Multiscale wireless communication
Abstract
This record has no abstract on file.
Term
Term ended
Expired 18 February 2025, 1.6 years ago.
- Priority
- Filed
- Granted
- Expired
- Today
50 claims: 13 independent, 37 dependent
- 1変調器を備える送信機であって、 前記変調器は、少なくとも1つのパラメータセットを用いてウェーブレット変換を決定するように構成され、ただし前記パラメータセットは複数のウェーブレットから1つを選択するように構成され、 前記変調器は、前記ウェーブレット変換を用いて入力信号を変調出力信号に変調するように構成され、ただし前記変調出力信号は、無線チャネルを介する送信に適しており、 前記変調器は、フィードバック経路から受信した情報に少なくとも部分的に基づいて、推定したチャネル行列を変換するために用いた前記パラメータセットの少なくとも一部分を受信するように構成され、ただし前記情報は、前記パラメータセットの前記少なくとも一部分の中のパラメータ値を量子化したものに対応し、 前記変調器はさらに、前記推定したチャネル行列を受信機が形成するために使用されるトレーニングデータを決定するように構成され、 前記送信機は、前記変調出力信号と前記トレーニングデータとを前記無線チャネルへと送信するように構成される、送信機。
- 2前記ウェーブレット変換が、P個のダイレーションに対応する表現を有する、請求項1に記載の送信機であって、ただし前記ダイレーションの数は、再サンプリング段階の数と同じである、送信機。
- 3前記ウェーブレット変換が、変更されたパラメータセットを少なくとも部分的に用いて決定されたK点のウェーブレット変換を含む、請求項1に記載の送信機。
- 4前記変調出力信号がパラレルデータセットを含み、前記送信機が、 前記入力信号を受信するように構成されている入力ポートと、 前記パラレルデータセットを、前記チャネルを介して伝送するのに適するシリアルデータセットに変換するパラレル‐シリアル変換器とを更に含む、請求項1に記載の送信機。
- 5前記チャネルを介して前記シリアルデータセットを伝送する送信アンテナを更に含む、請求項4に記載の送信機。
- 6前記入力信号に対応する入力データを表すベクトルを、前記ウェーブレット変換を表す、前記推定したチャネル行列と乗算するように前記変調器が更に構成されている、請求項1に記載の送信機。
- 7複数のウェーブレットの1つを選択するように構成されるパラメータセットを少なくとも用いてウェーブレット変換を決定することと、 前記ウェーブレット変換を用いて、入力信号を、チャネルを介する伝送に適する変調出力信号に変調することと、 推定したチャネル行列を変換するために用いた前記パラメータセットの少なくとも一部分を、フィードバック経路から受信した情報に少なくとも部分的に基づいて、受信すること、ただし前記情報が、前記パラメータセットの前記少なくとも一部分の中のパラメータ値を量子化したものに対応する、前記受信することと、 前記推定したチャネル行列を受信機が形成するために使用されるトレーニングデータを決定することと、を含む方法。
- 8前記ウェーブレット変換が、P個のダイレーションに対応する表現を有する、請求項7に記載の方法であって、ただし前記ダイレーションの数は、再サンプリング段階の数と同じである、方法。
- 9前記ウェーブレット変換が、変更されたパラメータセットを用いて少なくとも部分的に決定されたK点のウェーブレット変換を含む、請求項7に記載の方法。
- 10前記変調出力信号がパラレルデータセットを含み、前記方法が更に、 前記入力信号を受信することと 受信された入力信号を前記パラレルデータセットから、前記チャネルを介する伝送に適するシリアルデータセットに変換することとを含む、請求項7に記載の方法。
- 11前記チャネルを介して前記シリアルデータセットを伝送することを更に含む、請求項 10 に記載の方法。
- 12前記変調することが、前記入力信号に対応する入力データを表すベクトルを、前記ウェーブレット変換を表す行列と乗算することを更に含む、請求項7に記載の方法。
- 13コンピュータにより実行されると、 該コンピュータに、 請求項7~12 のいずれかに記載の方法を実行させるように構成されるプログラムコードを含む、 コンピュータプログラム。
- 14複数のウェーブレットの1つを選択するように構成されるパラメータセットを少なくとも用いてウェーブレット変換を決定する手段と、 前記ウェーブレット変換を用いて、入力信号を、チャネルを介する伝送に適する変調出力信号に変調する手段と、 推定したチャネル行列を変換するために用いた前記パラメータセットの少なくとも一部分を、フィードバック経路から受信した情報に少なくとも部分的に基づいて変更する手段、ただし前記情報が、前記パラメータセットの前記少なくとも一部分の中のパラメータ値を量子化したものに対応する、前記手段と、 前記推定したチャネル行列を受信機が形成するために使用されるトレーニングデータを決定する手段と、を備える装置。
- 15前記ウェーブレット変換が、変更されたパラメータセットを用いて少なくとも部分的に決定されたK点のウェーブレット変換を含む、請求項14に記載の装置。
- 16前記変更する手段が変更操作を実行する前に、前記変調する手段が、元のパラメータセットを少なくとも用いて決定されたウェーブレット変換を用いて、テストデータを含む入力信号を変調するように更に構成され、前記変更する手段が、前記受信された情報に少なくとも部分的に基づいて前記元のパラメータセットの少なくとも一部分を変更するように更に構成され、前記変調する手段が、変更された元のパラメータセットを少なくとも用いて決定されたウェーブレット変換を用いて、他のデータを含む入力信号を変調するように更に構成されている、請求項14に記載の装置。
- 17復調器を備える受信機であって、 前記受信機は、チャネルを介して信号を受信するように構成され、 前記復調器は、複数のウェーブレットの1つを選択するように構成されるパラメータセットにより更に特徴付けられ、 前記受信機は、所定のアルゴリズムを前記受信した信号に適用することによりチャネル行列を推定し、前記パラメータセットの少なくとも一部分を用いることにより、前記推定したチャネル行列を変換し、前記パラメータセットから、所定の基準に最もよく適合する前記少なくとも一部分を選択するように構成され、 さらに前記受信機は、フィードバック経路を介する伝送に適する情報を、前記所定の基準に最もよく適合する前記パラメータセットの前記少なくとも一部分に基づいて生成するように構成される、受信機。
- 18前記ウェーブレット変換が、P個のダイレーションに対応する表現を有する、請求項17に記載の受信機であって、ただし前記ダイレーションの数は、再サンプリング段階の数と同じである、受信機。
- 19前記ウェーブレット変換が、変更されたパラメータセットを少なくとも部分的に用いて決定されたK点のウェーブレット変換を含む、請求項17に記載の受信機。
- 20前記復調器により操作される前記受信された信号がパラレルデータを含み、前記受信機が、受信されたシリアルデータを前記パラレルデータに変換するシリアル‐パラレル変換器を更に含む、請求項17に記載の受信機。
