TWI287366B - Blind signal separation using I and Q components - Google Patents

Abstract

A communications device for separating source signals provided by M signal sources includes an antenna array comprising N antenna elements for receiving at least N different summations of the M source signals. A respective in-phase and quadrature module is connected downstream to each antenna element for separating each one of the N different summations of the M source signals received thereby into an in-phase and quadrature component set. A blind signal separation processor forms a mixing matrix comprising at least 2N different summations of the M source signals, with each in-phase and quadrature component set providing 2 inputs into the mixing matrix. The mixing matrix has a rank equal up to 2N. The desired source signals are separated from the mixing matrix by the blind signal separation processor.

Description

1287366 IX. Invention Description: This lx month is related to the field of ^ number processing, and more specifically related to (BSS) technology to mix and separate source signal money. Prior Art A blind ## separation involves recovering a source signal from a combined signal, wherein the combined signal comprises a source of money. The separation is "f 1 2 signal, signal source and conduction frequency, ^ ^ ^ 宁 ^, hanged since the implementation. Μ 于 于 ( 四 四 _ _ _ _ _ _ - - - - - - 例 例 例 例 例 例 例 例 例 例 例 例 例 例 例The familiarity of the "Cocktail Club," effect when all the sound combinations in the room are separated. Blind source separation is particularly applicable to mobile and personal wireless communication devices, where many of the frequency bands that typically coexist in the same spectrum are confusing by many RF transmitters. The problem with co-channel transmitters is expected to only be more efficient with year-on-year low-power, such as the development of unlicensed wireless technologies in Bluetooth and other personal area networks. The three commonly used blind signal separation techniques are principal component analysis (PCΑ), independent component analysis (ICA), and single value decomposition (SVD). The principal component analysis involves the first and second moment statistics of the source signal, and I is used when the signal to noise ratio of the source signal is high. In addition, the independent component analysis is performed using the third and fourth moment statistics of the source signal after the principal component analysis process. Alternatively, the single value decomposition can be used to separate the source signal from the source library number mixture based on its characteristic value. Regardless of the blind signal separation technique being applied, complex sensors are used to receive different source signal mixtures from various sources. Each sensor is a source signal mixture whose output source signal is uniquely summed. Usually, the receiver does not know the channel 1287366 coefficient and the original source signal. The unique summation of the signals is used to assemble the mixing matrix. A suitable blind signal separation technique is then applied to the mixing matrix to separate the expected source signal from the source signal mixture. As an example, U.S. Patent No. 6,799,170 discloses the use of independent component analysis to separate an independent source signal from a source signal mixture. The complex sensor receives the source signal mixture, and the processor takes a sample of the source signal mixture at any time and stores each sample as a data vector to create a data set. Each sensor outputs a source signal mixture that is uniquely summed. The independent component analysis module performs independent component analysis of the resource vector to separate the independent source signals in the source signal mixture from the other signals. The sensors are spatially separated from one another, and the processor creates a data set for each sensor to create a data set. The '170 patent also discloses that the sensor number N is equal to or greater than the source number Μ, that is, the N>M can aggregate the data set. One problem with this implementation is that as the number of sources increases, the number of sensors N also increases. The small portable communication device does not have a large volume to accommodate a large number of sensors, and the installation of the sensor is a problem for the user outside the communication device. U.S. Patent No. 6,931,362 discloses another method of separating a separate signal using a blind signal. The disclosed blind signal separation technique forms a mixing matrix with a weight of the adaptive matrix pen array that minimizes the mean square error caused by the interfering transmitter and Gaussian noise. This blending weight maximizes the signal interference plus the noise ratio. As in the '170 patent, the sensors are also spatially separated from each other, and the number of sensors is equal to or greater than the number of sources to assemble the mixing matrix. Moreover, each sensor provides a single input to the mixing matrix, resulting in a larger available volume for the portable communication device. SUMMARY OF THE INVENTION In view of the above background, it is an object of the present invention to provide a communication device comprising a small antenna array that can receive a mixture of source signals by a blind signal separation technique such that an intended source signal can be separated.

In accordance with this and other objects, features and advantages of the present invention, a source signal provided by a separate source is provided by a communication device that includes an antenna array that can receive different sums of the source signals. A receiver or receiver component is coupled to the antenna array and a blind signal separation processor is coupled to the receiver to form a mixing matrix. The hybrid matrix includes the V received by the antenna array [the source signals are differently summed. The blind signal separation process then separates the expected source signal from the mixing matrix. In addition to using a spatially separated sensor to provide a different source signal for the hybrid matrix, a small antenna array can be used instead. For portable communication devices, blind signal separation techniques can be used because the antenna array provides more than one input to the mixing matrix while still maintaining a tight charge. In particular, signal segmentation can be used to further aggregate the mixing matrix without having to add additional antenna elements to the antenna array. The antenna array may include a Ν antenna element to receive at least the sum of the source signals. When the U-antenna elements are connected downstream by the in-phase and quadrature modules, they are added to the in-phase and quadrature component groups differently from the received Μ__ to Ν. The "receiver component is connected downstream to each of the in-phase and quadrature modules to receive the at least two in-phase and quadrature component groups at least differently. The dead signal separation process H is downstream connected to the (four) component. Forming a mixing matrix comprising at least 2 Ν different sums of the source signals, each in-phase and quadrature component slave 1287366 blind signal separation processor can provide the expected source signal to provide 2 inputs to the mixing matrix number and separate the mixing matrix. The number of antenna elements can be selected. The number of twice the number of antenna elements is equal to the number of sources 疋 2N-M. Alternatively, the number of antenna elements may be two times the number of source signals 'that is 2N> M. Another configuration is when the order of the mixed matrix is equal to K% ' Wherein κ<2Ν' then the blind signal separation processor separates the chirp from the hybrid matrix. The modified antenna element may comprise an associated antenna element. The antenna element may comprise an active antenna element such that the antenna array The two-phase array can be obtained. Alternatively, the solid-associated antenna element can include at least one active antenna 7G component, and at most one passive antenna component, so that the antenna array can be cut. The beam antenna has different differences between the field type and the beam when the source signals are received and received. In the example, the antenna array can form at least one antenna beam to receive at least one different signal of the source signals. In total, each antenna beam system has a maximum ❿ increasing point of the next 3 db point 'can be rejected in at least one direction of the incoming signal. The other-radiation antenna array can form at least - an antenna field type to purely source signals. The N different sums, at least, the at least one antenna field does not substantially have the maximum gain point down to 3db points, and thus no signal rejection is generated in either direction of the incoming signal. The total is linear. The blind signal separation processor can separate the expected source signal from the mixing matrix based on at least one of the principal component analysis 'independent component analysis and single value decomposition. One of the enhancements to the I and Q embodiments also involves The code is based on the 1287366 v 信号 signal segmentation. Different summed signals can be split using the unrolling code. If the summed signal has a k-expansion code, the specific summed signal can be processed to provide correlation. k plus the total signal. Another enhancement to the I and Q embodiments involves an array bias that can receive additional signals for use in the mixing matrix without having to add additional antenna elements. Array biasing involves controlling azimuth and/or elevation directions. Antenna field type. Another enhancement to the I and Q embodiments involves path selection. The path selection is performed so that all source signals used to assemble the mixing matrix are summed to produce an association (first and second). Moment) and/or statistics (third and fourth moments) are independent. The antenna beam is selectively formed such that the incident signal is selectively selected to provide a new sum of the source signals in the mixing matrix instead of being associated and/or Or statistically independent summation. A further feature of the invention relates to a method of operating a communication device for separating a source signal provided by a source of signal. The present invention will now be described in more detail with reference to the accompanying drawings in which, However, the present invention may be embodied in a number of different forms, and is not limited to the embodiments described herein. Of course, these embodiments are provided so that this disclosure will be thorough and complete, and the scope of the invention may be fully conveyed to the skilled person. Similar figures of the figures represent similar elements, while primary elements (9) and (4) designate similar elements in alternative embodiments. In the communication network, there are other sources within the same frequency band that are expected to be used for a particular communication device, which are capable of producing unusable signals, and having source signals intended for operation in the phase communication device. There is also a source of noise, a signal that is communicated but also received by the communication device. In order to facilitate decoding of the source signal, blind signal separation is used to separate the signals received by the arrested device. As mentioned above, the term "blind," is a fact that the signal can be separated without the need to know the nature of the signal or the transition due to the interaction between the signal and the communication channel.

