンボル期間TSを含む区間にわたってサンプリングするステップと、前記データストリームにおいて前記有効区間だけ互いに離れた1対のストリーム部分のサンプルデータからなる第1の比較ブロックについて第1の分散を測定するステップと、前記第1の比較ブロックの前記1対のストリーム部分と所定区間だけ離れた1対のストリーム部分のサンプルデータからなる第2の比較ブロックについて第2の分散を測定するステップと、前記第1の分散及び前記第2の分散の大きさの比較に関する統計値を計算するステップと、前記統計値に基づいて前記ストリームを同期させて、高速フーリエ変換によって信号を復元するステップと、を含む方法。
3
symbolPeriod TSData withsymbolA method of decoding a stream ofsymbolHas an effective section and a guard section, and the guard section is a replica of a part of the effective section.symbolPeriod TSEvaluation ofsectionToLeave, The first step of sampling the stream at sampling intervals and multiple samplesversusThe second step to associate withPreviousRecordEach of multiple pairsHas a first component and a second component, eachIn pairsThe first component is the effective section from the second component in the stream.Just away, SaidMultiple pairsFirst component and saidMultiple pairsThe second component belongs to the first block and the second block, respectively, and the first block and the second block, respectively.SectionIs not larger than the guard section.PreviousRecordEach of multiple pairsAbout the aboveversusThe first component and the second component ofRegarding the difference between the absolute values ofA third step of determining a function and outputting a first output signal indicating the function, a fourth step of measuring the variance of the first output signal, and a second output signal indicating the variance. The fifth step to output and the first block and the second block for the streamOnly for the specified sectionDisplace and repeat the second to fifth steps.FirstThe sixth step of outputting the output signal of 3 and the second output signal described above.as well asThe third output signalAbout the size ofComparisonNanaThen, the seventh step of outputting the fourth output signal indicating the result of the comparison and the eighth step of detecting the boundary between the guard section and the effective section as a predetermined reference in the fourth output signal. And a ninth step of synchronizing the FFT window using the predetermined criteria for signal reproduction. シンボル期間TSを有するデータシンボルのストリームを復号化する方法であって、前記シンボルは有効区間とガード区間とを有し、前記ガード区間は前記有効区間の一部の複製であることを特徴とし、前記方法は、前記シンボル期間TSの評価区間において、サンプリング間隔で前記ストリームをサンプリングする第1のステップと、サンプルを複数の対にして関連付ける第2のステップであって、前記複数の対の各々は第1の構成要素及び第2の構成要素を有し、各対において前記第1の構成要素は前記ストリーム中で前記第2の構成要素から前記有効区間だけ離れ、前記複数の対の第1の構成要素及び前記複数の対の第2の構成要素は夫々第1のブロック及び第2のブロックに属し、前記第1のブロック及び前記第2のブロックの区間は前記ガード区間より大きくないことを特徴とする前記第2のステップと、前記複数の対の各々について、前記対の前記第1の構成要素及び前記第2の構成要素の絶対値の差分に関する関数を決定し、前記関数を示す第1の出力信号を出力する第3のステップと、前記第1の出力信号の分散を測定する第4のステップと、前記分散を示す第2の出力信号を出力する第5のステップと、前記ストリームに対して前記第1のブロック及び前記第2のブロックを所定区間だけ変位させ、さらに、前記第2乃至第5のステップを繰り返して第3の出力信号を出力する第6のステップと、前記第2の出力信号及び前記第3の出力信号の大きさに関する比較をなし、当該比較の結果を示す第4の出力信号を出力する第7のステップと、前記第4の出力信号中の所定の基準としての前記ガード区間と前記有効区間との境界を検出する第8のステップと、信号再生のために前記所定の基準を用いてFFTウィンドウを同期させる第9のステップと、を含む方法。
4
symbolPeriod TSData withsymbolA method of decoding a stream ofsymbolHas an effective section and a guard section, and the guard section is a cyclic extension of a part of the effective section.symbolPeriod TSEvaluation ofsectionToLeave, The first step of sampling the stream at sampling intervals and multiple samplesversusThe second step to associate withPreviousRecordEach of multiple pairsIs the first componentAnd the firstHas 2 components, eachIn pairsThe first component is the effective section from the second component in the stream.Just away, SaidMultiple pairsFirst component and saidMultiple pairsThe second component belongs to the first block and the second block, respectively, and the first block and the second block, respectively.SectionIs not larger than the guard section.PreviousRecordEach of multiple pairsAbout the aboveversusThe first component and the second component ofRegarding the difference between the absolute values ofA third step of determining a function and outputting a first output signal indicating the function, a fourth step of measuring the variance of the first output signal, and a second output signal indicating the variance. The fifth step to output and the first block and the second block for the streamOnly for the specified sectionDisplace and repeat the second to fifth steps.FirstThe sixth step of outputting the output signal of 3 and the second output signal described above.as well asThe third output signalAbout the size ofComparisonNanaThen, the seventh step of outputting the fourth output signal indicating the result of the comparison and the eighth step of detecting the boundary between the guard section and the effective section as a predetermined reference in the fourth output signal. Steps and signal restorationforA method comprising a ninth step of synchronizing FFT windows using the predetermined criteria. シンボル期間TSを有するデータシンボルのストリームを復号化する方法であって、前記シンボルは有効区間とガード区間とを有し、前記ガード区間は前記有効区間の一部の循環的伸長であることを特徴とし、前記方法は、前記シンボル期間TSの評価区間において、サンプリング間隔で前記ストリームをサンプリングする第1のステップと、サンプルを複数の対にして関連付ける第2のステップであって、前記複数の対の各々は第1の構成要素及び第2の構成要素を有し、各対において前記第1の構成要素は前記ストリーム中で前記第2の構成要素から前記有効区間だけ離れ、前記複数の対の第1の構成要素及び前記複数の対の第2の構成要素は夫々第1のブロック及び第2のブロックに属し、前記第1のブロック及び前記第2のブロックの区間は前記ガード区間より大きくないことを特徴とする前記第2のステップと、前記複数の対の各々について、前記対の前記第1の構成要素及び前記第2の構成要素の絶対値の差分に関する関数を決定し、前記関数を示す第1の出力信号を出力する第3のステップと、前記第1の出力信号の分散を測定する第4のステップと、前記分散を示す第2の出力信号を出力する第5のステップと、前記ストリームに対して前記第1のブロック及び前記第2のブロックを所定区間だけ変位させ、さらに、前記第2乃至第5のステップを繰り返して第3の出力信号を出力する第6のステップと、前記第2の出力信号及び前記第3の出力信号の大きさに関する比較をなし、当該比較の結果を示す第4の出力信号を出力する第7のステップと、前記第4の出力信号中の所定の基準としての前記ガード区間と前記有効区間との境界を検出する第8のステップと、信号復元のために前記所定の基準を用いてFFTウィンドウを同期させる第9のステップと、を含む方法。
5
symbolPeriod TSData withsymbolA method of decoding a stream ofsymbolHas an effective section and a guard section, and the guard section is a cyclic extension of the final portion of the effective section.symbolPeriod TSEvaluation ofsectionToLeave, The first step of sampling the stream at sampling intervals and multiple samplesversusThe second step to associate withPreviousRecordEach of multiple pairsIs the first componentAnd the firstHas 2 components, eachIn pairsThe first component is the effective section from the second component in the stream.Just away, SaidMultiple pairsFirst component and saidMultiple pairsThe second component belongs to the first block and the second block, respectively, and the first block and the second block, respectively.SectionIs not larger than the guard section.PreviousRecordEach of multiple pairsAbout the aboveversusThe first component and the second component ofDifferences regarding the absolute value ofAnd saidDifferenceThe third step of outputting the first output signal indicating, and the first output signalsizeA fourth step