- 21前記チャネルを介して前記シリアルデータのセットを受信する受信アンテナを更に含む、請求項20に記載の受信機。
- 22前記受信された信号に対応する入力データを表すベクトルを、前記ウェーブレット変換を表す変調行列と乗算するように前記復調器が更に構成されている、請求項17に記載の受信機。
- 23前記受信された信号がトレーニングデータを含み、前記チャネル推定値が少なくとも前記トレーニングデータを用いて決定される、請求項17に記載の受信機。
- 24変調器が、前記パラメータの前記少なくとも一部分を変更すると、前記パラメータセットにおいて干渉を最小限に抑えるパラメータを決定するように構成されている、請求項17に記載の受信機。
- 25前記変調器が、前記パラメータの前記少なくとも一部分を変更すると、前記パラメータセットにおいて符号間干渉を最小限に抑えるパラメータを決定するように構成されている、請求項24に記載の受信機。
- 26前記変調器が、前記パラメータの前記少なくとも一部分を変更すると、前記パラメータセットにおいてキャリア間干渉を最小限に抑えるパラメータを決定するように構成されている、請求項24に記載の受信機。
- 27変調器が、前記パラメータの前記少なくとも一部分を変更すると、前記パラメータセットにおいて受信信号エネルギーを最大化するパラメータを決定するように構成されている、請求項17に記載の受信機。
- 28変調器が、前記パラメータの前記少なくとも一部分を変更すると、前記パラメータセットにおいて符号間干渉を最小限に抑え、キャリア間干渉を最小限に抑え、受信信号エネルギーを最大化するパラメータを決定するように構成されている、請求項17に記載の受信機。
- 29前記復調器が、前記パラメータセットの前記少なくとも一部分におけるパラメータの値を量子化するように更に構成されている、請求項17に記載の受信機。
- 30限られた数のビットが前記情報に使用でき、前記限られた数のビットに、量子化した値を適合するため、前記復調器が、前記パラメータセットの前記少なくとも一部分における前記パラメータの値を量子化するように更に構成されている、請求項29に記載の受信機。
- 31複数のウェーブレットの1つを選択するように構成されるパラメータセットを少なくとも用いてウェーブレット変換を決定することと、 チャネルを介して信号を受信することと、 所定のアルゴリズムを前記受信した信号に適用することによりチャネル行列を推定することと、 所定の基準に最もよく適合する前記パラメータセットの少なくとも一部分に基づくウェーブレット変換を用いて、前記チャネル行列を変換することと、 前記所定の基準に最もよく適合する前記パラメータセットの前記少なくとも一部分に基づいて、フィードバック経路を介する伝送に適する情報を生成することとを含む方法。
- 32前記ウェーブレット変換が、P個のダイレーションに対応する表現を有する、請求項31に記載の方法であって、ただし前記ダイレーションの数は、再サンプリング段階の数と同じである、方法。
- 33前記ウェーブレット変換が、変更されたパラメータセットを用いて少なくとも部分的に決定されたK点のウェーブレット変換を含む、請求項31に記載の方法。
- 34前記受信された信号がパラレルデータを含み、前記方法が、受信されたシリアルデータを前記パラレルデータに変換することを更に含む、請求項31に記載の方法。
- 35前記チャネルを介して前記シリアルデータのセットを受信することを更に含む、請求項 34 に記載の方法。
- 36前記復調することが、前記受信された信号に対応する入力データを表すベクトルを、前記ウェーブレット変換を表す変調行列と乗算することを更に含む、請求項31に記載の方法。
- 37前記受信された信号がトレーニングデータを含み、前記チャネル推定値が前記トレーニングデータを用いて決定される、請求項31に記載の方法。
- 38前記変更することが、前記パラメータセットにおいて干渉を最小限に抑えるパラメータを決定することを更に含む、請求項31に記載の方法。
- 39前記パラメータを決定することが、前記パラメータセットにおいて符号間干渉を最小限に抑えるパラメータを決定することを更に含む、請求項38に記載の方法。
- 40前記パラメータを決定することが、前記パラメータセットにおいてキャリア間干渉を最小限に抑えるパラメータを決定することを更に含む、請求項38に記載の方法。
- 41前記変更することが、前記パラメータセットにおいて受信信号エネルギーを最大化するパラメータを決定することを更に含む、請求項31に記載の方法。
- 42前記変更することが、前記パラメータセットにおいて符号間干渉を最小限に抑え、キャリア間干渉を最小限に抑え、受信信号エネルギーを最大化するパラメータを決定することを更に含む、請求項31に記載の方法。
- 43前記パラメータセットの前記少なくとも一部分におけるパラメータの値を量子化することを更に含む、請求項31に記載の方法。
- 44限られた数のビットが前記情報に使用でき、前記限られた数のビットに、量子化した値を適合するため、前記量子化することが、前記パラメータセットの前記少なくとも一部分における前記パラメータの値を量子化することを更に含む、請求項 43 に記載の方法。
- 45コンピュータにより実行されると、 該コンピュータに、 請求項31~ 44のいずれかに記載の方法を実行させるように構成されるプログラムコードを含む、 コンピュータプログラム。
- 46複数のウェーブレットの1つを選択するように構成されるパラメータセットを少なくとも用いてウェーブレット変換を決定する手段と、 チャネルを介して信号を受信する手段と、 所定のアルゴリズムを前記受信した信号に適用することによりチャネル行列を推定する手段と、 所定の基準に最もよく適合する前記パラメータセットの少なくとも一部分に基づくウェーブレット変換を用いて、前記チャネル行列を変換する手段と、 フィードバック経路を介する伝送に適する情報を、前記所定の基準に最もよく適合する前記パラメータセットの前記少なくとも一部分に基づいて生成する手段とを含む装置。
- 47前記ウェーブレット変換が、変更されたパラメータセットを用いて少なくとも部分的に決定されたK点のウェーブレット変換を含む、請求項46に記載の装置。
- 48前記受信された信号がテストデータを含み、前記推定したチャネル行列が前記テストデータを用いて決定される、請求項46に記載の装置。
- 49前記変更する手段が、前記パラメータセットにおいて干渉を最小限に抑えるパラメータを決定する手段を更に含む、請求項46に記載の装置。
- 50前記変更する手段が、前記パラメータセットにおいて符号間干渉を最小限に抑え、キャリア間干渉を最小限に抑え、受信信号エネルギーを最大化するパラメータを決定する手段を更に含む、請求項46に記載の装置。
Independent claims50
92 paragraphs, as filed
The present invention relates to the field of wireless communication, and particularly to the field of wave packet systems that use both time domain and frequency domain separation.