Shih' Any knowledge that can be obtained is usually developed. In this case, the number separation is semi-blind. " Two commonly used blind signal separation techniques are principal component analysis, independent component analysis and single value decomposition. As long as the signals are independent of certain measurable characteristics, and if their signal summations are linearly independent of each other, one or more of these blind signal separation techniques can be used to separate the independent or expected source signal from the source signal mixture. The measurable characteristic is typically a combination of the first, second, third or fourth moments of the amb. • Principal component analysis can whiten the signal, use the first and second moments, and rotate the data set with the associated characteristics. If the signal to noise ratio of the source signal is high, the signal separation process can be stopped by principal component analysis. Right source 彳5 "5虎之彳§ U tiger noise ratio is very low', then independent component analysis can be based on the statistical properties of the third and fourth moments of the source signal to separate the source signal. Since the source signal is Gaussian, the third and fourth moments depend on the first and second moments. As an alternative to independent component analysis and principal component analysis, 'single value decomposition can separate the source signal from the source signal mixture based on its eigenvalue. 12 1287366 A typical scheme is depicted in Figure 1, in which a plurality of signal sources 2 are coupled to a source signal 22. The source signal 22 is transmitted in one direction based on the antenna wave 24 generated in association with each of the signal sources 2A. The complex signal source 20 includes a first signal source 20(1) to a μth signal source 20(Μ). Similarly, each source signal is appended with reference 20(^)40(^^), and the corresponding antenna beam is attached with reference 24(1)_24(M). More direct implementation is typically used in communication networks in the form of omnidirectional antenna fields or directional antenna patterns. The antenna array 32 for the communication device 30 receives the linear combination (mixing) of the source numbers 22 from the signal source 20. Antenna array 32 includes a plurality of antenna elements 34, each antenna element providing at least one linear combination (mixing) of source signals 22 from signal source 2A. The antenna element 34 includes a first antenna element 34(1) to a Nth antenna element 34(n). The received signal source 22(1)-22(M) is initially formed as a hybrid matrix 36. The communication device 3 uses a blind signal separation technique to determine the separation, array 3 δ to separate the source signals in the mixing matrix. The separated signal system is indicated by reference numeral 39. The device 彳 〇 〇 装置 装置 装置 装置 装置 装置 装置 装置 装置 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉 藉The output of each antenna element 34 is summed by the source 22 after the channel pulse / should, i.e., the output of the source 20 and the output = propagation path of the antenna element 34 plus the addition of Gaussian noise. The source signal provided by the No. 4 source ^00)-20(]^) 13 1287366 The communication device 30 will now be discussed in more detail with reference to FIG. The antenna array includes N antenna elements 34(1) seven (n) to receive at least N different sums of m source signals, wherein the mouth M is greater than the work. Antenna array 32 is not limited to any material configuration. The antenna array % may include one or more antenna elements 34. For example, as will be discussed later, the antenna components 34 can be configured to form the antenna array 32 to form a phase array or beam switching antenna. Transceiver 4 is coupled downstream to antenna array 32 to receive at least N different sums of up to one source signal 22. The processor is downstream of the transceiver 40. Even if the processor 42 is greened and transceived, it can be included in the transceiver. The different sums of the M source signals received by the transceiver 40 are used to assemble the matrix 36. The mixing matrix 36 is then processed by the processor or the UI separation processing modules 44, 46 and 48. The blind signal separation processing module includes a principal component analysis module ’-independent component analysis module, group 46 and a single value decomposition module. Some modules 44, 46 » μ π, 丄 can be configured as blind signal separation processing crying Q ju_- \ -a - σπ different force two knife main t amount, the analysis module 44 can be received by the source signal, The brother of (10) operates on the basis of the second moment, and the independent component module 46 can operate based on the third and fourth moments of the same signal; the early value solution & group 48 can be received by the source signal. The signal is separated based on the characteristics of different twists. . The pair of forces, the correlation processing performed by the principal component analysis module 44 can determine the start separation matrix 38(1), and the independent 14 1287366 ★4 - 析人模模 and 46 Next, it may be decided to strengthen the separation matrix 38(2) to separate the composite matrix 3 6 Φ, the original / ancient % # 7 士咕八源心唬. The right k is also determined by the single value decomposition module 48: and the bar code off matrix 38(3) to separate the received source signals in the mixing matrix 36 differently. 10 = each separation matrix 38(1)_38(3), and the separated signal is represented by the 荼39 word. The skin separation signal % # receives the signal analysis of the signal analysis group 5G to determine what signal is useful and what signal is dry

Disturber. The application dependent processing module 52 processes the signal output from the signal analysis module 50. The decision that the signal is useful is not always related to the last decoded signal. For example, the application may need to identify the interferer and take it out of the different sums of the received source signals, and then feed the reduced signal to the mode decoder. In this case, the useful signal is the one that is ultimately rejected. The information fed to the principal component analysis module 44 is the only sum of the signal illusions. Assume that the N linear mixture of independent components Xl, .. Xn is sacrificed to W +...+...屮(〇夂(')=%&(') + ...+ ···~~(,) ~1 Heart (') + + ... Λ - ~ (,) Generally, the transceiver 40 does not know the channel coefficient ajk and the original signal sk. The matrix mark of the upper group equation can be reduced to x = As, where A is a mixed matrix. Model x = As is also known as independent component analysis 15 1287366 • Model. Traditional techniques attempt to find independent component analysis module 46 DS-A'lx. y = W(As) = Wx. Vector y is a measure: : Giant array W, and set. If all signals are inseparable, and the unknown order s sub y=W(As)+Wn = , the dragon is more general type can be the residual noise of other sources. The additional n terms are due to the inability to identify the independent knives. The grading is a productivity. It indicates how the observed data is mixed by the sub-syntax. At the same time, the mixed matrix A is falsely tweeted and cannot be directly observed. \,^ and 5^:;;? The observer is estimated based on X. The starting point of independent component analysis is that the component ^ is statistically independent. Suppose the independent component Sk is There is a Gaussian distribution-- and one of the signals is limited by the Gaussian signal, and the fourth moment is indistinguishable between the Gaussian signals. For ===The matrix A is assumed to be square. Therefore =, observe the number of mixtures. However, this assumption can sometimes be relaxed: the matrix W can be determined for some measurable characteristic towels. "The ranking of the matrix = A determines how many signals can be actually separated." A 'mixed matrix with 4 sorts means that 4 source signals can be used. To be sure, the ordering of the mixture matrix A should be at least equal to the signal source. The larger the order, the more signals can be separated. The number of antenna elements required has also increased. It is discussed in the background section 16 1287366, and, as the patent case, it is revealed that the antenna element is larger than the number of signal sources, that is, N^M, otherwise, it is not dead. The technique will be used to separate the signals. The industry standard that separates and creates the linear independent summation of the signals is such that the associated sensors, that is, the sensors are isolated from each other at least, the wavelength is based on the operating frequency of the communication device 30. .heart Spatially uncorrelated 'but polar and angularly related ^:: The sensor provides linear independent signal summation, where each sense provides a single entry into the mixing matrix A. 糸 creates a source for the hybrid matrix A The description or elaboration of the linear independent summation of the signal will be discussed first with reference to Figure 3. After a brief introduction, the methods will be discussed in more detail below. Block 1 indicates that the sensor is not associated with each sensor. Providing m = input to the mixing matrix 块 block 102 is a phase (four) contiguous antenna array, wherein the array is used for multiple inputs to aggregate the mixing matrix A. The block 〇4 also represents an antenna array, wherein the antenna element is partially pure (four) And the antenna elements have different polarities that can be combined with the mixing matrix A. Different combinations of sensors and antenna arrays proposed by blocks _, m and 1 〇 4 can be combined in block 106 to further aggregate the mixing matrix a in block 116. The second paragraph of the illustration is intended to enhance the antenna configuration provided in the first paragraph. The enhancement causes the source signal to be linearly added or replaced, and collected to further assemble the mixing matrix A. Block 108 is related to the antenna field elevation angle being changed to receive the source signal plus the sum of the array biases. Any combination of the blocks 均可6 can be used in the array deflection block 1〇8 17 1287366 4 ο in the block 110, the path selection is performed so that the aggregate matrix A is used to aggregate all the source signals are correlated Union (first and second moments) and / or, first count (second and fourth moments) independent. That is, the incident signal is selectively selected to receive the source signal summed to replace the unassociated sums that are not associated and/or counted independently. The block i〇 can be fed by the block 〇6 and the blocker (10), and δ is fed. Blocks 108 and 11 can be fed directly to the lube matrix block 116. The third paragraph of the explanatory diagram proposes signal segmentation to further aggregate the mixing matrix in block 116. For example, 'block 112 uses a spreading code to split the different summed signals. If the summed signal has a k-expansion code, then the particular summed signal can be processed to provide the associated summed signal. The expansion code can be applied in conjunction with the outputs of blocks 106' 108 and 11〇. The block ^^ system divides the different summed signals into in-phase (1) and quadrature (Q) components to further aggregate the mixing matrix A. The I and Q components are therefore used as the 2 multiplier for the missing moment matrix, and the outputs of the combined blocks 1〇6, 1〇8 ′ 11〇 and ιΐ2 are applied. The final selection of the illustrated map is formed in the hybrid matrix A in block 116. As shown in the explanatory diagram, the mixing matrix a may be added differently by the aggregated source signals based on any of the above blocks. The advantage of the antenna array configuration in the first paragraph is that a small antenna array can be formed to assemble the mixing matrix A. The antenna array configuration advantages in the second and third paragraphs are n antennas, where N is less than the source signal number M, and can be used to add more to the M-stacking matrix or source signal. This is a discussion of the antenna configuration discussed in the accompanying drawings, including N associated antenna elements to receive M source signals. At least N different summing antenna arrays will be discussed, where N&M is greater than 〗. In an embodiment, as shown in FIG. 4, the antenna array is a beam switching antenna 14A. The beam switching antenna 40 can produce a complex antenna pattern comprising a directional antenna pattern and an omnidirectional antenna pattern. The beam switching antenna 14A can be configured with an active antenna element 142 and a pair of passive antenna elements 144. The active and passive antenna elements 142, 144 differ depending on the intended application. Reference is made to U.S. Patent Application Serial No. 11/65,752 for further discussion of beam-switched antenna arrays. This patent application is hereby incorporated by reference in its entirety in its entirety in its entirety in its entirety in its entirety in its entirety in its entirety in Each passive antenna element 144 includes a first half 14 buckle and a lower half 144 b. The upper half of the passive antenna element 144 is connected to the ground 146 via a reactive load 148. The reactive load 148 is a variable reactance, / can be changed from a capacitor to an inductor by using a time-consuming, transmission line or swap. The radiation pattern can be altered by changing the reactive load 148. ^ There are two passive antenna elements 144, so four different antenna types can be formed. The two antenna field type can be used to receive the unique sum of the signal illusions. The fourth field type is a linear combination of the other three, so it cannot be regarded as the entrance in the mixing matrix A. Therefore, since the three antenna elements are used, the two unique sums of the χ"" are input to the mixing matrix a. The advantage of the beam switching antenna is that the mixing matrix of the sorting 3 can be supported by using the three elements 142 and 144. 19 1287366 As shown in FIG. 5, in another embodiment, the antenna array includes N associated active antenna elements that enable the antenna array to form a phase array 160. Phase array 160 includes a plurality of active antenna elements 162 and a complex weight control component 164 coupled to the active antenna elements. The weight control component 164 can adjust the amplitude and/or phase of the received signal to form a combined beam. Splitter/combiner 166 and controller 168 are coupled to weight call control 1164. Reference is made to U.S. Patent No. 6,473, No. 36 to discuss the active array 160 in more detail. This application is assigned to the present disclosure, the contents of which are incorporated herein by reference. The number of active components 162 supports a hybrid matrix A having the same order. Compared to the conventional method of using an antenna element that is not associated with more than one wavelength, even if the source number Μ is equal to the number of active elements n, that is, Μ=Ν, since the active element ι62 is associated with space and polarity, the active array 1 〇〇 is very close. • In another embodiment, the mixing matrix ordering may be Κ, where Κ< Ν ' enables the blind signal separation processor 49 to separate the κ of the μ source signals from the mixing matrix. As discussed in more detail below, Ν may also be greater than Μ. In the beam switching antenna 140 and the phase array 16A, the distance between the individual antennas π, 142, 144 and 162 is set to contribute to a favorable front-to-back ratio. This is due to the fact that these antenna arrays use the lower system to reject unintended signals (that is, to approach the near side) and to strengthen the expected signal (that is, the front approach). However, in order to build a hybrid matrix, the target creates a different sum of signals. The useful signal in this application can actually be separated forever from the stalk of 20 1287366. Because this #intention distance does not need to be a specific separation. , between the antenna array: the line 70 can be moved into the step or closer together to produce the field type, and still quite suitable for the mixed m two signal source separation application usually reported a superior %, good use before and after the mouth system Tracking signal direction: Only when the expected signal and/or the back end are #, , the temple is different from the field type of the direction difference, and the use of each green, ancient, and page gain is not traced by the signal. The beam can be defined to have a signal from the maximum gain point that is close to at least - the direction of the two sigma points. The antenna pattern can be defined as a substantial benefit number, ,, and bar. Similarly, there is no signal close to either side...down to the location. In many applications, this offset component is lower overall antenna array size. Use it for the shore, "(4) can greatly reduce the number of L-eight in his application. 'The distance between the actual and expected parts can be used to reduce the problem of tracking, but increase the production; the signal is not related. Attached-implemented, as shown in Figure 6, antenna = τ: the source signal to the μ different sum of the fixed antenna element. At least two of the 1823 antenna elements are associated, and Having different purity, (d) at least two of the N different sums of pure additions, at 1. Other antenna elements ma, 184b in array 180 may be associated by fish antennas τ 182a, 182b. The other pair of antenna elements 1 8 4 & 18 4 b are traced, and these elements still have the same polarity of 1287366. Again, these elements are also not associated with each other. The different polarities of the antenna elements 182a, 182b can be mutually In another configuration, antenna elements 182a, 182b include a third element 182c that enables three polarities to be received to receive three different sums of M source signals. The following discussion supports the use of polarization to assemble a hybrid matrix A. Three different polarized antenna elements 182a, 1 82b, 182c can receive three lines of Lu and independent 彳 & number sum. As shown in Figure 7, the X, y and z axes are defined... The meaning and relationship will be used. For example, the following relationship exists: x = ScosO^sin (4) y = S ^ιη(θ)Βΐη(ψ) 2 = 5 cos(^) Simplify the hypothesis that the signal has linear polarization, the signal is linearly independent, and each has three linear antenna elements on the orthogonal axis. For example, the antenna element 182a is located on the X-axis, the antenna element 182b is located on the y-axis, and the antenna element 182c is located on the z-axis. φ By placing the trilinear antenna elements l82a, lMb, and l82c on the orthogonal axes, the mathematics system is Simplified. In the actual configuration, the antenna elements 182a, 182b' 182c need not be completely orthogonal, and they need not be satisfied with the common point. The removal of this assumption invalidates the general conclusion, and the change of the sorting is insufficient. The definition is applied, where the numerical subscript relates to the signals i, 2, 3:

Si, S2, S3: signal of the incident antenna element; β β 2 ' 0 3 · · X, Y plane E-field angle of the signal; 22 1287366 HA: Z-axis E-domain angle of the signal;

Xx, Xy, Xz: The summed product of the signal of the incident antenna element. Therefore, the vector component is: X y component "X": 1 0 component "y" : 0 1 component: 0 0 Si coefficient: (10) (9)) sin (color)

Z 0 0 1

S2 coefficient: cos〇92)sin(m) sini^sin(for) sii^Jsin(m) c〇s( φλ) cos(m) S3 coefficient: cos(03)sin(ii3) sin(^)sin( ^3) cos(^) takes the product of the point of each antenna element and signal, (Ζ·7 = ¥2+*^2+ζΑ) determines the relative Ε domain component that is added to the component. These values are used to create the blending matrix: X" cos(0) sin(private)cos(02) sin(#2) cos(6>3) sin(Lu 3) = sin^) sin(color)sin( ^2) sin(^2) sin(^3) sin(^3) cos(A) cos(#2) cos(^3) β, where det 23 1287366 cos(A) sin(A ) sin(02) Sin(6) c〇s(h ) + cos(6>2) sin(Lu 2) sin(6>3) sin(b) cos%) + cos(03) sin(#3) sin(A) sin % ) cos(#2) -cos(4) sin(02) sin(h) cos(03) sin(^3) - cos(h) sin(03) sin(b) cos(0)) sin(h ) - cos(h) sin(A) sin%) cos(6>2) sin(6) =cos(4) sin(02) sin%) sin(^) cos(A) + cos(02) sin( h) cos(^) sin(^2) sin(h) + sin(A) cos(h) sin(A) cos(h) sin(03) -sin(&) cos(^) cos% )sin (&) sin(#3) - cos(A) sin(03) sin(A) cos〇2) sin(h) - sin(0))cos(02) sin(A ) sin(^2) cos (#3) =cos(4) sin(02) sin%) sin(02) cos(h) - sin(h) cos(02) sin%) sin(#2) cos(03) + cos(h) Sin(%) cos%) sin(#2) sin(h) - sin%. Cos(4) cos%) sin(02) sin(A) + sin(4) cos(i93 ) sin(A) cos(>2) sin(b) - cos(A) sin(h) sin% ) cos(#2) sin(#3) =sin%) sin(Lu 2) cos(b )[cos(4) sin(& ) - sii^A) cos(&)] + cos(^ ) Sin(^2 )sin(^3 )[cos(^2 )sin(^3) - sin(^2) cos(/93)] + sin(^) cos(^2) sin(^3 )[sin (01) cos(^3) - cos(^) sin(<93)] =sin(^) sin(^2) cos(^) sin(^2 -^)

+ cos(^|) sin(^2) sin((^3) sin(03 - ^) + sin(^) cos(^2) sin(^3) sin(0j - ) Insufficient sorting will now be discussed When the decision is equal to 0, the mixed matrix is undersorted. This occurs in the following example: 1) θ θ 2 = Θ 3 'X' and 'y' components receive the same contribution from all three signals. 2)