of measuring the variance, a fifth step of outputting a second output signal indicating the variance, and the first block and the second block with respect to the stream.Only for the specified sectionDisplace and repeat the second to fifth steps.FirstThe sixth step of outputting the output signal of 3 and the second output signalSize andOf the third output signalIn F-test regarding sizeA seventh step of determining the F ratio and outputting a fourth output signal indicating the F ratio, and a predetermined position in the fourth output signal.sizeEighth step to detect the boundary between the guard section and the effective section, and signal restorationforThe predeterminedsize9th step to synchronize FFT windows with, and how to include. シンボル期間TSを有するデータシンボルのストリームを復号化する方法であって、前記シンボルは有効区間とガード区間とを有し、前記ガード区間は前記有効区間の終期部分の循環的伸長であることを特徴とし、前記方法は、前記シンボル期間TSの評価区間において、サンプリング間隔で前記ストリームをサンプリングする第1のステップと、サンプルを複数の対にして関連付ける第2のステップであって、前記複数の対の各々は第1の構成要素及び第2の構成要素を有し、各対において前記第1の構成要素は前記ストリーム中で前記第2の構成要素から前記有効区間だけ離れ、前記複数の対の第1の構成要素及び前記複数の対の第2の構成要素は夫々第1のブロック及び第2のブロックに属し、前記第1のブロック及び前記第2のブロックの区間は前記ガード区間より大きくないことを特徴とする前記第2のステップと、前記複数の対の各々について、前記対の前記第1の構成要素及び前記第2の構成要素の絶対値に関する差分を決定し、前記差分を示す第1の出力信号を出力する第3のステップと、前記第1の出力信号の大きさの分散を測定する第4のステップと、前記分散を示す第2の出力信号を出力する第5のステップと、前記ストリームに対して前記第1のブロック及び前記第2のブロックを所定区間だけ変位させ、さらに、前記第2乃至第5のステップを繰り返して第3の出力信号を出力する第6のステップと、前記第2の出力信号の大きさ及び前記第3の出力信号の大きさに関してF検定におけるF比を決定し、前記F比を示す第4の出力信号を出力する第7のステップと、前記第4の出力信号中の所定の大きさとして前記ガード区間と前記有効区間との境界を検出する第8のステップと、信号復元のために前記所定の大きさを用いてFFTウィンドウを同期させる第9のステップと、を含む方法。
6
It is a timing synchronization method for demodulating a signal modulated by frequency division multiplexing, and the signal issymbolPeriod TSData withsymbolHas a stream ofsymbolHas an effective section and a guard section, the guard section is a replica of a part of the effective section, and the method is at least described above.symbolPeriod TSThe first step of temporally sampling the stream at sampling intervals and a plurality of samples.versusThe second step to associate withPreviousRecordEach of multiple pairsIs the first componentAnd the firstHas 2 components, eachIn pairsThe first component is the effective section from the second component in the stream.Just away, SaidMultiple pairsFirst component and saidMultiple pairsThe second component belongs to the first block and the second block, respectively, and the first block and the second block, respectively.SectionIs not larger than the guard section.PreviousRecordEach of multiple pairsAbout the aboveversusThe first component and the second component ofDifferences regarding the absolute value ofAnd saidDifferenceThe third step of outputting the first output signal indicating, and the first output signalsizeA fourth step of measuring the variance, a fifth step of outputting a second output signal indicating the variance, and the first block and the second block with respect to the stream.Only for the specified sectionDisplace and repeat the second to fifth steps.FirstThe sixth step of outputting the output signal of 3 and the second output signalSize andOf the third output signalIn F-test regarding sizeA seventh step of determining the F ratio and outputting a fourth output signal indicating the F ratio, and a predetermined position in the fourth output signal.sizeEighth step to detect the boundary between the guard section and the effective section, and signal restorationforThe predeterminedsize9th step to synchronize FFT windows with, and how to include. 周波数分割多重により変調された信号を復調するためのタイミング同期方法であって、前記信号はシンボル期間TSを有するデータシンボルのストリームを有し、前記シンボルは有効区間とガード区間とを有し、前記ガード区間は前記有効区間の一部分の複製であることを特徴とし、前記方法は、少なくとも前記シンボル期間TSについて、サンプリング間隔で前記ストリームを時間的にサンプリングする第1のステップと、サンプルを複数の対にして関連付ける第2のステップであって、前記複数の対の各々は第1の構成要素及び第2の構成要素を有し、各対において前記第1の構成要素は前記ストリーム中で前記第2の構成要素から前記有効区間だけ離れ、前記複数の対の第1の構成要素及び前記複数の対の第2の構成要素は夫々第1のブロック及び第2のブロックに属し、前記第1のブロック及び前記第2のブロックの区間は前記ガード区間より大きくないことを特徴とする前記第2のステップと、前記複数の対の各々について、前記対の前記第1の構成要素及び前記第2の構成要素の絶対値に関する差分を決定し、前記差分を示す第1の出力信号を出力する第3のステップと、前記第1の出力信号の大きさの分散を測定する第4のステップと、前記分散を示す第2の出力信号を出力する第5のステップと、前記ストリームに対して前記第1のブロック及び前記第2のブロックを所定区間だけ変位させ、さらに、前記第2乃至第5のステップを繰り返して第3の出力信号を出力する第6のステップと、前記第2の出力信号の大きさ及び前記第3の出力信号の大きさに関してF検定におけるF比を決定し、前記F比を示す第4の出力信号を出力する第7のステップと、前記第4の出力信号中の所定の大きさとして前記ガード区間と前記有効区間との境界を検出する第8のステップと、信号復元のために前記所定の大きさを用いてFFTウィンドウを同期させる第9のステップと、を含む方法。
7
It is a frequency division multiplexing signal demodulator, and the signal issymbolPeriod TSData withsymbolHas a stream ofsymbolHas an effective section and a guard section, the guard section is a duplicate of a part of the effective section, and the demodulator includes a timing synchronization circuit for detecting a boundary position of the guard section.Timing synchronizationThe circuit provides a first predetermined interval delay, a first delay circuit that receives the sampled input signal, and a first subtractor that subtracts the output of the first delay circuit from the input signal. A first circuit connected to the first subtractor to generate a signal having a unipolar amplitude proportional to the input, a second delay circuit connected to the output of the first circuit, and the first circuit. An addition / subtraction circuit having a first input connected to the output of the circuit, a second input connected to the output of the second delay circuit, and a third input, and an output of the addition / subtraction circuit. The register is a register connected to the third input, and is addressed according to the output of the addition / subtraction circuit, and the addition / subtraction circuit is used.Subtraction circuitA third delay circuit that provides a second predetermined interval delay and has an input connected to the lookup memory, and a third delay circuit connected to the lookup memory. A second subtractor having one input and a second input connected to the output of the third delay circuit, and the second subtractor connected to the output of the second subtractor. A demodulator with a comparison circuit that compares the output of the with a threshold. 周波数分割多重信号の復調器であって、前記信号はシンボル期間TSを有するデータシンボルのストリームを有し、前記シンボルは有効区間とガード区間とを有し、前記ガード区間は前記有効区間の一部分の複製であり、前記復調器は前記ガード区間の境界位置を検出するタイミング同期回路を備え、前記タイミング同期回路は、第1の所定間隔の遅延を提供し、サンプリングされた入力信号を受け取る第1の遅延回路と、前記入力信号から前記第1の遅延回路の出力を減算する第1の減算器と、前記第1の減算器に接続され、入力に比例する単極性振幅を有する信号を生成する第1の回路と、前記第1の回路の出力に接続された第2の遅延回路と、前記第1の回路の前記出力に接続された第1の入力と、前記第2の遅延回路の出力に接続された第2の入力と、第3の入力と、を有する加減算回路と、前記加減算回路の出力を蓄積するレジスタであって、前記レジスタは前記第3の入力に接続されるレジスタと、前記加減算回路の出力に従ってアドレスされ、前記加減算回路の前記出力の対数を呼び出すルックアップメモリと、第2の所定間隔の遅延を提供し、前記ルックアップメモリに接続された入力を有する第3の遅延回路と、前記ルックアップメモリに接続された第1の入力と、前記第3の遅延回路の出力に接続された第2の入力と、を有する第2の減算器と、前記第2の減算器の出力に接続され、前記第2の減算器の出力を閾値と比較する比較回路と、を備える復調器。