Wireless communication involves several forms of signal modulation prior to transmission over the mobile channel. Some examples of the types of processing involved in radiomodulation include temporal processing (eg, spread spectrum), spectral processing (eg, orthogonal frequency division multiplexing or OFDM) and spatial processing (eg, spatiotemporal coding). Be done. This type of processing occurs in one dimension (eg, time, frequency or space) and is sometimes referred to as single-scale modulation. Usually, the type of processing involved is often selected based on the type of mobile channel state experienced. For example, non-diversity mobile channels (eg, single path fading) may be addressed with spatial processing. On the other hand, spectral modulation such as OFDM can be used to better handle mobile channels experiencing multipath (eg, frequency selective fading).
Multiscale modulation involves processing signals in two dimensions, ie, in time and frequency. Therefore, the output of a multiscale modulator is indexed by both the time range and the frequency bin. Conditioning of this type of signal can match a two-dimensional radio channel rather than a one-dimensional one.
Multiscale modulation can be visualized using time-frequency tiling diagrams. Figure 1 shows the sample tiling derived from reference (1) below. Frequency f in this example<sub>0</sub>Sine curve (sinusoid) and impulse t<sub>0</sub>The time-frequency tiling of the waveform consisting of f<sub>0</sub>And t<sub>0</sub>All subband energies are generated in the time-frequency domain that intersects both. Wavelet decomposition can be used to obtain a signal with a time-frequency representation.
Previous studies have addressed the use of wavelet decomposition in digital communications. For example, Wornell (8) developed the concept of fractal modulation for multi-scale communication. In addition, studies such as (9) and (10) have selected optimal wavelet decomposition to account for specific types of channel conditions or transmitter defects. A particular problem with many previous studies in the field of multi-scale communications is that the issue of multipath channel equalization in receivers is often not addressed in particular. This is most likely due to the difficulty of adaptively equalizing channels in two dimensions. Therefore, it is desirable to be able to match a particular wavelet to the instantaneous channel state with minimal interaction (ie, feedback) between the receiver and transmitter. However, previous studies did not challenge utilizing the compact realization of a large wavelet family to match wavelets with radio channel states based on the selection of one or more scalar values. In this study, a parameterized wavelet modulation method has been developed that selects the parameters that best match the radio channel state, based on the well-known compact wavelet decomposition.
Usually, the wavelet decomposition is defined in a continuous region, and the so-called scaling coefficient φ (x) is first derived from the following references (2) and (3).
<maths num="1"><img file="JP4580977B2_D0001.tif" /></maths>
Where {c<sub>k</sub>} Is a real number series. Series {c<sub>k</sub>} Has an even length and must satisfy the following items.
<maths num="2"><img file="JP4580977B2_D0002.tif" /></maths>
{c<sub>k</sub>Another important wavelet characteristic that determines the "smoothness" or continuity of the sequence defined by} is the number of vanishing moments. If the wavelet has M disappearance moments (M 1), the following items can be applied. That is,
<maths num="3"><img file="JP4580977B2_D0003.tif" /></maths>
Next, the corresponding wavelet can be defined as follows.
<maths num="4"><img file="JP4580977B2_D0004.tif" /></maths>
here,
<maths num="5"><img file="JP4580977B2_D0005.tif" /></maths>
Therefore, the dilation and transform of the wavelet function forms an orthonormal basis. That is,
<maths num="6"><img file="JP4580977B2_D0006.tif" /></maths>
Wavelet has compact support, so series {c<sub>k</sub>} Is assumed to have a finite length and a series length of 2N. Then two equal length series {c<sub>k</sub>} And {d<sub>k</sub>You can define the discrete wavelet transform starting with }. You can think of these two series as filters. That is, these series together form a fully reconstructed filter bank.
A parameterized structure of a wavelet with (M N) vanishing moments and a scaling factor for any N value have been proposed in references (3) and (4) below. Now, for the value of N, set the filter factor to {c<sub>k</sub><sup>N</sup>Let's assume that it is displayed as }. N-length wavelet parameter set {α<sub>i</sub>} (-π α<sub>i</sub>Considering <π, 0 i <N), the coefficient {c<sub>k</sub><sup>N</sup>} Is derived by repetition.
<maths num="7"><img file="JP4580977B2_D0007.tif" /></maths>
Parameter set generally [-π, π]<sup>N</sup>The wavelet configuration in reference (7) is limited in the sense that it cannot be defined in scope and still produces wavelets with at least one vanishing moment. However, Pollen said that for any N (at the expense of smoothness) [-π, π]<sup>N</sup>It proved that the wavelet can be defined in the range. To check these types of configurations for a given N, the filter bank matrix F<sub>N</sub>Is defined as follows.
<maths num="8"><img file="JP4580977B2_D0008.tif" /></maths>
Next, the filter bank matrix with N = 1 is
<maths num="9"><img file="JP4580977B2_D0009.tif" /></maths>
The matrix in reference (9) may also be known as the Givens rotation matrix. Similarly, a filter bank matrix with N = 2
<maths num="10"><img file="JP4580977B2_D0010.tif" /></maths>
The N = 2 and N = 3 filter bank equations are sometimes also known as pollen filters because Polen first proposed these two expressions (see reference (7)). Similarly, α<sub>0</sub>If = π / 4, the filter bank matrix F<sub>1</sub>Becomes a Haar matrix, α<sub>0</sub>If = π / 6, then F<sub>2</sub>Is a Daubechies 4-tap filter bank. As N increases, the filter selectivity improves, at the expense of having to determine a large set of parameters to define the wavelet. This is {c<sub>k</sub><sup>3</sup>Can be seen in the equation parameterized to.
<maths num="11"><img file="JP4580977B2_D0011.tif" /></maths>
Two parameters must then be determined before specifying the filter bank. In practice, as the number of coefficients increases, the filter selectivity improves, but the complexity involved in setting the parameters required to form the filter bank increases.