Φι Φι Φι 0 0 0 0 0 90° 0 90° 0 90° 0 0 90° 90° 90c Add 180 degrees to any combination of the table entry of another sort of insufficiency. These occur when the signal is not summed up by the antenna elements in sufficient combination. 3) All individual sums are not equal to 0 every 1 or 2, but: 24 1287366 sin(0))sin(Lu 2) cos(念)sin(& - <9)) + cos%) sin(6) Sin(03) sin(<93 - %) + sin%) cos(02) sin(03) sin% - &) = 〇 角 angle, almost equal to the signal pole ' or the occurrence of two components may happen by chance The signal-injection is the small separation between the signals, and some other from the calibrated signal energy level on the opposite side of the array is very inferior to the above. The first paragraph of the figure shows the antenna configuration. The above-mentioned antenna configuration including unassociated sensors can be provided in various combinations of configurations] V [source source missing + i ^ died from 1 / original 5th summed signal to the mixing matrix 〇 Now referring to Figure 8, the separation of the source signals provided by the M sources can be discussed. The antenna array includes at least N different summed antenna elements that can receive at least one source signal, N and Μ greater than one. The Ν antenna τ 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 The two associated antenna elements 206 are not phased with the antenna elements 2〇4. The antenna array can include additional antenna elements in various combinations that are associated, uncorrelated, and polarized. Receiver 210 is coupled to antenna array 2〇2 for receiving at least a different sum of the μ source signals. The blind signal separation processor 2丨2 is coupled to the receiver to form a mixing matrix 214 comprising at least N different sums of the source signals. The hybrid matrix has a sequence of equal to at least N, and the blind signal separation processor 212 separates the expected source signal 2i6 from the disc matrices A. b The second paragraph of the diagram presents an enhancement to the antenna configuration provided in the first paragraph. This enhancement causes the source signal to be linearly appended = summed up to further aggregate the mixing matrix A. An enhancement involves an array bias that can receive additional signals and add to the mixing matrix A without having to add additional antenna elements. Array biasing involves controlling the antenna pattern in the azimuth and/or elevation directions. σ A communication device 240 that can separate the source signal provided by the chirp source using an antenna array will now be discussed with reference to FIG. The antenna array includes a chirp antenna element 244 that produces a different sum of the initial antenna pattern for receiving the chirp signal. The antenna array 242 also includes a selectable elevation controller 246 for selectively changing the elevation angle of at least one of the first day of the field to generate at least one additional antenna pattern to cause at least one additional difference in the source signal. The sum is received. Receiver 248 is coupled to antenna array 242 and uses the initial antenna pattern to receive a different sum of the source signals, and also uses at least one additional antenna pattern to receive at least one additional sum of the source signals. The blind signal separation processor 250 is coupled to the receiver 248 to form a mixing matrix 252 comprising at least a different summation of the source signals and a different sum of the μ source signals. The hybrid matrix has at least one additional different sum total equal to at least Ν plus the use of an additional antenna pattern to receive the L L number. The processor 250 can separate the pre- 26 Ϊ 286366 period 254 from the mixing matrix. In general, a signal summation suitable for increasing the ordering of the mixing matrix is provided. Any antenna array device can use a biasing mechanism. This bias produces two different and hybrid array usable signals for the pin and antenna array devices. So there are twice the multiplication effect using this technique. If the array deflection is divided into different regions of the κ different regions connected to the antenna, two independent deflection regions and the entrance to the mixing moment = can be provided. For example, if the antenna array can provide its own total and have different deflection regions, then the total number of signals in the mixing matrix can be referred to as FIG. 10 for the trace system, wherein the beam switching antenna 100 shown in FIG. 4, It has been modified so that the antenna pattern can be tilted up or down. In particular, the passive antenna elements 1〇4, each of which is connected to the ground 1〇6 via a reactive load 108. The passive antenna element 104, each of which is also connected to the ground 106' via a reactive load 1〇8. The passive antenna element 1〇4 has a reactance system that increases or shortens the effect of the passive antenna element. The inductive load increases the capacitive load and shortens the electron length of the passive antenna element 1〇4. The antenna beam is tilted upward and downward according to the reactance load 108, the upper half l4a, and the lower & anti-load 118' lower half i4b'. As shown in the figure, by adjusting the ratio, the antenna pattern can point to the upper 97 or the downward 99. When the antenna field elevation angle is adjusted to receive the mixed signal, at least one additional order can be added to the mixing matrix A. If the array is biased, it is not necessary to increase the number of N elements of the antenna element, and more signals can be received and mixed with the matrix 1 A. The specific implementation has two reactances 118, which control 2 different deflection regions. It is a three-independent field type, so the total number of signals that can be used to create a hybrid matrix is 12 (2*2*3). Referring to the above-mentioned U.S. Patent No. 11/65,752, it discloses in more detail how to adjust the antenna beam. Elevation angle. Array deflection techniques can be applied to any of the antenna array embodiments described above or any other antenna array that is sensitive to ground interaction. As shown in Figure 12, another embodiment of the elevation controller is coupled to the antenna. The ground 272 of the component 274 can control the RF suppression coil 270. It will be readily understood by those skilled in the art that the antenna field associated with the antenna element a" is controlled by the RF suppression coil 27 Move. A communication device 300 that can separate the M signal source providing support signals based on the path selection will be discussed with reference to FIG. This is an enhancement to the antenna configuration provided in the first paragraph of the illustration and to the above array bias. The communication device 300 includes an antenna array 〇2 including N elements 304 that can be used to receive at least N different sums of at least N antenna beams, N and Μ greater than two. Controller 306 is coupled to the antenna array for selective formation to the 'antenna beam receiving component 308 is coupled to the antenna array 3 to receive at least the summation of the source signals. The blind signal separation processor 310 is coupled to the receiver component 3〇8 to form a mixing matrix 3 12 comprising at least a different sum of the up-source signals. 28 1287366 % The blind signal separation processor 310 also determines whether the different summation signals of the source signals are associated or statistically independent, and if not, cooperate with the controller 3% to form a new different sum of the receivable source signals. Different beams are substituted for different sums of the mixed source matrix 312 that are not associated or statistically independent. The expected source signal 314 is then separated from the mixing matrix 312. A rake receiver is a wireless receiver designed to calculate multiple fading effects. This is accomplished by tuning each of the multiple components using a number of independent receivers that are each slightly delayed. It is accessible by most wireless access network types. A special benefit of the type of modulation expansion code has been found. Its ability to select a particular incident signal path is such that it can be used as a means of changing the path that is fed to the blind signal separation process. Those skilled in the art will readily appreciate that selective N-antenna beams as described above can be applied to all wireless access networks. For a code division multiple access (CDMA) system, the receiver component 3〇8 includes a receiver 316. Each of the (4) receivers 316 includes a k-loss finger that can add a total of k different multi-path components for the M-source signals received by the respective antenna elements that are connected. In this configuration, the blind signal separation process crying 310 is connected to the N-dissipation receiver 316 to form a hybrid matrix. The mixing matrix 312 includes at least N different fueling signals to different multipath components' and the hybrid matrix has an equalization order. In particular, when a split-wave multiple access waveform is transmitted, it is usually 29 1287366 ^ multipath from source to destination. The rake receiver 3 16 is specifically used to capture several of these individual instances and combine them for more powerful L唬 decoding. When the original signal is passed along each path, its characteristics are corrected by the unique characteristics of the path. In some cases, the phase of the received signal, and/or the statistical corrections will be large enough to be considered as a separate signal stream. The modified rake receiver 316 can be used to retrieve each corrected stream and treat it as the only entry to the mixing matrix 312. Although this increased ordering device is not always available, but when it is the most It is easy to obtain in high multi-path environments when needed. As discussed in Figure 13, although the rake receiver 316 can develop different paths, the more general application to any modulation technique is beamforming. This is different from the rake receiver 3 1 波束 because beamforming is used for the expected signal % and expected signal rejection. However, the difference is that the rejected signal can actually be expected to be used to receive another version. However, the receiver component 3〇8 must detect the same spring version of these same transfer path versions to build the hybrid matrix 3 12 to full ordering. The third paragraph of the explanatory diagram proposes to further aggregate the mixed matrix number segmentation. In one method, the summed signal is extracted using a spreading code. In another method, the sum signal is split using J and Q modules. The division of signals using the unwrapping code will now be discussed with reference to Figure 14. The depicted communication device 400 includes an antenna array 4〇2 that includes n antenna elements 30 1287366 4〇4 that can form at least N different sums for receiving M source signals. The coded frequency reducer 406 is coupled to the N antenna elements 4〇4 for solution, the M source signals being at least N differently summed. Each of the N different sums can be provided, and the M source signal k of this link is different from the total k code. Receive benefit component 408 is coupled to code downconverter 406 to receive the source signal at least kN differently summed. The blind signal separation process is connected to the receiver component to form a mixing matrix 412 comprising at least different sums of the source signals. The mixing matrix 412 has a ranking of up to kN. The blind signal separation processor 41 is separated from the mixing matrix 412 by the number 4M. Predicting the goods: depending on the modulation of the signal, the above signal segmentation can be used; the ordering of the matrix A does not increase the number of antenna elements N. Two IS_95, CDMA2 〇()() and wcdma are examples of the spread spectrum coded spectrum communication system. Commonly used threads are code-only coded to process data on a larger frequency band. The same unfolding stone system is processed by the sum of the received signals (expected signal, period signal and unknown noise source). This causes the expected reconstruction to return to its original bandwidth, while the interference is spread over the broadband. The built-in CDMA implementation actually has a simultaneous use of multiple signal streams. Each signal flow system is coded theoretically with all other positive parents. If this condition is satisfied at the decoder, the soil :: signal will be despread. If the total κ signal is "de-spread", then the total received signal plus xk will be mostly composed of amplitude = term and unchanged or lower value term. 曰 plus coded multiple access signals are usually between There are some associations, so the 31 1287366 disturbing tiger system is reconstructed to some extent along with the expected signal. This is usually caused by the delay experienced by the ft number and the multipath of the signal. Some non-pre-dragons, especially the points The code multiple accessor will increase the value. The increase will not be as obvious as the expected signal, but it will increase the total noise value and reduce the signal noise ratio. The spread frequency equation and the money itself are designed to satisfy the blind signal. Processing criteria. One of the right expansion codes on the Gobbe, which is individually applied to the communication device 400, is received by the communication device 400, and the received ^ number is received, and the independence is obtained. Analyze the individual requirements of the model requirements. Therefore, there are many columns of the matrix used for the mixing matrix, and it is assumed that each of them produces a linear independent saliency value. This will increase the mixing matrix to greater than the encoding. One of the numbers, N days The element and M code can provide NM matrix columns. In order to describe the '3 coding system, it is assumed that the 3 signal system retains its orthogonality. The code frequency reducer 4〇6 towel, the $丘睡A system has various causes. The antenna flows through the stream by the 3 ^ ^ ^ · ^ ^ - Γ ^ ^ , and the flat code solves the spreading of the upper three columns and the lower three 卩. It is not in the diagonal Q (four) because of the positive father of the encoding. 5 and 6 are used for the same finger ·', example. U heart 軚 unknown signal general ^ α\\ 0 0 ^14 αΐ5 α\6 "7 广ή χ2 0 α22 0 α^Α 24 α25 α26 S2 χ3 = 0 0 α33 α34 α35 α36 χ4 α41 0 0 ^44 α45 α46 Χ5 0 α52 0 α54 α55 α56 3J 0 0 α63 α64 α65 α66 J The signal corresponding to the entrance of the column 4'5 and 6 can be known as other