Independent claims6

73 paragraphs, as filed

The present invention relates to a receiver of an electromagnetic signal that employs multicarrier modulation. More specifically, the present invention relates to the synchronization of timed intervals required to compute the Fast Fourier Transform in a receiver that employs orthogonal frequency division multiplexing.

Coding Orthogonal Frequency Division Multiplexing (COFDM) has been proposed for digital audio and digital video broadcasting, both of which are efficient use of limited bandwidth, as well as some. Despite the impact, we need a reliable transmission method. For example, the impulse response of a typical communication path can be modeled as the sum of multiple Dirac pulses with different delays. Each pulse is subject to a multiplication factor, where the amplitude generally follows Rayleigh's law. Such pulse trains can extend over several microseconds, making uncoded transmissions at high bit rates unreliable. In addition to random noise, impulse noise, and fading, other major problems with high data rate digital terrestrial transmission are multipath propagation and adjacent channel interference, which results in signal changes with high correlation of proximity frequencies. including. COFDM is particularly suitable for these applications. In a real COFDM configuration, a relatively small amount of data is modulated into each of a large number of carriers placed close together at frequency. datasymbolThe interval increases at the same rate as the number of carriers or subchannels, therebysymbolInterference is significantly reduced.

[0003] FIGS. 1 and 2 show COFDM multiplexing, where the spectrum of a single COFDM carrier or subchannel is shown by line 2. A set of carrier frequencies is shown by the overlapping waveforms in FIG. 2, where the orthogonality condition is satisfied. In general, the two real-valued functions are orthogonal if:

[0004] [Number 1]<img file="JP4142758B2_D0001.tif" />Here, K is a constant, and if p q, then K = 0; if p = q, then K 0. The actual coding and decoding of the signal by COFDM relies heavily on the Fast Fourier Transform (FFT), as can be seen from the equation below.

[0006] The signal of carrier c is given by the following equation.

[0007] [Number 2]<img file="JP4142758B2_D0002.tif" />[0008] Here, AcIs the data at time t, ωcIs the carrier frequency, φcIs the topological. The N carriers in the COFDM signal are given by:

[0009] [Number 3]<img file="JP4142758B2_D0003.tif" />[0010] [Number 4]<img file="JP4142758B2_D0004.tif" />[0011]symbolWhen sampled over a period, [0012] [Equation 5]<img file="JP4142758B2_D0005.tif" />[0013] [Number 6]<img file="JP4142758B2_D0006.tif" />At a sampling frequency of 1 / T, the result signal is represented by the following equation.

[0015] [Number 7]<img file="JP4142758B2_D0007.tif" />[0016] ω0= 0, one datasymbolWhen sampling over the period T = NT, [0017] [Equation 8]<img file="JP4142758B2_D0008.tif" />This is comparable to the general form of the inverse discrete Fourier transform.