Wavelet decomposition can then be specified for a series of filter banks and resampling steps. Input series a<sub>i</sub>Given (n), the output sequence can be derived as in the process shown in FIG.
In modern technology, filtering is performed digitally using integrated circuits suitable for digital signal processing or computational systems such as general purpose computers.
The number of resampling steps in the wavelet decomposition is sometimes referred to as the number of dilations. This process can also be expressed as a conversion of the input series by a unitary matrix. Input series a in time index i<sub>i</sub>(n) has an (even) length K (ie 0 n <K) and a particular filter bank matrix F<sub>N</sub>Discrete wavelet transform (DWT) matrix T of size K × K for<sub>k</sub>Suppose you have to define. Further assume that the conversion has the desired P dilations. Next, using the above configuration, we can find an iterative method for deriving the transformation matrix. By defining the time scale index as 1 (0 1 P), a K × K filter bank matrix can be defined for each time scale. That is,
<maths num="12"><img file="JP4580977B2_D0012.tif" /></maths>
In equation (12), 0<sub>m × n</sub>Is a zero matrix of m rows x n columns, I<sub>R</sub>Is the identity matrix of R rows x R columns. For each dilation, the permutation matrix Pν (1 ν P) can be defined in the same way. That is,
<maths num="13"><img file="JP4580977B2_D0013.tif" /></maths>
Therefore, for P dilations, the unitary transformation matrix T<sub>K</sub>(P) can be determined as follows.
<maths num="14"><img file="JP4580977B2_D0014.tif" /></maths>
Using this matrix next, the input vector a<sub>i</sub>= [a<sub>i</sub>(0) ... a<sub>i</sub>(K-1)]<sup>T</sup>(I is a sign index) can be modulated. Illustratively, the input signal vector a<sub>i</sub>The elements of are the coefficients of a set of basis functions that represent data, eg speech.
Therefore, the output sequence of such modulation is multiplied by a matrix and a vector x<sub>i</sub>= T<sub>K</sub><sup>T</sup>(P) a<sub>i</sub>Can be formed as a result of. As mentioned above, a single net matrix T<sub>K</sub>(P) represents the wavelet of P inner products in the data. F<sub>N</sub>Is presumed to result in the compact realization of the wavelet, T<sub>K</sub>(P) is the wavelet parameter set {α<sub>i</sub>} Is a function of, i.e. x<sub>i</sub>= T<sub>K</sub>(P, {α<sub>i</sub>}) a<sub>i</sub>Can also be estimated.<nplcit num="1"><text>Wavelets and Subband Coding, by Vetterli, Martin and Jelena Kovacevic, Englewood Cliffs, NJ, Prentice-Hall Inc., 1995</text></nplcit><nplcit num="2"><text>By Gilbert Strang and Truong Nguyen, "Wavelets and Filter Banks", (Wellsley, Massachusetts), Wellesley-Cambridge Press, 1996</text></nplcit><nplcit num="3"><text>"Parameterization of Compactly Supported Orthonormal Wavelets", IEEE Transactions on Signal Processing, Vol. 41, No. 3, March 1993, by Zou, Hehong and Ahmed H. Tewfik, p.1428-1431</text></nplcit><nplcit num="4"><text>"On the Parameterization of the Coefficients of Dilation Equations for Compactly Supported Wavelets" by Schneid, J and S. Pittner, Computing, Vol. 51, May 1993, p.165-173</text></nplcit><nplcit num="5"><text>"Pollen Bases and Daubechies-Lagarias Algorithm in MATLAB" by Vidakovic, Brani, Jacket's Wavelets website, http://www.isye.gatech.edu/~brani/datasoft/DL.pdf.</text></nplcit><nplcit num="6"><text>"Analytical Optimization of CQF Filter Banks" by Silva, Vitor and Luis de Sa, IEEE Transactions on Signal Processing, Vol. 44, No. 6, June 1996, p. .1564-1568</text></nplcit><nplcit num="7"><text>Pollen, D, Parameterization of Compactly Supported Wavelets, Aware Inc. technical report AD890503, May 1989</text></nplcit><nplcit num="8"><text>Wornell, Gregory W, "Emerging Applications of Multirate Signal Processing and Wavelets in Digital Communications," Proceedings of the IEEE, Vol. 84, No. 4, April 1996, p.586-603.</text></nplcit><nplcit num="9"><text>"Wavelet Packet Modulation for Orthogonally Multiplexed Communication", IEEE Transactions on Signal Processing, Vol. 45, No. 5, May 1997, p.1336-1339, by Lindsey, Alan R.</text></nplcit><nplcit num="10"><text>Wavelet Packet Division Multiplexing and Wavelet Packet by KM Wong (Wong, KM), Jay Wu (Wu, J), TN Davidson (TN) and Jin, Q. Design under Timing Error Effects , IEEE Transactions on Signal Processing, Vol. 45, No. 12, December 1997, p.2877-2890</text></nplcit>
Summary of invention
The present invention relates to a multi-carrier communication system that adaptively selects a set of wavelets that match the channel state.
A feature of the present invention is the use of compact parameterization that allows the generation of an infinite number of wavelets and a scaling filter pair that uses a finite set of parameters such as filter parameters and dilation numbers.
Another feature of the present invention is to remove the equalizer in the receiver by changing the wavelet parameters to pre-compensate for changes in channel state.
Detailed explanation
Input vector a<sub>i</sub>= [a<sub>i</sub>(0) ... a<sub>i</sub>(K-1)]<sup>T</sup>Starting with the output vector x, similar to an OFDM system<sub>i</sub>= T<sub>K</sub><sup>T</sup>(P, {α<sub>i</sub>}) a<sub>i</sub>Wavelet-based transmission systems can be formed. However, unlike OFDM systems where simple equalization structures can be used by using the periodic convolution characteristics of the underlying discrete Fourier transform modulation, wavelet-based systems are sensitive to frequency-selective radio channels. Therefore, the equalization problem with wavelets can be complicated.
Output vector x<sub>i</sub>Is transmitted continuously. L tap channel vector h<sub>i</sub>= [h<sub>i</sub>(0) ... h<sub>i</sub>(L-1)]<sup>T</sup>(h<sub>i</sub>Can represent a radio channel by means of a unit norm (which is considered to have a unit norm).<sub>i</sub>Can be expressed as follows (assuming L <K).