S 32 !287366 : two: t ' or other cell signals known to be encoded. At the same time, one, the second is: the other signal is obeyed by the central limit theorem to make it a sub-code multi-access signal group of the 4 channel. Interfering rabbits can also be ▲ Tigers will be added to the Gaussian signal. number. After the non-Nans signal source or the network-unknown multi-Gaussian signal divider 406 makes it known for encoding, the Blind Letter 3 V2 receives the mixing matrix 412 of the sort 6. Since the sorting of the known horses of '6' is applied to the blind signal separation processor by the 6 signal system with the 2 antenna elements being up to 3 factors, and the middle matching matrix 412 is formed. Blind signal separation processing: The separation matrix W can be determined only by the k channel: X=AS corrected received signal. In the example, the 6 signal system is separable. The blind signal separation processor 41() can select (4) to be decoded. The signal can be dropped' and the expected signal is all versions. The (4) is applied to the demodulation group to demodulate the hacker system. It can be used to combine the same signal multi-channel version of the well-known technique 0. In the general case, 'for simplicity, the above shows 0 is not right: the value can be consistently Is a non-zero value. This will be a more useful example when the encoded signal is not perfect. This can indicate additional noise for each number. However, as previously shown, the matrix;:: separates these signals, so its value after the blind signal separation process: 33 1287366 is significantly reduced. This results in reduced noise, increased signal-to-noise ratio, and increased channel capacity as indicated by Shannon's law. Referring now to Figure 15, another method of increasing the order of the mixing matrix a without increasing the number of antenna elements N separates the received mixed signal into I and Q components. The j & Q component of the coherent RF signal is a component of the same amplitude but separated by 9 degrees. The overnight device 500 comprises an antenna array 5〇2 comprising N antenna elements 5〇4 that can receive at least N different sums of the source signals. Each worker and Q module 506 is located downstream of each antenna element 5〇4 and is used to separate the N different adders of the received source into j and q component groups. The receiver component 508 is located downstream of each of the J & Q modules 5〇6 to receive at least N different sums of at least N and 〇 component groups of the source signal. The blind signal separation processor 51 is located downstream of the receiver component 5〇8 to form a hybrid matrix 512 comprising at least 2N different sums of the source signals. Each I and q component group provides a 2 input to the mixing matrix 512. The mixing matrix 512 has an order equal to 2N, and the blind signal separation processor 510 can separate the expected source signal 514 from the mixing moment 2. One of the Q modules 5〇6 downstream of the antenna element 502 is depicted in Fig. 16. The mixed hues received at the antenna elements 5〇2 are divided by a pair of mixers 520. The jkq component is usually applied to the two simultaneous samples by the phase difference of 90 degrees from the reference signal; the converted intermediate frequency (IF) signal is generated for another frequency range. At the same time, the ^ and 〇 signals can save the phase signal contained in the IF signal, thereby contributing to 34 1287366. The signal with eight f frequencies is different from the signal with negative frequency. Chen Dawei: Separation of the received mixed signal into I and Q components, and the mixing centroid increases by a factor of 2. As long as the 1 and Q components are treated with different data = stone horses, the mixed signal received by any antenna element can be divided into two different mixed signals. In different coding examples, the modulation property must be analyzed to determine whether the linearity requirement is met. For example, for Global System for Mobile Communications (GSM), when Gaussian Minimum Shift Keying (GMSK) coding is properly used for filtering, it can be assumed to be linear, and when processed in a receiver, it is like bi-phase shift keying. (BPSK) encoding. Once the dual phase shift keying meets the blind signal separation processing requirements, the illustrated and Q processed can be used. And the Q component can be used by any of the above antenna array embodiments to integrate, (7) / 3⁄4 the matrix A. When I and Q are used, the mixing matrix A can be used by the current and m π if the number of antenna elements is twice. Another example can use two antenna elements (2 factors) that are not associated with _ unpolarized polarization (2*2 factor), and combine with I and Q components (2*2*2 factors) to add 8 independent mixed signals. Can always be produced. This mechanism can also be used by antenna array bias technology to create more signal summation. These sums are also separated into I and Q components in order. Another feature of the present invention relates to multiple input and multiple output (ΜΙΜΟ) antenna techniques that can exploit multiple uses of the same RF channel. Interference Cancellation Receiver processing techniques can minimize the number of antennas required by developing field diversity, rather than using antenna diversity to achieve increased signal transmission and 35 1287366 data rates. The antenna array has a changeable weight in its receiver path. When these weights are changed, the receiving antenna pattern is corrected. By using techniques similar to these blind signal separation proofs, the expected signal can be extracted from receiver data containing signals from several interferers. As shown in Fig. 17, no matter how the field type is formed, it is possible to replace the antenna diversity system in the receiving structure of the multi-input and multi-output implementation with the field type φ set. The K field numbers are theoretically equal to the number of N antenna elements. However, K field patterns can be generated with L antenna elements which are lower than the N antenna elements required by the prior art. Similar to the existing antenna array multi-input and multi-output implementations, Μ and K are equal only in the case where all transmitted M spatial channels can be identified by K receiver patterns. Since this is usually only used for fixed transmitter and receiver examples, it is necessary to exceed the receiver field type or the transmitter field type to achieve a minimum space gain. Multi-user detection processing techniques are used to separate data channels from the receiver system. The method of constructing the hybrid matrix described above can be considered as part of this implementation. Another feature of the invention relates to inter-symbol interference (Isi). The limitation of using the Fourier transform to reduce intersymbol interference is proposed by Figure 18 = configuration. The following block is added to the transmission side to improve the Fourier transform method for reducing inter-symbol interference: coffee coding, repetition/interleaving, and block redundancy interleaving have been added to the transmission side. On the receiving side, the following blocks are added: blind signal separation interference removal, block deinterlacing, go to 36 1287366 repeat/de-interlace and Vi t e r b i decoding. "Viterbi coding" has strong redundancy that overcomes data decoding processing errors. Coded alternatives such as turbo coding can also be applied. "Repeat or interspersed" can cause the source data rate and the data block of the transmitted data rate gate to match. "Block interleaving, the random arrival of the source data can be randomized to maximize the appropriate decoding probability." It can improve the return channel of the transmission channel. This will be able to randomly distribute the data stream from far more block errors. The error-recovering Viterbi decoder previously introduces block errors due to severe fading by assigning the block error. The "blind signal separation interference removal" can reduce the signal to the expected signal before converting to the time domain. The best way to handle non-conforming (PAR level) is to add a non-light matcher (used to equalize the signal level across the frequency) to the fast zebra leaf. Convert the output and add a reverse conversion to the reverse fast Fourier transform input. In addition, this signal is usually modulated and aggregated to the transmit frequency in an ideal case, so add the modulator, upconverter and Down, the converter 'demodulation ϋ can complete the image. The boundary between the transmitted waveforms = has discontinuity. This can be mitigated in several ways. One can:, add, bring between the modes Wherein - the curve is interpolated to the frequency component produced by the minimization of the mode j X. All of the above methods of constructing the hybrid p can be considered as part of this implementation. J. Another feature of the invention is related to support Field type communication for layer space communication Referring now to Figure 19, in the preferred embodiment, the transmitter changes the power level of each hierarchical space by means of 37 1287366:: stone. The stream thus arrives at the s\power level. Quasi-received 11 where it provides the appropriate difference of the received signal: to assemble the matrix suitable for blind signal separation processing. Since all power is positively reached at the transmitter, the L antenna at the receiver: the number of cattle For the 1' and the receiving person does not need the field type to generate the hardware or software component 0., this method also satisfies the prior art, in which the small angle between the arrival signals and the degree difference is no longer necessary to create a field contour that is sufficient to identify the signal. Problem. In another embodiment, the 'has a distinct dryness from the intended transmitter. If there is a single such interferer, the difference between it and the expected expected transmitter wavefront will be sufficient to separate the blind signal separation process. All letters. There is one right More than one significant interferer may be insufficient for matrix ordering. System performance can be improved by creating additional field variations at the receiver. Although this deviates from the preferred embodiment, it still requires significantly fewer field types than before, so In another embodiment of the spring, the complex data streams are summed together via a power amplifier via a single antenna element. On a time-sharing basis, the relative power between the signals is summed. The level is changed in a manner suitable for blind signal separation and decoding at the receiver. The advantage of this method is that each signal stream in the combined signal experiences the same transmission path effect, which means that the relative signal relationship can be maintained at the transmitter and reception. Between the devices provides a very powerful decoding at the receiver. This concept can measure 'several of the individual signals can be transmitted via different antennas 70. The strong signal separation can be combined with the multi-channel diversity gain 38 Ϊ 287366 and / or the valley The amount of gain comes. The power ratio is theoretically approximately equal to the average signal level for the peak-to-average signal: the method of the metering matrix is (4). All of the above-mentioned constructions are part of this practice. Another feature of the invention relates to the field type. Now refer to the 20m difficulty, and the wave of the dan modulating the _ type of the complex device is the difference of the transmitted signal: rate == the unintended access point will receive the information required version of the separated signal. This provides blind signal separation. The penalty #^r can be as simple as changing the power transmitted. This can be achieved by independent field = 全, so omnidirectional, divided by fan (four) and even beam = field type can be used. Changing the transmit beam aperture line can also be used. ^ The most efficient method is to use the transmission of the calibrated slot to cry. This point can be set by using the internal clock in the device, or up to the same time; If the signal does not calibrate when it arrives at the burglary, the ability of the blind signal to separate and separate the signal is reduced. Calibration can be adjusted by determining the distance to the device or the measurement time delay. The access device can then use the point-in-time or delay technique. False " The gain change of the received k is considered to be the target of the blind ## separate installation access point and other examples of interference use, then the registration gain may change. If there is no entire network coordinates, the receiver is expected to be calibrated. If the entire network coordinates are present, the measurement may show that the best method is to make the signal easier to remove the interfering sub-, while still providing sufficient separation to the intended receiver. If there are other (4) technologies that do not use RF power level modulation technology: ΓΓ" Tiger rejection technology can be used. Alternatively, the ':= field type or other ^ to increase the f signal score == (four) set is used, the matrix information is derived two: also greatly reduce the branch implemented at the access point receiver = above The method of constructing the hybrid matrix can be used as the implementation of the 1 knife 0 f. This is the deduction of the 彳 系 有关 调整 调整 调整 调整 调整 调整 调整 调整 调整 调整 调整 。 。 。 。 。 。 。 。 。 。 The signal that needs to be separated to decode the useful stream is low. In general, the decoding matrix ordering determines the number of tigers to be separated and the remaining signals are considered as noise. Therefore, this 2 must be the minimum value that contains the signal to be decoded. It is possible to lower the noise component with a higher minimum value, so that the ratio of signal to noise is allowed to be subject to the decoding error rate. & Figure 21 depicts a receiver implementation that operates only. Figure 22 is a superset of the Ang 21 diagram and also contains information from the transmitter to the receiver and information from the receiver to the transmitter. If the fill matrix is selected to exceed the order of operations f, the antenna array control can reduce the number of choices used. Some of the choices from the available groups may be better than others, and the best choices may result in lower matrix ordering. This group can be determined by examining the L number from various options compared to other choices, by fallacy techniques (such as comparing the results of being used and not being used to select k), or by tracking the situation and results history. The combination of the method t, i λ a ^ ^ A used in 1287366 can also be determined based on the assumption of known conditions and historical evidence. § The device is known to be located within a significant signal range from several sources, ^ occurring in the covered overlap region, then the highest power signal can be expected to be "in different directions from the moon. This selection should be chosen to provide significant signal differences in these directions. Regarding coding, the error correction coding determines the error rate that can be tolerated in the original Lu decoding stream. Since the original error rate is also a matrix-filled, sub-aggregate function, there is a permutation relationship between these settings. The feedback and control loop between the stone machine and the decoder can be selected to optimize the mutual setting. The β right receiver is found not to be in a power limited condition (e.g., line voltage power) and the decoder can increase its matrix ordering. This can be used for a variety of purposes. Higher ordering reduces noise, which reduces the signal-to-noise ratio, which in turn reduces the error rate. Reduced noise can be used to increase transmission rate, reduce error correction coding, or improve overall chain reliability. The padding burden of the transfer matrix to the receiver can also reduce the burden on the transmitter, which can be explored if there is a control loop between the two. Conversely, 'battery-based devices can try to transfer the order to create a more powerful supply.曰到 By changing the time point setting, the most powerful operation requires the decoding moment to be recalculated for each symbol. However, the coherence time usually exceeds the 浐 number, making the measurement rate only slightly faster than the coherence time. = The occurrence of low decoding matrix decisions will save power and processor overhead. At 41 1287366, depending on the occurrence of a $saw + i less - the matrix change is used to determine how often the matrix hangs. Decoding often has individual coherence times. Each sub-frequency two = two: channel pass Very large decoding heart needs =: the fastest required rate is recalculated - always less than the usual use, the measurement of the sub-matrix matrix is used - the measurement of the larger one is added. The straight moment sends the 'create source field' and the receiver can adjust to sort the sufficient matrix. The receiver can be exemplified; the % code is measured by the material, or the transfer setting with the source. Transfer == Resource restrictions can also be considered, so any one: ΐ 5 is more burdened to remove the other. To solve the problem, the decoding matrix will change too much from one calculation to the next. The previous value can therefore be used as a solution: , = the seed of the 疋 其 其 其 其 其 其 其 ' ' ' ' ' 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器 处理器The same is true when it is decided to be in an unknown state. This is to solve the large order 'quantity full fill matrix. Usually, all of the above combinations are visible components, corrected coding levels, suitable equipment and other factors that affect reasonable operation. All of the above methods of constructing a hybrid matrix can be considered as part of this implementation. - The invention is characterized in that the wave field type supports the effective area. For field-type transmission, the basic concept is to cover the field type by using the architecture address at s 42 1287366. It is used by the actual 扃 ga , , π ^ ^, and the difference between the demand and the cost factor. The actual implementation can be any large quantity. The sector itself can be subdivided into azimuth or elevation planes. The key benefit of using the Aug. A 1P angle or the knives of the knives is to reduce the reliance on the _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ The ^ domain gives another transmitter that is thus used by the prior art to produce the blind money separation process. The technique uses appropriate techniques to make the appropriate blind signal: no need to generate any wave field type. In other implementations, it means the wave field. The number of fonts is significantly reduced. ^F secret is used by a two-information system for a transmission point. This embodiment proposes that I do not know if the source is also operating. Referring to (4), the transmission type contour is attached to the pure instrument (4) The sequence towel is made into a wave. = The change in the knowledge type is timed to match the transmission symbol. The field contour is for each time slot and the text is fixed and fixed. Yang μ # There is no foreseeable memorial message to be dealt with. The domain Asia does not clarify the change, and the recipient will experience a change in age due to the change of age. The blind wave has eight wavefront power generated by the contour of the wheel. The difference between each signal stream at different relative gain values is collected. The important signals received by m-in-laws are from the 5th-day thieves sent by the wave signal, and the receiver only uses the sample for the change of the field type 43 1287366. And use the final data if there is - use waves ^ The transmitter does not use this wave signal to send his orthodox signal separation technique to explain it &&w using the user detection method can be:] = such as: beamforming and more usually more powerful. The knife-off method is sufficient to add a field-type blind device to perform field-type deformation and generate the number of separated signals: No. 1 separation matrix ordering, the ordering is =, for blind signal separation decoder implementation and the receiver must generate When two other signals are received, the ^^~ contour is separated to separate the interferences that are unfavorable to each other. If the transmission n does not generate its own corridor, the implementation burden on the receiver is always reduced. The device is following the single ramie catch, ... σ, 士 ^ ^ ^ ^ early road rainbow to convey the early flow, then the field contour group 2:: r connected ... the signal early power changes the entire field without changing Contour shape. Port ^ = stream is added to the receiver, even if its amplitude is fixed, blind can be separated. This is due to the power high-frequency vibration source. Received: change: If more than one other stream is received , the generation of the type:: with other separation devices or Its self-wave field, otherwise it seems to act as a single-packet interferer for the separation of blind signals. The field-type transmitter in the receive mode will now be discussed, because the blind signal separation processing of multiple 1287366% contours is excellent for signal separation. The method, so the same technique used to generate the transmission field type can also be used to generate multiple receiver values. When the transmission is already supported, the only cost factor for blind signal separation reception is the blind signal separation processing expenditure. The device receiver feedback to the transmitter will now be discussed. Although not strictly required, the feedback from the user equipment receiver can be used to improve the entire chain operation. For example, the receiver can determine the Xin% profile. The extent to which the changes provide useful information. This information is fed back to the transmitter. The transmitter can then adjust its operation to improve the bond, use less power, or produce less interference to other communication links. A certain fading can be: what sequence and symbol are used for each field type, and how much change is made during the transmission process (that is, from Μ to N contour change). Each contour change adjustment will have to be transmitted to the receiver for best results