[0019] [Number 9]<img file="JP4142758B2_D0009.tif" />[0020] In the above equation,AnejφnIs the input signal in the sampled frequency domain, Ss(kT) is the time domain display. The increase in FFT size is longsymbolIt provides a period of time and improves the ruggedness of the system in terms of echoes that exceed the length of the guard section. But Nlog2N increases computational complexity and is a practical limitation.

[0021] Caused by a transmission linesymbolOrthogonality between signals is not maintained in the presence of inter-interference. One approach to this problem is for each section in the time domain that exceeds the memory of the communication path and any multipath delay.symbolWas intentionally sacrificed to some extent by radiant energy. Any "guard section" so selectedsymbolLarge enough to absorb inter-interference, eachsymbolCan be established by preceding a part of the replica. This replica is typicallysymbolIt is a cyclical extension of the terminal part of. See Figure 3 for datasymbol4 issymbolIt has an effective section 6 that includes all the transmitted data inside. The final part 8 of the effective section 6 is used as the guard section 10.symbolRepeated at the beginning of. The COFDM signal is indicated by the solid line 12.symbolAt the end of, the initial part of effective interval 6 can be repeated cyclically.

Transmission of COFDM data can be achieved by the known general scheme shown in FIG. The serial data stream is converted into a series of parallel streams 16 within the serial-parallel converter 18. Each parallel stream is grouped by x bits, each of which produces a complex number, where x determines the data array of its associated parallel stream. After external coding and interleaving within block 20, a pilot carrier is inserted by the signal mapper for use in synchronization and channel estimation within the receiver. Pilot carriers typically come in two types. The continuous pilot carrier is transmitted in the same position in each symbol with the same phase and amplitude. In the receiver, these are used for phase denoising, automatic frequency control, and time / sampling synchronization. Scattered pilot carriers are distributed within the symbol and their position typically varies between symbols. They are mainly useful for channel estimation. The complex number is then modulated in the baseband within block 24 by an inverse fast Fourier transform (IFFT). After that, a guard section is inserted at block 26. The discrete symbols are then converted to analog, typically lowpass filtered, and then upconverted to radio frequency at block 28. After that, the signal is transmitted through the communication path 30 and received in the receiver 32. As is well known in the prior art, the receiver applies the reverse of the transmission process to obtain the transmission information. In particular, the FFT is applied to demodulate the signal.

A modern application of COFDM is the European Telecommunications Standard DRAFT pr ETS 300744 (May). Proposed in 1996), it specifies framing structure, channel coding, and modulation for digital terrestrial television. This standard is designed to adapt digital terrestrial television to existing spectral allocations for analog transmission and provides adequate protection against even higher levels of mutual and adjacent channel interference. Flexible guard intervals are specified so that the system provides adequate network configuration and adequate protection against mutual and adjacent channel interference from existing PAL / SECAM services while maintaining high spectrum efficiency. Can be done. The above European telecommunications standards specify two modes of operation: "2K mode" suitable for single transmitter operation and small single frequency networks with limited transmitter distances. "8K mode" can be used for either single transmitter operation or large single frequency networks. Various levels of quadrature amplitude modulation (QAM) are maintained, they have different internal code rates to balance the bit rate with respect to ruggedness. The system is intended to adapt the transport layer according to the Moving Picture Experts Group (MPEG) and is directly compatible with MPEG-2 encoded television signals (ISO / IEC 13818). Has sex.

[0024] In the European telecommunications standards described above, the data carrier of the COFDM frame uses quadrature phase shift keying (QPSK), 16-QAM, 64-QAM, non-uniform 16-QAM, or gray mapping. It can be either non-uniform 64-QAM.

[0025] An important problem in receiving a COFDM transmission is that it is difficult to maintain synchronization due to phase noise and jitter, and the phase noise and jitter are pre-transmission. It is caused by the up-conversion of the above, the down-conversion at the receiver, and the front-end oscillator in the tuner, which is typically a voltage controlled oscillator. Other than providing a pilot carrier to assist in synchronization during demodulation, these issues are not specifically addressed by the above European telecommunications standards and are left to the practitioner's resolution.

[0026] Basically, there are two types of phase disturbances. First, the noise component that interferes with adjacent carriers in a multi-carrier system is called the "external noise contribution" ("FNC"). Second, the noise component that interferes with one's own carrier is called "self-noise contribution".

[0027] With reference to FIG. 5, the ideal sequence sample position is indicated by the x symbol 34. The effect of external noise contribution is stochastic and produces Gaussian noise. Samples that perturb in this way are shown by circle 36 in FIG. The effect of self-noise contribution is the common rotation of all array positions, shown as the displacement of each "x" symbol 34 and its associated circle 36. This is called a "common phase error" and it issymbolSignificantly changed between, so eachsymbolPeriod TSMust be calculated in. Also, the common phase errorsymbolPeriod TSIt can be interpreted as the average phase deviation in.

[0028] The receiver 32 is the data in the actual system.symbolEach data to processsymbolMathematical operations are performed on the complex signal indicating. Generally, this is an FFT. Received FFT interval data for valid resultssymbolCertain forms of timing synchronization are required to align with.

[0029] A first object of the present invention is reception data in a signal transmitted by COFDM.symbolIs to provide an improved way to synchronize with the FFT window.

[0030] Another object of the present invention is to improve terrestrial transmission of COFDM signals.

[0031] Yet another object of the present invention is to provide an improved device for receiving COFDM signals.

[Means for Solving the Problems] These and other purposes of the present invention are:symbolPeriod TSData withsymbolAchieved by the method of decoding the stream of.symbolDefines an effective section and a guard section, which is a duplicate of a part of the effective section, preferably a cyclic extension of its end. Stream issymbolPeriod TSEvaluation ofsectionToLeaveIt will be sampled. sampleAs a pairAssociated and oneversusThe component of is the valid intervalOnlyIn the data streamis seperated。versusData belong to the first and second blocks, respectively, and each block is not larger than the guard section. Of each dataversusAbout, preferablyversusThe function, which is the difference between the components of, is determined and output as the first output signal. Then, the variance of the first output signal is measured and output as the second output signal. The first and second blocks are then displaced relative to the stream, and the preceding steps of associating samples, determining functions, and measuring variance are repeated. The result is output as a third output signal. The second output signal is compared with the third output signal and the result is output as the fourth output signal. Then, a predetermined criterion indicating the boundary between the guard section and the effective section in the fourth output signal is detected. The reference can be a predetermined amplitude, maximum value or minimum value.