<maths num="15"><img file="JP4580977B2_D0015.tif" /></maths>
In equation (15), n<sub>i</sub>(k) is an additional noise term. A typical equalization technique for this type of received signal is decision aid equalization, which takes into account intersymbol interference (ISI), so that the previously transmitted code a<sub>i-1</sub>To take an estimate of and take into account intercarrier interference (ICI) x<sub>i</sub>Requires an estimate of the individual inputs of. This equation can also be expressed in matrix-vector format. That is,
<maths num="16"><img file="JP4580977B2_D0016.tif" /></maths>
In equation (16), H<sub>i</sub>, H<sub>ISI_i</sub>And n<sub>i</sub>Is expressed as follows.
<maths num="17"><img file="JP4580977B2_D0017.tif" /></maths>
<maths num="18"><img file="JP4580977B2_D0018.tif" /></maths>
<maths num="19"><img file="JP4580977B2_D0019.tif" /></maths>
Instead, we propose to pre-equalize the channels using wavelet compact parameterization. In other words, a given filter bank matrix F<sub>N</sub>If there is an estimate of a set of parameters that maximizes the quality of the received signal for, then this information is used in the modulation matrix T.<sub>K</sub>Can be changed. Channel estimate at receiver ^ h<sub>i</sub>Assuming that there is sufficient training data to form, additional training data and a K × K channel estimation matrix can be used to find the best wavelet parameter set. That is,
<maths num="20"><img file="JP4580977B2_D0020.tif" /></maths>
It is clear that applying unitary transformation does not reduce the output of additional Gaussian noise. However, unitary transformations such as DWT can be applied to minimize the two sources of interference found in frequency selection channels, ICI and ISI, with multicarrier systems. Using this channel estimation matrix, we can find the optimal wavelet parameter set that minimizes ICI for a given P as follows.
<maths num="21"><img file="JP4580977B2_D0021.tif" /></maths>
Additional criteria can be established to minimize ISI. This involves selecting a wavelet transform that allows the adverse effects of ISI from one wavelet code to the next to be negligible. We define a K × K channel matrix based on the adverse effects of ISI apparent in Eq. (15). That is,
<maths num="22"><img file="JP4580977B2_D0022.tif" /></maths>
Therefore, another optimization criterion that minimizes ISI can be considered as follows. That is,
<maths num="23"><img file="JP4580977B2_D0023.tif" /></maths>
In addition to the selection of wavelets that minimize ICI and ISI, the wavelets themselves cannot be considered nearly equalized unless the energy generated along the resulting diagonal is maximized. As a result, the received signal energy is directly maximized. The criteria for maximizing the energy generated along the diagonal (assuming Λ is a diagonal operator) are as follows: That is,
<maths num="24"><img file="JP4580977B2_D0024.tif" /></maths>
It cannot be assumed that a single parameter set minimizes ISI and ICI while maximizing the channel energy recovered. Therefore, wavelet parameter selection should be based on criteria that minimize all residual interference.
<maths num="25"><img file="JP4580977B2_D0025.tif" /></maths>
Finally, the wavelet transmission method is as shown in FIG.
In summary, the processing according to the present invention is as follows.
Start with an estimated channel matrix and an initial parameter set α that specifies the initial wavelet (assuming P is fixed).
The training signal is transmitted from the transmitter (base station) to the receiver.
The receiver repeats (or otherwise calculates) the values of the wavelet parameters that are adjusted to minimize ICI, ISI and total residual interference.
The adjusted parameters are returned to the transmitter along the feedback path.
The adjusted parameters are used for transmission during the next period until the next adjustment.
It should be noted here that residual interference may still be present after the wavelet selection. This results from effects such as feedback delays on channel coherence time, reduced parameter search space, and so on. As a result, even if the wavelet best matches the channel state, some limited form of interference removal may still be required. Furthermore, the matrix of Eq. (22) does not maximize the diversity seen in the system, that is, only optimizes ICI and ISI.
The receiver must provide the transmitter with information about the channel state, that is, how to minimize the throughput required to relay the information needed for accurate waveform selection from the receiver to the transmitter. In adaptive waveform design for wireless transmission in systems that must, the design of parameterized wavelets poses a typical problem. This is usually not a problem in time-divided dual (TDD) systems. The reason is that it is generally assumed that the TDD transmitter can estimate the channel state seen in the TDD receiver without feedback. However, paired-band systems generally do not have sufficient interrelationships between the transmission and reception frequencies used by a given transmitter / receiver. Therefore, it is important to minimize the amount of information relayed by the receiver to the transmitter for waveform selection.
However, the wavelet filter bank matrix selectivity improves significantly as the 2N digit increases. This requires a large amount of parameterization and therefore has the potential to increase the amount of feedback information required. Therefore, given the maximum feedback data payload R (bits), there is a trade-off between increasing filter bank selectivity and decreasing feedback parameter quantization noise. Parameter set {α<sub>i</sub>Assuming that each value of} is uniformly quantized between 0 and π, the average quantization error can be found as follows.
<maths num="26"><img file="JP4580977B2_D0026.tif" /></maths>
In equation (23), the quantization error is derived from the typical result for uniform quantization of uniformly distributed random variables. Unfortunately, due to the throughput limitation on the feedback of wavelet parameter selection, this variation in error does not provide sufficient insight into the degradation of wavelet-based communication systems. However, it is possible to indicate the optimal filter bank order for which a particular feedback limit is given. Quantization noise for filter orders of 2 and 4<u style="single">Figure 3</u>Shown in. In the left region of the curve, the quantization noise increases to unacceptable levels as the filter order increases. However, in the right region of the curve, the quantization noise is at an unacceptable level as the filter order increases (although the quantization noise still worsens as the filter order increases for any R value). It tends to go down. This is not surprising. The reason is that it makes sense that an increase in the throughput payload allows an increasing number of wavelet parameters to be accurately quantized and, therefore, a higher order filter bank.
For a given R value, it makes sense that the filter order to be selected is the maximum order below this level, so the related problem constitutes what constitutes the "unacceptable level" of quantization noise. Is it done?
FIG. 2 shows a process of forming a coefficient that generates a predetermined wavelet. On the left, a set of signal coefficients a representing voice or other data<sub>i</sub>Is input and processed (eg, by a general purpose computer system) to generate the final coefficients that define the particular wavelet used by the transmitter.