At month b. The second embodiment relates to a multi-transfer point known to use the above method. • The receiver operation of the multi-transmitter address is basically the same as the unit address. The field patterns produced by the different transmitters can be counted at the receiver for blind signal separation. ^ However, more powerful operations can be obtained by receiving information from the network about the parameters transmitted by the co-agreement. For example, a matrix ordering of the required number of fields can be adjusted in sequence. When available, the field generator of the receiver can be adjusted for each. Calling Wireless Resource Management You can use the information that is fed back to the user device to establish a full-domain field type. All of the above methods of constructing a hybrid moment 45 1287366 can be considered as part of this implementation. Another feature of this test is the blind signal separation and field fluctuations to assist in the separation of coded multiple access signals. For blind signal separation calculations, the effective knife is separated from the k-' Xi receive signal and the signals received at the antenna must be aggregated with relatively different weighting factors associated with each signal. This can be achieved in the transmit state, the receiver or both. Whether the weighting factor is changed to the transmitting end or the receiving end, it can be changed per wafer or adjacent wafer group. The basic requirement is that the aggregated signal is adjusted per symbol at least as much as the separated signal. Fig. 24 shows an example in which the symbol is changed 12 times (12 chips). The changed parameters are fixed for 4 wafers. Three variables per symbol Metaphor Three different signals can be separated from the aggregated received signal. If the transmitter is transmitting a single along a single path, the field profile group does not need to be rotated or different. This is because the signal detected at the receiver is changed relative to all other received signals. Therefore, transmitting crying can change the entire field type with simple power without changing the contour profile: Ports - Other streams are added to the receiver, even if their amplitude is fixed, the blind = number separation can separate them. This is due to the power high frequency vibration source providing the required changes in its operation. If more than one other stream is received, it may appear to be a single-group interferer that separates the blind signals, unless the receiver itself uses other separation devices or has its own self-wave field generation capability. Although it is not strictly necessary, the feedback from the use of the device can be used to improve the entire link. For example, the receiver 46 1287366 determines that each change in the field profile is sent back to the transmitter. The transmitter is then 2 degrees per week. This good link uses less power, or °°正一# to change the interference. Although there are many changes to the power wheel = the letter, which produces less sequence for each field type; the symbol 2 ^ ^ some adjustments can be how to modulate and high frequency vibration power to each::: two: The change adjustment will have to be transmitted to the receiving::: The actual power amplifier is best used ^ ^ with a large peak-to-average power ratio, linear work: 'drying. Being reduced - caused by power amplification crying = #范围范围::: When using the transmission parameters, this consideration can be made by two parties: 1时:::t when electricity is used when more than one radiator is supplied by the same amplifier = The 6-off change can be H-stepped with all signal power and sustaining modes. That is to say, some transmission increases are reduced and offset. If the power is transferred close to the crystal _ chain #安, s A1 - , and thousand I values, then the super- and force-rate hoist can be absorbed by the solution surface storage element and cause a small continuous wave. Excess power can be transferred to a distributed load. / Two or three subfield types can be made by a number of transmit and receive antenna devices, including the delay and power level of the phased array antenna. : Parasitic antenna elements for exchangeable loads; power planes that generate field bias Load change; mechanical movement of the component or reflector; and any upper, and σ. All of the above methods of constructing the hybrid matrix can be considered as part of this 47 1287366 implementation. The single parasitic antenna can be coupled to the high speed digitizer and the downhill state to provide a multi-space independent channel to the baseband processing structure. (4) By using a single low noise amplifier (lna), Hybrid H 1 local vibrator (L 0), low pass filter (L p F) and analog digital conversion (ADC) are provided. The multi-space independent channel obtained by this technology can come in more:: pick: virgin The principle can be as follows: a blind signal separation or multiple input. The principle of the system is described as follows: Connection = change component into inductor and capacitor core: shot: = wave limit is presented to the low noise amplifier band and low Noise ^ 隹 放大器 放大器 放大器 并非 并非 并非 并非 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器 放大器Change the 'optional local oscillator exchange and solve the multi-guard exchange=hunting is driven by the same digital sequence generator, so that the heart bottle is changed again: the antenna diversity mode is manufactured. This produces a single from the mix: to the digital converter Channel RF input: 頬 ratio digital converter (not shown) and driveable = same; bit sequence generator, optional local stimulator and consider carrier frequency Fc and frequency conversion wide B - 丄 signal , 48 1287366 t multi: 乍 for the pulse shape of the downsampling operation, the array of ten pairs of Ν components, analogy = Γ, Β. Because each sample only one -: Demodulation in the benefit The chain is connected, so Ν is necessary. 2* Β,, a certain frequency r; q:: sampling theorem is required. Therefore, the frequency of signals received by this system is also limited by the switching speed of the device. Alternate samples and each 1"parallel demodulator in BBP, if In the group, it is continuously allocated. For example, there are two antenna diversity selections (left, right and omnidirectional) on the right, such as 3. From the analog digitizer number 1, 2, 3, 4, 5, ό 7 8 12 The sample will be assigned as 4, 7, to the first solution «11 to the second demodulator link; and 3, 6, 9, 12 to the third demodulator link. As mentioned above, the demodulator can For coherent combination, blind signal separation or any two common multiple input multiple output demodulation techniques. This can be a combination of a single demodulation circuit or an expected spatially independent channel. The same person can make a soft decision weight or hard decision Manipulation. Some implementation limitations are: skin time theory. It includes signal noise ratio (SNR) considerations, noise index 'impedance matching, and received signal power. If you assume that the antenna array has a bandwidth that matches the received signal, the in-band signal-to-noise ratio remains the same. However, the in-band signal shovel has been reduced by a factor N2 compared to the conventional array. Because the low noise amplifier is the first active element in the signal path after the antenna array, so when the switching array starts with a simple diode

< S 49 Ϊ 2Β7366, the noise index is not an important consideration. Since each channel receives 1/N signal power after the multiplexer is demultiplexed, the low noise amplifier gain requirement is increased by 1 〇l〇g1〇N to maintain the comparable signal amplitude at the mixer output. The exchange between different antenna elements introduces changes in impedance matching characteristics. This does not always have an antenna embodiment as the "active" antenna element that is directly connected to the RF path. Other "parasitic, antenna elements only affect the RF path. ,

W • Alternative to some MIMO and other parallel path transmission schemes. The alternative embodiment is to integrate the tuned local oscillators into different carrier frequencies and switch to different antenna array diversity modes. This can be achieved synchronously or independently. It must still occur at the same time, but the states (array mode vs. carrier frequency) do not need to be in phase. This can be an 802.11g+ waveform, where two regular 8〇11 llg waveforms are transmitted on two parallel different carriers. In this case, you can alternately download the wave frequency on the local oscillator and in different field types, and the parent replaces the different antenna array diversity modes. * The mixer can be set to down convert RF waveforms to IF or to baseband DC. This changes some analog to digital converter sampling requirements. Deliberate, name, and other technologies can perform intermediate frequencies under sampling and still recover pre-emptive content. ' / This method also considers the dual use of the antenna for receiving and transmitting functions.胗 Yes: Some satellite receiving applications do not require a transfer function. For the collection and transmission is not the same time division duplex system, system (such as the milk, 50 1287366 螓

WiMAX, WCDMA-TDD, TD-SCDMA, etc.) or non-time dual-band duplex systems (such as GSM/GPRS), when the transmission mode can be considered, the receiving antenna system can be multiplexed. For a full-duplex system (such as CDMA2000 or boundary ^)^^41)1))' transmission function can be done by an independent antenna. Any of these air interfaces can use any of the enabling demodulator techniques (coherent combination, blind signal separation, multiple input multiple output). Another special policy of the present invention relates to the blind signal separation applied to the processing of the code division multiple access receiver. The antenna array that is sufficiently separated between the antenna elements is suitable for feeding the decoding link. For those skilled in the art. Other literature is discussed as single antenna interference cancellation (SAic) technology.