[0033] Preferably, the absolute value of the sample is calculated. However, according to one aspect of the invention, it is necessary that only the real or imaginary components of the stream be sampled.

The F ratio is used to compare the second and third output signals of the two comparison blocks. Preferably, the F ratio is evaluated by subtracting the logarithm of the numerator and denominator of the F ratio. The F ratio of consecutive pairs of comparison blocks is evaluated for the peaks that mark the boundaries of the guard interval. Peak determination includes a statistical significance test.

[0035] In the first operation mode, the evaluation interval has a fixed first value, and in the second operation mode, the evaluation interval has a second value, making it possible to adapt to the changing signal state. To do. In one aspect of the invention, in the second mode of operation, it is continuous.symbolThe variance of the position of the guard section inside is stored, and the value of the evaluation interval is adjusted according to the stored variance.

[0036] In another aspect of the present invention, the detection of the boundary of the guard section is followed by subsequent detection.symbolNew before the start of detection of the boundary of the guard section insideversusSamples are sampled and received for the valid interval.

The present invention provides a frequency division multiplexing signal demodulator in which the signal issymbolPeriod TSData withsymbolIs a stream ofsymbolIsValid sectionas well asGuard sectionTo specify. The guard section is a replica of a portion of the effective section, preferably a cyclic extension of its terminal portion. A timing synchronization circuit is provided in the demodulator to detect the boundary position of the guard section. The circuit is preferably implemented as a first-in-first-out memory (FIFO) and the input signal s.iCan store L samples, preferably the size of the FFT window, connected to the input signal s only in the interval equal to the FFT window.iIt has a first delay circuit that delays. This delayed version of the input signal is then subtracted from the input signal. The absolute value of the input signal or difference signal is calculated. The circuit has a second delay circuit, which delays the absolute value by the first predetermined interval. The add / subtract circuit subtracts the delayed absolute value from the absolute value and adds the feedback signal obtained from the register that stores the result of the add / subtract circuit. The lookup memory, which is addressed according to the output of the add / subtract circuit, holds the logarithm of the result produced by the add / subtract circuit. A third delay circuit that provides a second predetermined interval delay has a first input connected to a lookup memory. The second subtractor has a first input connected to the lookup memory and a second input connected to the output of the second delay circuit. The comparison circuit thresholds itWhenConnected to the output of the subtractor for comparison.

[0038] From one point of view of the present invention, the circuit is an input signal s.iIt has a fourth delay circuit connected to, which can store preferably less than L samples while providing a delay for the effective interval. The selector selects the first delay circuit or the fourth delay circuit.

[0039] According to a further aspect of the present invention, there is a module that calculates the absolute value of the input signal connected to the first delay circuit and the addition / subtraction circuit.

[0040] According to another aspect of the invention, the control circuit selects a given sample of the input signal for processing in the timing synchronization circuit and is in the data stream.symbolAllows selection of evaluation interval for.

BEST MODE FOR CARRYING OUT THE INVENTION For a better understanding of these and other objects of the invention, a detailed description of the invention as an example should be referenced and read with the accompanying drawings.

With reference to FIGS. 3 and 4 again, according to the present invention, a statistical method is applied to the COFDM signal to find the terminal portion of the guard interval 10. Although this method is described with reference to the European telecommunications standards described above, it is applicable to many forms of frequency division multiplexing with pre- or post-guard sections. Thereby, given only the received sample complex signal (solid line 12) and the magnitude of the effective interval 6, the receiver 32 can find the final part of the guard interval. The method is that the guard section 10 is the datasymbolBased on the fact that it is a copy of the last part of 4. In receiver 32, guard section 10 and data due to echo and noise from the communication path and local oscillator error.symbolDifferent from the last part of 4. If the errors introduced are random, statistical methods can be applied. According to the present invention, the received complex signal is sampled at about the same rate as that used in the transmitter. The difference signal is as close as possible to the effective interval 6.Only the section is separatedReceived sampleversusIs found out about. This period should be equal to the size of the applied Fast Fourier Transform (FFT) (ie, 2048 or 8192 samples).

[0043] [Number 10]<img file="JP4142758B2_D0010.tif" />Assuming [0044], SiThe difference signalWhenThen siAnd si-fftsizeIs the current and previous input samples for which the absolute value is obtained. That is, the subscript "i" indicates the linear time order of the input values. Assuming the input signal is random, SiWill also be random. Within the guard section, it will not be the same due to the influence of the communication path, but siAnd si-fftsizeApproximates. Therefore SiIs a random signal with a small variance. The term "variance" as used herein generally refers to the spread of values and is not limited to any particular mathematical definition. Generally onesymbolThe effective part ofsymbolNot related to the effective part of. Outside the guard section, SiIs random and has a very large variance. Difference signal S to find the end of the guard sectioniMonitor the variance of and look for significant increases that occur at the boundary between guard section 10 and effective section 6. The inventors also observed that a large decrease in variance was observed at the beginning of the guard section 10.

[0045] In a preferred embodiment of the invention, at least one sample of the input signal is sampled.symbolPeriod TSMemorize over the section containing. Difference signal SiThe variance of is calculated over one block of the sample. Go back in time for multiple samples, n, and recalculate the variance. These two blocks are referred to herein as "comparison blocks". The ratio of the current variance in the first comparison block to the variance in the previous comparison block is found. Then use the F-ratio significance test to find significant differences in the variance between the two comparison blocks. The F ratio is given by the following formula, [0046] [Equation 11]<img file="JP4142758B2_D0011.tif" />Where n is a positive integer, i is an index of the input sample, and VAR (i) is the variance of the block of values of the length N sample. The variance is defined as follows.

[0048] [Number 12]<img file="JP4142758B2_D0012.tif" />Although the F-ratio significance test is used here in a preferred embodiment, other functions of the two variance values that give a signal associated with the change in variance can also be used. There are many such functions. The advantage of the F ratio is that it has a known probability distribution for random input signals, allowing convenient statistical analysis for performance analysis and system design purposes. Also, the F ratio essentially normalizes the signal, and the result depends on the signal level.