During operation, in the system according to the invention, the transmitter periodically transmits a reference signal (training sequence) to the receiver. The receiver applies a known algorithm for estimating the channel matrix to the received signal. The receiver then transforms the channel matrix using various experimental parameters and selects the parameters that give the result of optimizing the criteria as represented by equation (22), which minimizes ICI and ISI. To do.
The "optimized" parameter is relayed to the transmitter along a feedback channel with a limited number of bits.
The transmitter then prepares a packet with the parameters transmitted by the receiver to the transmitter during a period of time.
To test the feasibility of adapting wavelets to instantaneous channel states based on parameterization, we tested several different types of wavelet decomposition for specific channel profiles. The transmission system was tested at a wavelet code rate of 250 kHz, assuming that an input vector of magnitude 32 was modulated by a square wavelet transform matrix (in this case, the code rate is 32 wavelets resulting from a single input vector. All of the coefficients mean the transmitted rate). BPSK signaling in the 2-tap and 4-tap wavelet models of (9) and (10) in a multipath fading channel (channel tap output profile of [.8.1.1]) with an assumed carrier frequency of 5 GHz and a speed of 3 km / hr. Was inspected for performance. In this system modeling, wavelet parameters were selected for every 50 codes. Under these conditions, the performance of 2-tap and 4-tap wavelets for feedback quantization<u style="single">Figure 4</u>as well as<u style="single">Figure 5</u>Each is shown in.
Under these conditions, as the number of bits increases, the benefit of the natural increase in feedback decreases. In this environment, increased feedback results in 6 dB of quantization denoising, but it is clear that this cannot be converted into the same benefit in terms of overall performance. It should be noted that the results are also graphed for the raw BPSK bit error rate, i.e., additional error correction code may turn into less benefit for an increase in the number of feedback bits. Another thing to note is that the performance of the 2-tap and 4-tap wavelet systems was about the same.
In addition, the number of dilations could in some cases affect the performance of the proposed wavelet transform method. In this case as well, the results of both the 2-tap wavelet and the 4-tap wavelet assuming 3-bit feedback quantization are obtained.<u style="single">Figure 6</u>as well as<u style="single">Figure 7</u>Shown in. Increasing the number of dilations did not increase the benefits of improving system performance. This means that the channel state under consideration is already fairly compact in frequency and time, that is, the greatest performance improvement of this method is the basic wavelet rather than changing the time-frequency resolution of the transmitted signal. Most likely due to the fact that it is to change the filter bank.
To better understand the maximum achievable benefits of multiscale communication, we simulated an OFDM system using exactly the same channel conditions and exactly the same input vector size. Compare with a 2-tap wavelet with 1-bit feedback<u style="single">Figure 8</u>Shown in. It can be seen that the performance improvement by wavelet modulation is as much as 3 dB. Although OFDM systems are orthogonally transmitted, it is clear that the spectral efficiency of wavelet systems has the potential to be much higher than that of OFDM systems, given that wavelet systems do not require orthogonal transmission for BPSK signaling. is there.
Further results were obtained for the adaptation method in the fading channel and compared with the use of the fixation method (in this case, Haar basis and Dubuchi 4 taps). Using a [.8.2] channel tap profile with a code rate of 125 kHz and an input vector size of 64, comparing fixation and adaptation methods under fading conditions with a carrier frequency of 5 GHz and a speed of 150 km / hr. did. We used a rate 1/2 convolutional code, which means that the 32-bit input segment was encoded into 64 binary code (BPSK signaling) input vectors. These results<u style="single">Figure 9</u>as well as<u style="single">Figure 10</u>Shown in.
Both methods appear to converge at low signal-to-noise ratio. This is due to the AWGN overriding the error source in this region of operation, i.e., under such conditions, no adaptation can improve the link. Moreover, it should be noted that both methods result in an error floor. In the case of the fixed method, this is due to the use of suboptimal wavelets for transmission. In the case of adaptive methods, this is in part due to the use of limited feedback. However, even assuming the infinitely accurate quantization of the parameterized space, it should be reiterated that the selected wavelet is still an approximation of the equalization transform.
The above wavelet selection method, in particular the criteria given in Eq. (22), was examined in contrast to the fixation method in the static channel profile of [.5.3.1.1]. In this case, the feedback quantization was increased to 5 bits to accurately check whether the wavelet selection actually selects the best wavelet. In particular, under the assumption of coarse quantization, the possibility of selecting an accurate wavelet using the wavelet selection method has increased. Therefore, the accuracy of this method is not so easy to evaluate under such conditions. 2 Tap method results<u style="single">Figure 11</u>Shown in. It should be noted that the wavelet selection method maintained the gain of the adaptive method as opposed to the use of fixed wavelets. Ten<sup>-4</sup>And 10<sup>-3</sup>It should also be noted that although it is a bit error rate between and, there is still an error floor.
Although the present invention has been described for a limited number of embodiments, those skilled in the art will appreciate that other embodiments may be constructed within the spirit and scope of the claims above.
<u style="single"> Finally, examples of the invention described in the claims on the international filing date are described below.</u><u style="single">[1] Means for receiving input vector signals and</u><u style="single"> A controllable change of variables that modulates the input signal with a transformation having P dilations and a compact parameter representation further characterized by a parameter set, wherein the transforming means has an output parallel dataset. Controllable change of variables and</u><u style="single"> A parallel-serial means for converting the parallel data set into a serial data set,</u><u style="single"> A transmission means for transmitting the serial data set along a channel and</u><u style="single"> A receiving means for receiving the serial data set and</u><u style="single"> Serial-parallel means to convert received serial data into parallel data,</u><u style="single"> With an inverse conversion means that demodulates the parallel data according to the compact parameter representation</u><u style="single">Communication system including.</u><u style="single">[2] The communication system according to [1], wherein the modulation process is achieved by multiplying a vector representing the input data with a modulation matrix representing the transformation.</u><u style="single">[3] The communication system according to [1], wherein the receiving means includes means for processing test data so as to extract parameters that specify the modulation matrix.</u><u style="single">[4] The communication system according to [3], wherein the modulation process is achieved by multiplying a vector representing the input data with a modulation matrix representing the transformation.</u><u style="single">[5] The communication system according to [3], wherein the means for processing test data to extract the parameters is to search for parameters that minimize ISI and residual interference, including ISI.</u><u style="single">[6] The communication system according to [3], wherein the means for processing the test data so as to extract the parameters searches for the parameters that minimize the ICI.</u><u style="single">[7] The communication system according to [3], wherein the means for processing the test data so as to extract the parameters searches for the parameters that minimize the ISI.</u><u style="single">[8] Means of receiving input vector signals and</u><u style="single"> A controllable change of variables that modulates the input signal with a transformation having P dilations and a compact parameter representation further characterized by a parameter set, wherein the transforming means has an output parallel dataset. Controllable change of variables and</u><u style="single"> A parallel-serial means for converting the parallel data set into a serial data set,</u><u style="single"> A transmission means for transmitting the serial data set along a channel and</u><u style="single"> A receiving means for receiving the serial data set and</u><u style="single"> Serial-parallel means to convert received serial data into parallel data,</u><u style="single"> An inverse conversion means that demodulates the parallel data according to the compact parameter representation, and</u><u style="single"> With an adjusting means that adjusts the parameter set according to changes in the channel state</u><u style="single">Communication system including.</u><u style="single">[9] The communication system according to [8], wherein the modulation process is achieved by multiplying a vector representing the input data with a modulation matrix representing the transformation.</u><u style="single">[10] The communication system according to [8], wherein the receiving means includes means for processing test data so as to extract parameters that specify the modulation matrix.</u><u style="single">[11] The communication system according to [10], wherein the modulation process is achieved by multiplying a vector representing the input data with a modulation matrix representing the transformation.</u><u style="single">[12] The communication system according to [10], wherein the means for processing test data to extract the parameters is to search for parameters that minimize ISI and residual interference, including ISI.</u><u style="single">[13] The communication system according to [10], wherein the means for processing the test data to extract the parameters is to search for the parameters that minimize the ICI.</u><u style="single">[14] The communication system according to [10], wherein the means for processing the test data to extract the parameters is to search for the parameters that minimize the ISI.</u><u style="single">[15] A method of operating a communication system, wherein the method is</u><u style="single"> The step of receiving the input vector signal and</u><u style="single"> In a controllable variable transforming means, a step of modulating the input signal using a transform having P dilations and a compact parameter representation further characterized by a parameter set, wherein the transforming means outputs parallel data. Steps with sets and</u><u style="single"> Steps to convert the parallel dataset to a serial dataset,</u><u style="single"> The step of transmitting the serial data set along the channel and</u><u style="single"> The step of receiving the serial data set and</u><u style="single"> Steps to convert the received serial data to parallel data,</u><u style="single"> With the step of demodulating the parallel data according to the compact parameter representation</u><u style="single">How to include.</u><u style="single">[16] The method of [15], wherein the receiving means processes the test data so as to extract parameters that specify the modulation matrix.</u><u style="single">[17] The method of [15], wherein the modulation process is accomplished by multiplying a vector representing the input data with a modulation matrix representing the transformation.</u><u style="single">[18] The method of [15], wherein the steps of processing the test data to extract the parameters search for parameters that minimize ISI and residual interference, including ISI.</u><u style="single">[19] The method according to [15], wherein the step of processing test data to extract parameters searches for parameters that minimize ICI.</u><u style="single">[20] The method of [15], wherein the step of processing the test data to extract the parameters finds the parameters that minimize the ISI.</u><u style="single">[21] A method of operating a communication system, wherein the method is</u><u style="single"> The step of receiving the input vector signal and</u><u style="single"> In a controllable variable transforming means, a step of modulating the input signal using a transform having P dilations and a compact parameter representation further characterized by a parameter set, wherein the transforming means outputs parallel data. Steps with sets and</u><u style="single"> Steps to convert the parallel dataset to a serial dataset,</u><u style="single"> The step of adjusting the parameter set according to the change of the channel state, and</u><u style="single"> The step of transmitting the serial data set along the channel and</u><u style="single"> The step of receiving the serial data set and</u><u style="single"> Steps to convert the received serial data to parallel data,</u><u style="single"> With the step of demodulating the parallel data according to the compact parameter representation</u><u style="single">How to include.</u><u style="single">[22] The method according to [21], wherein the receiving means processes the test data so as to extract a parameter that specifies the modulation matrix.</u><u style="single">[23] The method of [21], wherein the modulation process is accomplished by multiplying a vector representing the input data with a modulation matrix representing the transformation.</u><u style="single">[24] The method of [21], wherein the steps of processing test data to extract parameters search for parameters that minimize ISI and residual interference, including ISI.</u><u style="single">[25] The method according to [21], wherein the step of processing test data to extract parameters searches for parameters that minimize ICI.</u><u style="single">[26] The method according to [21], wherein the step of processing test data to extract parameters searches for parameters that minimize ISI.</u><u style="single">[27] A product comprising a computer-readable program storage medium that embodies instructions that can be executed by the computer in order to perform steps of the method in which the medium operates the communication system. But,</u><u style="single"> The step of receiving the input vector signal and</u><u style="single"> In a controllable variable transforming means, a step of modulating the input signal using a transform having P dilations and a compact parameter representation further characterized by a parameter set, wherein the transforming means outputs parallel data. Steps with sets and</u><u style="single"> Steps to convert the parallel dataset to a serial dataset,</u><u style="single"> The step of transmitting the serial data set along the channel and</u><u style="single"> The step of receiving the serial data set and</u><u style="single"> Steps to convert the received serial data to parallel data,</u><u style="single"> With the step of demodulating the parallel data according to the compact parameter representation</u><u style="single">Products including.</u><u style="single">[28] The method of [15], wherein the receiving means processes the test data so as to extract parameters that specify the modulation matrix.</u><u style="single">[29] The method of [15], wherein the modulation process is accomplished by multiplying a vector representing the input data with a modulation matrix representing the transformation.</u><u style="single">[30] The method of [15], wherein the steps of processing the test data to extract the parameters search for parameters that minimize ISI and residual interference, including ISI.</u><u style="single">[31] The method according to [15], wherein the step of processing test data to extract parameters searches for parameters that minimize ICI.</u><u style="single">[32] The method according to [15], wherein the step of processing the test data to extract the parameters finds the parameters that minimize the ISI.</u><u style="single">[33] A product comprising a computer-readable program storage medium that embodies instructions that can be executed by the computer in order to perform a step of the method in which the medium operates the communication system. But,</u><u style="single"> The step of receiving the input vector signal and</u><u style="single"> In a controllable variable transforming means, a step of modulating the input signal using a transform having P dilations and a compact parameter representation further characterized by a parameter set, wherein the transforming means outputs parallel data. Steps with sets and</u><u style="single"> Steps to convert the parallel dataset to a serial dataset,</u><u style="single"> The step of adjusting the parameter set according to the change of the channel state, and</u><u style="single"> The step of transmitting the serial data set along the channel and</u><u style="single"> The step of receiving the serial data set and</u><u style="single"> Steps to convert the received serial data to parallel data,</u><u style="single"> With the step of demodulating the parallel data according to the compact parameter representation</u><u style="single">Products including.</u><u style="single">[34] The method of [33], wherein the receiving means processes the test data so as to extract parameters that specify the modulation matrix.</u><u style="single">[35] The method of [33], wherein the modulation process is accomplished by multiplying a vector representing the input data with a modulation matrix representing the transformation.</u><u style="single">[36] The method of [33], wherein the steps of processing test data to extract parameters search for parameters that minimize ISI and residual interference, including ISI.</u><u style="single">[37] The method according to [33], wherein the step of processing test data to extract parameters searches for parameters that minimize ICI.</u><u style="single">[38] The method according to [33], wherein the step of processing test data to extract parameters searches for parameters that minimize ISI.</u><u style="single">[39] An input port that receives an input vector signal to be transmitted, and</u><u style="single"> A modulator that modulates the input signal using a transformation having P dilations and a compact parameter representation further characterized by a parameter set, and a modulator that supplies an output parallel dataset.</u><u style="single"> A parallel-serial converter that converts the parallel data set into a serial data set,</u><u style="single"> With a transmitting antenna that transmits the serial data set along the channel</u><u style="single">Transmitter including.</u><u style="single">[40] With a receiving antenna that receives the serial data set over the channel,</u><u style="single"> A serial-parallel converter that converts the received serial data into parallel data,</u><u style="single"> A demodulator that demodulates the parallel data according to a compact parameter representation with P dilations, wherein the representation is further characterized by a parameter set.</u><u style="single">Receiver including.</u><u style="single">[41] The receiver according to [40], further comprising a controller that adjusts the parameter set in response to changes in the state of the channel.</u>
<figref num="1">It is a figure which shows the time-frequency relationship of a wave packet system.</figref><figref num="2">It is a block diagram of a wavelet transmission system.</figref><figref num="3">It is a figure which shows the quantization noise characteristic.</figref><figref num="4">It is a figure which shows the feedback effect of a 2-tap wavelet system.</figref><figref num="5">It is a figure which shows the feedback effect of a 4-tap wavelet system.</figref><figref num="6">It is a figure which shows the dilation effect of a 2-tap wavelet system.</figref><figref num="7">It is a figure which shows the dilation effect of a 4-tap wavelet system.</figref><figref num="8">It is a figure which shows the comparison of BER between a wavelet and an OFDM system.</figref><figref num="9">It is a figure which compares BER in comparison of a 2-tap adaptive wavelet system and a fixed wavelet system.</figref><figref num="10">It is a figure which compares BER in comparison with a 4-tap adaptive wavelet system and a fixed wavelet system.</figref><figref num="11">It is a figure which shows BER in various systems.</figref>
Every citation, both ways
| Document | Relation | Office |
|---|---|---|
| WO02102130A1 | Cites | World Intellectual Property Organization (WIPO) |
| JP2003218831A | Cites | Japan |
| JP2006526370A | Cites | Japan |
| JP2005086444A | Cites | Japan |
12 members in 8 offices
Priority claims9
| Document | Office | Kind | Date |
|---|---|---|---|
| 10788188 | United States of America | – | |
| 78818804 | United States of America | A | |
| 78818804 | United States of America | A | |
| 2005000413 | International Bureau of the World Intellectual Property Organization (WIPO) | W | |
| 2005000413 | International Bureau of the World Intellectual Property Organization (WIPO) | W | |
| 2004788188 | – | – | – |
| 2005000413 | – | – | – |
| US20040788188 | – | – | – |
| WO2005IB00413 | – | – | – |
Members12
| Document | Office | Kind | |
|---|---|---|---|
| US2005185732A1 | United States of America | A1 | |
| WO2005086443A1 | World Intellectual Property Organization (WIPO) | A1 | |
| EP1721432A1 | European Patent Office (EPO) | A1 | |
| KR20070024497A | Republic of Korea | A | |
| CN1943195A | China | A | |
| BRPI0508277A | Brazil | A | |
| JP2007524319A | Japan | A | |
| RU2006132687A | Russian Federation | A | |
| RU2341030C2 | Russian Federation | C2 | |
| KR100904404B1 | Republic of Korea | B1 | |
| US7680208B2 | United States of America | B2 | |
| JP4580977B2This record | Japan | B2 |
12 legal events, as the office reported them to INPADOC
Over the term
Point at a mark for the eventEvents
| Event | Code | |
|---|---|---|
| Cancellation because of no payment of annual feesLAPS | LAPS | |
| Renewal fee payment (event date is renewal date of database)FPAY | FPAY | |
| Certificate of patent or registration of utility modelJAPANESE INTERMEDIATE CODE: R150R150 | R150 | |
| First payment of annual fees (during grant procedure)JAPANESE INTERMEDIATE CODE: A61A61 | A61 | |
| Written decision to grant a patent or to grant a registration (utility model)JAPANESE INTERMEDIATE CODE: A01A01 | A01 | |
| Written decision to grant a patent or to grant a registration (utility model)JAPANESE INTERMEDIATE CODE: A01A01 | A01 | |
| Decision of grant or rejection writtenTRDD | TRDD | |
| Written amendmentJAPANESE INTERMEDIATE CODE: A523A521 | A521 | |
| Notification of reasons for refusalJAPANESE INTERMEDIATE CODE: A131A131 | A131 | |
| Written amendmentJAPANESE INTERMEDIATE CODE: A523A521 | A521 | |
| Notification of reasons for refusalJAPANESE INTERMEDIATE CODE: A131A131 | A131 | |
| Report on retrievalJAPANESE INTERMEDIATE CODE: A971007A977 | A977 |
Numbers
- Publication
- 4580977
- Publication, DOCDB
- 4580977
- Publication, EPODOC
- JP4580977B
- Application
- 2007500311
- Application, DOCDB
- 2007500311
- Application, EPODOC
- JP20070500311
Titles2
- Japanese
- マルチスケール無線通信
- English
- Multi-scale wireless communication
Classification
- CPC, 3
- H04L27/0004
- H04L27/32
- H04L27/00
- IPC, 3
- H04J11 00
- H04B15 00
- H04L27 00