Surgery. These blind signal splitters are required to correlate and or statistically transform the independent I components to create a sort 2 matrix. Therefore, these decoders Γ separate a disturber and an expected signal. If there are two interferers, one is

The save-single antenna interference cancellation technique is not feasible. This is referred to as the use of a "virtual" second antenna. The I-first technique can obtain independent summation of signals by existing technology devices, and by virtue of not presenting and exploring the literature. Although the device may be implemented on some wireless access networks, it may not be suitable for code division multiple access coding. Any method that violates the mixing matrix can be considered as part of this implementation. Mb, although these techniques increase the ordering of available components for independent component analysis = which is more similar, independent component analysis applications also take the expected signal, L sub is not guaranteed. So only the detailed technique still needs to be used to select 51 1287366 for proper decoding of the bond. For example — if the right is too unfavorable to be dealt with by the knight, then you must exit 猸& 八 stomach, where I k 唬 add #一 ^, 乜 separate component analysis processing. In the case of the yoke yoke, different coding links are used. The node A is imaginary, the beer is produced 4^, and the t-group is shown in Figure 27. The 罝+ stalker is simply displayed, but _self->the stem and the heart-matrix matrix. The bottom of the noise is overtaken by narrow-band interferers, and the expected multi-access signal is below the bottom of the noise. As in the b-node at point 28, the interferer has been captured. "Device" determines that _ withdrawal is (four) is actually a disturber. If no interference occurs, then no signal is selected. If the signal has the expected signal characteristics, it is not selected. If one or more interferers are selected, they are presented to the "inverter, (C node). The independent component analysis takes the signal that can be received in the reverse or non-reverse direction, ie, whether each signal must be determined The received signal is matched in the reverse direction. The interference sub-band with the correct amplitude symbol is presented to the negative input of the D-node adder. The skilled person will of course understand the alternative, but the equivalent of the parent is feasible. For example, 'pure add-in The counter can be used in this phase, while the descendants are only used when the signal is captured in a non-reverse mode. The delayed version of the original received signal (Α node) is presented at the input of the other adder. The delay value is equal to the delay caused by the ICS, selection and "reverse" processing. The skilled person will of course understand the alternative, but equivalent implementation is feasible. For example, the delay and adder function blocks can be replaced by small blocks. , transforming and summing the two signals until the minimization is implemented. 52 1287366 At the D node in Fig. 29, the interferer has been removed. At the E node in the first picture, the 耙 type (the heart-shaped receiver) The signal has been frequency-reduced It is now presented to the baseband decoder. Further details of this embodiment are that the signals collected by the antenna structure can be obtained via selection in accordance with the previously discussed embodiments of the prior art. It should be understood that the structure shown in Figure 26 is only an implementation overview. One method of the present invention. Prior art implementations that select different paths before or after processing locations may also be used, rather than "selectors, present without any information." Replacement must be associated with processing delays, implementation costs, The overall operational robustness is somewhat related to the designer's choice. The basic concept of extracting the interfering from the signal stream only for the rake receiver must remain in all variations of the same invention. Also, although the previous explanation has been shown to be completely removed Interferers, but not all interferers can be removed. However, 'assuming the 耙 decoder can cope with the improved signal set, any interference removal will usually provide • Improved performance over the prior art. The code division multiple access signal itself is more Gaussian than its reduced version, making independent component analysis more difficult to detect. Since the signal is still statistically significant, it is possible to remove some of the material that is connected to the expected signal. The removal of the interference is usually very significant, and the entire gain is presented to the decoder. , full: the decoding process can be further enhanced by using the process increment method. It means that the signal can be checked in more detail whether it is included or excluded, and/or the number of removed signals can be incrementally increased or decreased and measured improved or Deterioration 53 1287366 Resulting the integrity of the decoded signal. ^ One of the key points of this example is that the independent component analysis is used for the L 其 it can be identified without being used for periods that are difficult to identify and/or Taking the coded multiple access signal before the processing. Another feature of the present invention relates to the mixed minimum mean square error matrix pen separation weight for blind signal processing via field type. Referring again to U.S. Patent No. 6,931,362 Where multiple sensors are required to provide a linear independent φ sum signal. The '362 patent is incorporated herein by reference. In addition to the multi-sensors, the antenna array described above can also be used, and the processing is still applicable after the disclosure of the 3 62 patent. A person skilled in the art will recognize many modifications and other embodiments of the present invention, which have the benefit of being presented in the above description and the associated drawings. Therefore, the invention is to be understood as not limited to the specific embodiments disclosed, and modifications and embodiments are intended to be included within the scope of the appended claims.

54 1287366 BRIEF DESCRIPTION OF THE DRAWINGS The block diagram of the scheme, the expected and unintended signal of its source. Fig. 1 shows a typical communication device according to the present invention receiving signals from its respective signals. Figure 2 is a more detailed block diagram of the communication device shown in Figure ith. Figure 3 is a diagram showing the different methods for creating a linear independent summation of the source signals for the hybrid matrix in accordance with the present invention. The # 帛 4 diagram is configured according to the present invention # for beam switching. Antenna array block diagram. Figure 5 is a block diagram of a day array configured as a phase array in accordance with the present invention. The array is a three-polar three-dimensional diagram of an antenna configured with a polar antenna element in accordance with the present invention. Blind signal separation processing for communication of antenna elements is provided. FIG. 7 is a view of the use according to the present invention. FIG. 8 is a diagram according to the present invention.

Block diagram of associated and unrelated devices for different signal summations. Figure 9 is a block diagram of a communication device providing different "wide array bias" operation for blind signal separation processing according to the present invention. Brother 1G® is a beam having a selectively changeable antenna pattern = towel angle controller according to the present invention. Switching the antenna block diagram is: Π糸 is the antenna field pattern in the azimuth direction and then:: the angular direction is rotated in response to the day of the elevation controller shown in Fig. 9 1 1 287366. FIG. 12 is a representation according to the present invention. The type is formed in the ground: the antenna angle is rotated in the direction of the elevation. The antenna element block of the "suppression loop" is shown in Fig. 13 as the basis for selecting the path of the blind signal in accordance with the present invention = different 14th The figure shows a communication device based on the f signal separation processing = two-card number ^ total expansion code according to the present invention. Port 15 is a two-character block diagram based on the in-phase and quadrature signal components of the blind signal component in accordance with the present invention. Fig. 16 is a more detailed block diagram of the antenna element and the orthogonal module shown in Fig. 15. » 17® is a block diagram of an input multiple output system based on field diversity in accordance with the present invention.夕 夕 Figure 18 is a block diagram showing a symbol-switched communication system in accordance with the present invention. Tian Licha, a Figure 19 is a block diagram of a communication system in accordance with the present invention, wherein the transmitter changes the power level of each hierarchical spatial stream on a time-sharing basis. Figure 20 is a block diagram of a communication system in accordance with the present invention in which a wave field type is used to support transmission of multiple transmitters to the same access point. / Figure 21 is a block diagram showing the optimum processing and power square of the receiver in accordance with the present invention. Fig. 22 is a block diagram of the receiver shown in Fig. 21 for coordinating its operation and transmitter. ^

S 56 1287366 Benefits of known timing undulations Figure 23 is a diagram of the transmission field pattern profile in accordance with the present invention. Figure 24 is a block diagram of a receiver in accordance with the present invention having 12 variables (i.e., a 12-wafer wafer fixed time line in which the symbol period is 丨, and the changed parameters are 4 according to the present invention. Figure 26 for a plurality of spatially independent channels is a block diagram of a decoder decoding chain according to the present invention. Figures 27 to 30 are amplitude versus frequency patterns corresponding to nodes a B, D and E in Fig. 26, respectively. Component Symbol Description 20, 22 Signal Sources 20(1) to 20(M) 1st to 2nd Signal Sources 22(1) to 22(Μ) 1st to 3rd Received Signal Sources 24 Antenna Beam 24(1) )~24(Μ) 1st to 2nd antenna beam 30 communication device. 32 Antenna array 34(1) to 34(Ν) 1st to 2nd antenna elements 36 Mixed matrix 3 8 , 3 8 (1 ) - 3 8 (3) Separation matrix 39 Separated signal 40 Transceiver 57 1287366 42 Processor 44 47 49 50 52 100 102 104 106 108 110 Principal component analysis module Independent component analysis module Single value decomposition module Signal analysis module Application dependent processing Module unassociated sensor associated antenna array polarized antenna array sensor and array combined array bias path Select 112 spread spectrum 114 I and (3 components 116, 214, 252, 312, 412, 512 hybrid matrix 140, 100' beam switching antenna 142, 162 active antenna elements 144, 104' passive antenna elements 144a, 104a' upper half 144b 104b' lower half 146, 106', 272 ground 148, 108', 118' reactive load 160 phase array 162 active antenna element 58 1287366 164 weight control component 168, 306 controller 180, 202, 242, 302, 40 2 Antenna array 206 210, 248 212 216 246 Two associated antennas Where the receiver blind signal separation processor is separated signal elevation controller

250, 310, 410, 510 processor 254, 314, 414, 514 split signal 270 control radio frequency adjustment device 272 ground 308, 408, 508 receiver component 316 target receiver 406 code frequency reducer 506 in-phase and quadrature Module 520 Mixer 182a, 182b, 182c, 184a, 184b, 204, 244, 3 04, 404 502, 504, 274 Antenna Element 200, 240, 300, 400, 500 Communication Device PC A Principal Component Analysis ICA Independent Component Analysis SVD Single Value Decomposition RR 接收 Receiver 59 1287366 RF RF LNA Single Low Noise Amplifier LO Local Oscillator LPF Low Pass Filter ADC Analog to Digital Converter