The method is described with reference to FIG. 6, where the sample in the current evaluation block in step 38 is described.versusThe first component of is measured. At step 40, a delay of one effective interval 6 (Fig. 3) is experienced. This is achieved by using a digital delayer such as a FIFO, or equivalently buffering a sample in memory for only one valid interval and accessing the appropriate memory cell. sampleversusThe second component of is measured in step 42 and the difference between the first and second components is determined and stored in step 44. The final part of the current block is tested in decision step 46. The size of the evaluation block should not exceed the length of the guard section and can be very small. If the end of the current block has not yet been reached, another sample is taken in step 48 and control returns to step 38.

When the end of the current block is reached, the variance of the current block is measured in step 50 and processed as one of the two data comparison blocks. At step 52, a test is made to determine if one group of two comparison blocks has been evaluated. If this test is negative, another block of data is acquired in step 54, after which control returns to step 38. Other blocks of data do not have to be adjacent to the block that just ended.

[0052] If the test in decision step 52 is positive, then step 56 calculates the F ratio for the two groups of comparison blocks. The result obtained in step 56 is used for peak detection in step 60. Peak detection optionally includes a statistical test of significance, as described below.

[0053] Once the peak is detected, then the boundary of the guard section is established in step 62 for the purpose of synchronizing the FFT window required for subsequent signal reconstruction. If no peaks are detected, repeat the above process with blocks of samples taken from other parts of the data stream.

【0054】Example 1 With reference to FIG. 7, a random number generator was used to generate a complex signal according to the European telecommunications standard described above, with white Gaussian noise (SNR = 3.7) added and transmitted through the Ricean channel mode. Then the datasymbolWas analyzed according to the method described above. 6 results datasymbolIs shown in FIG. 7, where spikes 66 and 61 at the start and end of the guard section are so large that the F ratio is plotted on the logarithmic axis as line 64 for convenience of display.

Although it is very clear from Figure 7 that the end of the guard section can be easily found using some known peak detectors, do the two blocks of the sample have the same variance? Statistical tests can be applied to answer the question ?, more accurately. This is the null hypothesis, H0That is, the variances are the same, and the spikes observed in F are due solely to random variation. H0If has a very low probability, it can be rejected, which will correspond to the detection of the start or end of the guard section. COFDMsymbolFrom the method of constructing, H for comparison blocks that completely belong to the guard section or effective section0Is expected to be true, but is expected to be false when the comparison block straddles the start or end boundary of the guard interval. Assuming that the comparison block of the random sample is derived from the same population, the probability of F is given by the following equation, [0056] [Equation 13].<img file="JP4142758B2_D0013.tif" />Here, I () is an incomplete beta function, [0058] [Equation 14].<img file="JP4142758B2_D0014.tif" />[0059] v1And v2Is the degree of freedom, by which the first and second variances are evaluated. In this example, v if n> = N1= v2= (N-1). The shape of the function is shown in Figure 8. From a statistical point of view, n should be large enough so that the two blocks do not overlap, i.e. n> = N. When the blocks overlap, the second variance calculation uses the sample used to calculate the first variance. This effectively reduces the degree of freedom, thereby reducing the significance of the result. It turns out that the n = N setting works well.

[0060] The function Q () in Eq. (13) actually gives one trailing probability. If F is very large or very small, then H0Can be rejected, which requires two trailing tests. In reality, the two endings are identical, so for the two ending tests, the probability is twice that given by Eq. (13). However, this yields a probability value greater than 1 for F <1. Therefore, the probability p is calculated as follows: [0061] [Equation 15]<img file="JP4142758B2_D0015.tif" />Therefore, if (p> 1), then p = 2-p. This probability is H0Reflects the survival of. Therefore, if p is small, then H0Is rejected and with certainty identified, it can be said that the comparative blocks came from a sample population with different variances. The European Telecommunications Standards document above states that the block size N should be 32 for the correlation algorithm. N = {32, 64} was tested and succeeded. The resulting probability function is shown in Figure 9 using these values for N. In a preferred embodiment, H0P <= 0.05 was set for the rejection of.

[0063] For exact execution, calculate F, then x, then incomplete beta function, then p, and then apply the threshold test. Beta functions are so complex that it is very difficult to implement this algorithm in hardware. In a preferred embodiment, setting the receipt threshold and N parameters is fairly simple and gives the same result, thus defining upper and lower limits for F. Then it is only necessary to calculate F and compare it with the upper and lower limits. To easily find the end of the guard section, we can safely assume F> 1. Only the upper limit of F is needed. To accurately calculate the F limit, a method of finding a suitable root, such as the Newton-Raphson formula, can be used. Typical values are shown in Table 1.

[0064] [Table 1]<img file="JP4142758B2_D0016.tif" />This method was successful in tests using a specific channel model with the addition of white Gaussian noise (SNR = 3.7).

The variance equation given in equation (12) requires a multiplier to run in silicon. The calculation of F is division, where, as long as the two blocks have the same size, (N-1 normalization constants cancel each other out. In silicon, exact multiplication and division are expensive. In certain embodiments, a simplification is performed for F that is less accurate but still gives a valid value.iIs presumed to have a zero mean, so there is no need to calculate the mean from a block of samples. This also increases the degree of freedom from (N-1) to N. Instead of calculating the variance using the sum of standard square equations, the variance can be estimated by mean absolute deviation. The formula for VAR (i) is as follows.

[0067] [Number 16]<img file="JP4142758B2_D0017.tif" />[0068] If two blocks have the same size, they can be eliminated by dividing the coefficient of (1 / N) in the calculation of F. However, the division of the two variances still remains, requiring a square. These can be dealt with by a logarithm with a base of 2. Substitution of equation (16) into equation (11) gives the following equation.

[0069] [Number 17]<img file="JP4142758B2_D0018.tif" />[0070] Taking the logarithm with 2 as the base, [0071] [Equation 18]<img file="JP4142758B2_D0019.tif" />[0072]. Therefore, it is only necessary to calculate y and compare it with the base 2 logarithm of the upper limit of F. For comparison, the logarithm of the limit value is 2 (log2sa-log2sb b) And compare with zero. The coefficient 2 can be absorbed within the limit value.

[0073] The calculation of the base 2 logarithm is relatively simple when the number is stored as a fixed-point decimal part. The decimal part can be divided into exponents and fractional mantissas: x = A2BIs. Taking the base 2 logarithm, logx = logA + B. Since A is a decimal part, it is realistic to find its logarithm by referring to a look-up table. The index B can be found from the position of the MSB (s)aAnd sb bBecause both are positive numbers).

[0074] Therefore, the calculation can be reduced so that only addition and subtraction arithmetic operations are required. When using this method, v1= v2The limit value should also be recalculated using = N. In practice, the significance level can be set preferably at p = 0.05 for a particular application.

[0075] For example, the standard deviation, using the skew, various moments, histograms, and other calculations known conventionally without departing from the spirit of the present invention, minute are possible using a variety of computational methods dispersion That will be understood by those skilled in the art.

[0076] In the first alternative embodiment of the present invention, the above method is used by using the real and imaginary parts of the signal instead of the absolute value. This embodiment achieves the economics of hardware.

[0077] In the second alternative embodiment of the present invention, the parameter of n in Eq. (11) is optimized. At the end of the guard interval, the two blocks straddle more than the transition to the effective interval, giving a clear increase during variance. The use of any value of n> 2 has the disadvantage that some contiguous points give a significant increase as subsequent blocks move to the boundary. This small problem can be easily overcome by introducing an invalid period after the detection of the boundary. That is, once a spike is detected, a set of samples of the same size as the FFT window is received before any attempt is made to place another spike. The invalid period has the additional benefit of not introducing false spikes. When using a larger value of n, H0The noisy F-signals remain nearly identical, but spikes 66 and 61 (Fig. 7) increase.

【0078】Example 2The maximum F-spike height as a function of n can be systematically measured with background changes in F. The results are shown in Table 2.

[0079] [Table 2]<img file="JP4142758B2_D0020.tif" />Table 2 was created using the first 5 frames of the signal analyzed in Figure 7. The statistics in columns (2) and (3) of Table 2 were performed by excluding all points where F> = 3.0 in order to exclude spikes from the calculation. Otherwise, the spikes will affect the mean and standard deviation, even if they are from different statistical populations.

The result is F, which is a background change of F.sdIs affected by n and asymptotically increases to a value of about 0.28. This seems to be the effect of overlapping blocks. For example, for N = 64 and n <64, the blocks for which the variance is calculated will contain several identical values and will therefore be correlated. To test this law, FsdIs evaluated for n> N and the results are shown in Table 3.

[0082] [Table 3]<img file="JP4142758B2_D0021.tif" />[0083] For n> = N / 2, the dependency is linear. This dependency is reduced if F is calculated per n samples instead of per sample. However, this raises the risk that for small guard sections, the first block is not completely within the guard section and the second block is completely outside the effective section.

A third alternative embodiment of the present invention is shown in FIG. 10, which schematically illustrates a timing synchronization circuit 70. The circuit receives a complex input signal 72 and has a circuit module 74 that makes the absolute value of that input, which is obtained from node 83. Circuit module 74 ensures that the value being subsequently processed is an unsigned number. The input to the circuit module 74 is a difference signal created by the subtractor 75, which takes the input signal 72 and the delayed version of the input signal 72 processed by the delay circuit 79 as inputs. The delay circuit 79 is preferably implemented as a FIFO 77 of length L, where L is the size of the FFT window. As mentioned above, this circuit can be operated even when the input signal 72 is a real, imaginary, complex, or complex absolute value. If the input signal 72 is real or imaginary, the circuit module 74 is transformed into any known circuit that removes the sign of the output of the subtractor 75, or the sign is set equivalently and the output accumulates monotonically. That is, the circuit can have a unipolar output. The output of circuit module 74 is finally clocked into a digital delay circuit, the delay circuit preferably running as FIFO 78. When the FIFO 78 is full, the signal SIG1 80 is output and the output of the FIFO 78 is available as indicated by the AND gate 82. Further, the addition / subtraction circuit 84 is connected to the node 76, and its output is stored in the register 86. The delayed version of the output of the add / subtract circuit 84 is obtained from register 86 and fed back to line 88 as the second input of the add / subtract circuit 84. When the signal SIG1 80 is output, the output version of circuit module 74, delayed by the first predetermined interval N, which is the number of samples in the comparison block, is subtracted from the signal on node 76.

The signal on line 88 is an index into a look-up table, preferably executed by read-only-memory (ROM) and shown as ROM 90. The address of ROM90 contains the base 2 logarithm of the amplitude of the signal on line 88, which then appears at node 92. Node 92 is connected to the subtractor 94 and is connected to a delay circuit designated as FIFO98, which delay circuit is used to make the denominator of the central term of equation (17).

[0086] The subtractor 94 is a predetermined threshold value F in the comparison circuit 106.LIMITIt produces a signal that is compared to a logarithm with a base of 2, and the comparison circuit is shown as an adder 108 connected to the comparator 110 for simplicity. The output signal SYNC112 is output when the position of the boundary of the guard section is detected.

Although not implemented in the current preferred embodiment, the size of the FIFO 77 can be dynamically configured to allow the size of the section being evaluated to be adjustable according to the operating state. This can be conveniently done by storing the values on the node 92 in RAM 114 to calculate the variance of those values.

[0088] In a fourth alternative embodiment described with reference to FIG. 11, configurations similar to those of the embodiment shown in FIG. 10 have the same reference numerals. The timing synchronization circuit 116 is similar to the timing synchronization circuit 70, except that the delay circuit 79 is implemented here as the FIFO 77 and the other FIFO 100 selected by the multiplexer 102. Both FIFOs 77 and 100 provide the same delay, but their capacities are different. The FIFO100 can store samples taken during a section equal to the size of the FFT window, and is usually the whole to detect the boundary position of the guard section during the first mode of operation, eg channel acquisition.symbolIs selected when it is necessary to evaluate. In the above European telecommunications standards, data storage up to 8K is required along with commensurate resource requirements. In the subsequent operation, the approximate position of the boundary of the guard section is ahead.symbolIt can be known from the history of. In the second mode of operation, therefore, it is only necessary to evaluate a fairly small section to determine the boundary position of the guard section. The number of samples used to calculate the variance can be maintained at a small number, preferably 32 or 64, so a fairly small FIFO 77 is chosen to hold the calculated value. The resources saved thereby can be used for other functions in the modulator, and the memory used by the large FIFO 100 can be reassigned for other purposes.

[0089] The control block 81 is continuous.symbolIn the data stream ofsymbolIt can optionally accelerate the evaluation interval for boundaries and can also be used to delay for invalid periods. After all, the moving evaluation interval is the currentsymbolAcross the boundaries of the guard section of, then synchronization is determined. The size of the evaluation interval is chosen to be large enough to minimize memory usage and still achieve statistical significance during the evaluation interval. The evaluation interval and the size of the FIFO 77 can be statistically or dynamically configured.

Although the present invention has been described with reference to the structures disclosed herein, it is not limited to the details described and the present application is made in any modification or modification within the scope of the following claims. Intended to cover.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 shows spectra of COFDM subchannels.

FIG. 2 shows the frequency spectrum of multiple carriers in a COFDM signal.

FIG. 3 is a diagram of signals according to COFDM and data.symbolIndicates the format.

FIG. 4 is a block diagram of a COFDM system based on an FFT.

FIG. 5 shows a perturbation in the COFDM signal sequence.

FIG. 6 is a flow chart of a timing synchronization method according to a preferred embodiment of the present invention.

FIG. 7: Some data on coarse timing synchronizationsymbolIt is a plot of the F ratio test performed on.

FIG. 8 is a plot of incomplete beta function for different degrees of freedom.

FIG. 9 is a plot useful for understanding the statistical significance test of the present invention.

FIG. 10 is an electrical schematic of a synchronous circuit according to an alternative embodiment of the present invention.

FIG. 11 is an electrical schematic of a synchronous circuit according to another alternative embodiment of the present invention.

43 members in 17 offices

This recordMember, by publication datePriority claim7 more offices are in the list below

Priority claims5

Priority claims
Document	Office	Kind	Date
9622728	United Kingdom	A	1996-10-31
9622728	United Kingdom	A	1996-10-31
96227285	United Kingdom	–	1996-10-31
19969622728	–	–	–
GB19960022728	–	–	–

Members43

Family members, oldest first
Document	Office	Kind
GB9622728D0	United Kingdom	D0	1997-01-08
GB9720550D0	United Kingdom	D0	1997-11-26
IL122020A0	Israel	A0	1998-03-10
IL122020D0	Israel	D0	1998-03-10
CA2199111A1	Canada	A1	1998-04-30
EP0840485A2	European Patent Office (EPO)	A2	1998-05-06
GB2318953A	United Kingdom	A	1998-05-06
AU3991697A	Australia	A	1998-05-07
CA2270149A1	Canada	A1	1998-05-07
WO9819410A2	World Intellectual Property Organization (WIPO)	A2	1998-05-07
CN1181673A	China	A	1998-05-13
AU5147198A	Australia	A	1998-05-22
JPH10150428A	Japan	A	1998-06-02
EP0840485A3	European Patent Office (EPO)	A3	1998-06-10
ID19373A	Indonesia	A	1998-07-02
KR19980033375A	Republic of Korea	A	1998-07-25
WO9819410A3	World Intellectual Property Organization (WIPO)	A3	1998-08-27
SG55399A1	Singapore	A1	1998-12-21
TW363321B	Taiwan Province of China	B	1999-07-01
EP0938795A2	European Patent Office (EPO)	A2	1999-09-01
US5953311A	United States of America	A	1999-09-14
BR9712722A	Brazil	A	1999-10-26
AR008906A1	Argentina	A1	2000-02-23
AR008910A1	Argentina	A1	2000-02-23
IL129655A0	Israel	A0	2000-02-29
IL129655D0	Israel	D0	2000-02-29
CN1249099A	China	A	2000-03-29
KR20000052956A	Republic of Korea	A	2000-08-25
EP0840485B1	European Patent Office (EPO)	B1	2000-09-06
AT196218T	Austria	T	2000-09-15
ATE196218T1	Austria	T1	2000-09-15
DE69703016D1	Germany	D1	2000-10-12
IL122020A	Israel	A	2000-11-21
AU727726B2	Australia	B2	2000-12-21
DE69703016T2	Germany	T2	2001-04-05
AU737361B2	Australia	B2	2001-08-16
TW465234B	Taiwan Province of China	B	2001-11-21
JP2001527706A	Japan	A	2001-12-25
US6359938B1	United States of America	B1	2002-03-19
US2003142764A1	United States of America	A1	2003-07-31
US6687315B2	United States of America	B2	2004-02-03
JP2005045788A	Japan	A	2005-02-17
JP4142758B2This record	Japan	B2	2008-09-03

23 legal events, as the office reported them to INPADOC

Over the term

Point at a mark for the event

Events

Legal events, newest first
	Event	Code
2017-10-21	Cancellation because of completion of termEXPY	EXPY
2017-06-06	Receipt of annual feesR250	R250
2016-05-31	Receipt of annual feesR250	R250
2015-06-02	Receipt of annual feesR250	R250
2014-06-03	Receipt of annual feesR250	R250
2012-06-12	Renewal fee payment (event date is renewal date of database)FPAY	FPAY
2011-06-14	Renewal fee payment (event date is renewal date of database)FPAY	FPAY
2008-06-20	Renewal fee payment (event date is renewal date of database)FPAY	FPAY
2008-06-20	Certificate of patent or registration of utility modelR150	R150
2008-06-19	First payment of annual fees (during grant procedure)A61	A61
2008-06-05	Written permission of extension of timeA602	A602
2008-05-31	Written request for extension of timeA601	A601
2008-05-08	Written decision to grant a patent or to grant a registration (utility model)A01	A01
2008-05-01	Written decision to grant a patent or to grant a registration (utility model)A01	A01
2008-04-17	Decision of grant or rejection writtenTRDD	TRDD
2007-12-13	Written submission of copy of amendment under article 19 pctA524	A524
2007-09-18	Written permission of extension of timeA602	A602
2007-09-12	Written request for extension of timeA601	A601
2007-06-13	Notification of reasons for refusalA131	A131
2007-03-27	Request for written amendment filedA521	A521
2006-12-26	Notification of reasons for refusalA131	A131
2006-09-25	Report on retrievalA977	A977
2004-09-16	Written request for application examinationA621	A621