60

Claims (1)

1287366 X. Patent application scope: The source signal is connected to receive the VI. 1. A signal device that can be separated by one signal source, the communication device comprises: an antenna array comprising one antenna element, a source signal At least one different sum; individual in-phase and orthogonal mode π π 八 兀 And determining that each of the different sources of the received source signals is an in-phase and quadrature component group; a receiver component is connected downstream to each of the in-phase and quadrature modules to receive the And the at least one of the at least one in-phase and quadrature component groups of the plurality of source signals; and a blind signal separation processor coupled downstream to the receiver component to form at least the source signal 2 不同 different mixed matrix 'the respective in-phase and orthogonal component sets provide 2 inputs to: a mixing matrix having a rank equal to 2 ,, the blind signal separation processor can combine the expected source signal with the hybrid matrix For example, the communication device of claim 1 of the patent scope, wherein 2Ν=Μ. 3. The communication device of claim 1, wherein the order of the ::: is κ, where Κ&lt;2Ν, and The blind signal separation can separate the κ of the source signals from the hybrid matrix. 4. The communication device of claim 1, wherein 2 Ν &gt; 61 1287366 6. The communication device of claim 5 ,among them The n associated antenna elements comprise N active antenna elements such that the antenna array forms a phase array. 7. The communication device of claim 5, wherein the N associated antenna elements comprise at least the active antenna And the antenna array is configured to switch the antenna. The communication device of claim 1, wherein at least two of the N antenna elements are associated and different 9. The communication device of claim 8, wherein the different polarities are orthogonal to each other. 1〇. The communication device of claim 1 of the patent scope further includes an elevation controller 3 for selective Changing the less-increasing angle such that the one source signal is at least == 2 (four) is received; wherein one of the in-phase and quadrature modules is connected downstream of the antenna element from the output beam to the J-additional beam, separating: At least one additional difference is added to the in-phase and quadrature: component=where the receiver component receives the at least one additional in-phase and positive-in-one group; and the blind signal is separated Forming the mixing matrix ' to include the at least one additional in-phase and quadrature component group. As in the communication device of claim 10: the bean: the angle controller selectively changes the elevation angle of the fixed antenna field type, Let Ilf add an antenna pattern to receive the source signal "additional different sums, the ordering of the mixing matrix is now equal to 4N. 62 1287366 '· κ - 12. The communication farm according to claim 1, wherein the elevation controller is divided into p regions relative to the N antenna elements, each of which is independently controlled to change the The solid antenna field type allows N additional antenna field patterns to be generated to receive N additional different sums of the m source voxels, the ordering of the mixing matrix is now at 4PN ° 13 · as claimed in the first item The communication device, wherein the array of spring lines comprises a ground; and wherein the N antenna elements comprise: an active antenna element adjacent to the ground; a plurality of passive antenna elements adjacent to the ground, each passive antenna element comprising An upper half and a corresponding lower half, an upper variable reactance load connecting the upper half to the ground to change an antenna field azimuth; 4 the elevation controller includes a lower portion for each passive antenna element A variable reactance load connects the lower half to the ground and moves the N値 antenna pattern in an elevation direction by adjusting at least one of the lower variable reactance loads. 14. The communication device of claim 10, wherein the antenna array comprises a ground adjacent to one of the N antenna elements; and wherein the elevation angle comprises a controllable RF suppression coil coupled to the ground. The N antenna beams are moved in the elevation direction by controlling the RF suppression coil. 15. The communication device of claim 1 of the patent scope, wherein the day 63 1287366 ΙΙΗ· ' :洲心(10)~, 幻,灿似·|9 The more interesting, 、, t array can form at least N antenna beams And further comprising a controller coupled to the side antenna array to selectively form the at least N, milk beams, N and M are greater than 2; wherein the blind signal separation processor also determines different ones of the source signals Whether the total is associated or statistically independent, and if not, then operating with the edge controller to form a different beam that can receive the μ source signal φ, the new difference is added, instead of being uncorrelated or statistically independent in the mixing matrix The different sources of the source signal are summed up. 16. The communication device of claim 15 wherein the $receiving component comprises a rake receiver, and each of the rake receivers comprises a snap button for generating the source signal. Each of the different sums of μ plus a plurality of different multi-path components, the one source signal being received by the antenna elements connected thereto; and wherein the blind signal separation processing system is connected to the one-handed receiving And the mixing matrix comprises the at least two different sums of the one source signals to at least 2 different multi-path components, the n-combination matrix having a rank equal to 2ΡΝ. 17. The communication device of claim 3, further comprising a code downconverter coupled downstream to the one of the antenna elements to decode the at least one different sum of the one source signals, the v different Each of the k codes is included to provide k different sums of the one source signals connected to the respective in-phase and orthogonal modules, and the kNs of the received ones of the source signals are different. Separating into ^ 64 ! 287 366 in-phase and quadrature component groups; and wherein the blind signal separation processor forms a chirped mixture matrix to include the k in-phase and quadrature component groups such that the hybrid matrix has an order equal to 2 kN . The communication device of claim </ RTI> wherein the death signal separation processor has at least a principal component analysis (pCA), an independent component octave (ICA), and a single value decomposition (SVD). Based on one, the &quot;expected source signal is separated from the hybrid matrix. / / Lu ^ a method of operating a communication device to separate source sources provided by m signal sources 'the communication device includes an antenna array, a downstream connected to the antenna element in-phase and orthogonal modules, a downstream a receiver component coupled to the in-phase and quadrature module and a blind signal separation processor coupled downstream to the receiver component, the method comprising: receiving at least N different sums of the one source signals at the antenna array The antenna array comprises N antenna elements; using a respective in-phase and quadrature lu-module connected downstream to an antenna element to separate at least N different sources of each of the adjacent source signals into an in-phase and quadrature component group; Providing at least one different in-phase and quadrature component groups for the at least one different sum of the source signals to the receiver component, and processing by the blind signal separation processor for the at least one source signal Differently adding the at least one in-phase and quadrature component groups, the processing comprises: forming a mixing matrix, the hybrid matrix comprising at least two different sums of the one source signals, each in-phase and positive Group-based components provide 65 I287366 &gt; P 2 is input to the mixing matrix, the mixing matrix having a first-order, and would be expected to separate a source signal with the mixing matrix. 2. The method of claim 19, wherein 2N=M 21. The method of the patent scope (1), wherein the mixing is K' where K&lt;2N, and the blind signal is separated The κ of the source signals can be separated from the mixing matrix. 22. The method of claim 19, wherein the method of claim 19, wherein the antenna (4) comprises N associated antenna elements. Wherein the N jl is such that the N phase. 24. The method of claim 23, wherein the associated antenna element comprises N active antenna element arrays to form a phase array. 25. The method of claim 23, wherein the associated antenna element comprises at least an active antenna element and at most one passive antenna element 'such that the antenna array forms a beam switching line 0, ^6. The method of claim 19, wherein at least two of the N Tianxin elements are associated and have different polarities. The method of claim 19, wherein the communication assembly includes an elevation controller that selectively changes an elevation angle of the at least one of the at least one of the lines to generate the M source signals At least one additional different summation is received, wherein - the individual in-phase and quadrature modules are connected downstream from the antenna element generating the 66 1287366 two: additional beam, separating the one source source nickname by at least one additional different summing An additional in-phase and quadrature component group; wherein the receiver component receives the at least one additional in-phase and quadrature component group; and wherein the blind signal separation processor forms the hybrid matrix to include the at least one additional in-phase and quadrature Component group. The method of claim 27, wherein the elevation control selectively alters the elevation angle of the one antenna pattern, such that an additional antenna pattern can be generated to receive the additional source signals. In total, the ordering of the mixing matrix is now equal to 4Ν. The method of claim 27, wherein the elevation controller is divided into opposite ones of the antenna elements. Areas, each of which is independently controlled to change the amplitude of the one antenna pattern = the transmission can generate an additional antenna pattern to receive the additional source sum of the source signals The method of claim 19, wherein the antenna can form at least one antenna beam; the communication device further includes a controller coupled to the antenna array to selectively form the to-line beam 'Ν and Μ greater than 2 Where the blind signal &lt; is still determined (4) whether the different sums of the source signals are related or erected, and if not, then 7 and the (four) device operate together to form a new difference that can be received by the source The different beams are summed to replace the different sums of the M source signals that are not associated or statistically independent in the mixing matrix 67 1287366 _ : 乂 2...W..:.vt. 31. The method of claim 3, wherein the receiver component comprises N rake receivers, each rake receiver comprising 1&gt; fingers for generating the N of the M source signals Each of the different summed plurality of different multipath components, the one source signal being received by the antenna elements connected to $; and wherein the blind signal separation processing thief is connected to the one a receiver to form the mixing matrix, the mixing matrix comprising the at least one of the plurality of source signals and the method of adding at least 2 different multipath components, and the hybrid matrix and the method of claim 19 , wherein the communication J step comprises a coded frequency reducer, and the downstream is connected to the one antenna element to decode the one source, the source, the μ 曼 曼 曼 Μ Μ Μ Μ Μ Μ Μ Μ , , , , , , , , The different sums each include k-codes ^ ^ ^ ^ ΛΛ 1 , οη to provide the Μ : = 细 细 细 细 细 = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = The component matrix and the sum of the sub-dividers can form the hybrid matrix: the packet number is separated Li groups, so that the mixing matrix having a first-class;:,: two-phase and quadrature components to achieve dry nettle sort of 2kN. 33. If the scope of the patent application is Μ, the separation processor is analyzed by the principal component analysis #T method, and the blind signal is separated. Based on the basis, the expected source signal and the mixing matrix are divided into 68 1287366. 7. The representative representative figure: (1) The representative figure of the case is: (2) Figure (2) The symbol of the symbol of the representative figure is simple: 30 Communication device 34 ( 1)~34(N) 1st to Nth antenna elements 32 Antenna array 36 Mixed matrix 38, 38(1)-38(3) Separation matrix 39 Separated signal 40 Transceiver 42 44 46 48 50 φ 52 PCA ICA SVD Processor principal component analysis module independent component analysis module single value decomposition module signal analysis module application dependent processing module principal component analysis module independent component analysis module single value decomposition module 8. If there is a chemical formula in this case, please Reveal the chemical formula that best shows the characteristics of the invention: