TWI284465B - Blind signal separation using correlated antenna elements - Google Patents

Abstract

A communications device for separating source signals provided by M signal sources includes an antenna array comprising N correlated antenna elements for receiving at least N different summations of the M source signals, with N and M being greater than 1. A receiver is connected to the antenna array for receiving the at least N different summations of the M source signals. A blind signal separation processor is connected to the receiver for forming a mixing matrix comprising up to the at least N different summations of the M source signals, and for separating desired source signals from the mixing matrix. The mixing matrix has a rank equal to at least N.

Description

'1284465 IX. Description of the invention: [Technical field to which the invention pertains] The present invention relates to the field of signal processing. In particular, the present invention relates to the application of blind signal separation (BSS) techniques to separate desired source signals from mixed source signals. [Prior Art] Φ Blind Signal Separation (BSS) includes a reply source signal from a self-mixing signal, wherein the mixed signal includes a mixed source signal. In this way, the signal is separated, and it is "blind". Because of this "signal separation," the implementation usually only has the limited information of the signal, the source of the signal, and the effect of the transmission channel on the signal. An example is familiar. The "Chicken Dance Party," the effect, that is, the members of the ball can separate a single voice from all the voices in the house. In particular, the blind source separation (BSS) system can be applied to mobile phones and personal wireless communication devices. When many frequency bands are filled with various radio frequency transmitters, various radio frequency transmitters usually coexist in the same frequency spectrum. Under the trend of low-power, unlicensed wireless technologies (such as Bluetooth) and other personal area networks, the problem of common channel transmitters in the next few years seems to be only worse. The three commonly used blind signal separation (BSS) techniques are principal component analysis (PCA), independent component analysis (10), and single numerical decomposition (8\〇3). The main component analysis (PCA) includes the first and second momentum statistics of the source signal, and the principal component analysis (PCA) can be applied to the signal to noise ratio (SNR) of the source signal. . In addition to this 1284465, the Independent Component Analysis (ICA) system can be applied, which includes the main component analysis (PCA) processing, and follows the third and fourth momentum statistic of the source signal. Alternatively, a single numerical decomposition (SVD) can be applied to separate the desired source signal from the mixed source signal based on the eigenvalue. Regardless of the blind signal separation (BSS) technology applied, the complex sensor system can be applied to receive different mixed source signals through various signal sources. Each sensor is a mixed source signal, which is the unique sum of the source signals. In general, neither the channel coefficient nor the original source signal is known to the receiver. The unique sum of the source signals can be used to fill the hybrid matrix. Subsequently, the appropriate blind signal separation (BSS) technique can be applied to the hybrid matrix to separate the desired source signal from the mixed source signal. In one example, U.S. Patent No. 6,679,917 discloses a method of applying independent component analysis (ICA) to separate independent source signals from mixed source signals. The complex sensor receives the mixed source signal, and the processor samples the mixed source signal over time and stores each sample as a data vector to generate: New #合。. Each _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Independent Component Analysis (ICA) Module _ Implementation of the independent component analysis of the data vector to separate the independent source signal and other signals from the mixed source signal. The sensors are spatially separated from each other, and the processor generates a single data vector for each _ to generate a data set. In addition, the US patent number 脳799 also reveals that the number N of sensors is equal to or greater than the number M of sources to fill the data set. The problem with this 1284465 implementation is that as the number of sources M increases, the number of sensors... increases. Small portable communication devices have almost no usable capacity to accommodate large numbers; N, and the provision of a wire outside the portable communication device may also cause user confusion. In view of this, U.S. Patent No. 6,931,362 discloses another method by which a blind signal separation (BSS) technique is applied to separate the desired source signals. The disclosed blind signal separation (BSS) technology forms a hybrid matrix with hybrid matrix - pencil adaptive array weights, thereby minimizing the simultaneous interference based on the chirp and the muse noise. The squared error is minimized. Hybrid matrix-pencil adaptive army weights minimize signal interference noise ratio (SINR). Similar to U.S. Patent No. 6,799,170, the sensors are also spatially separated from each other, and the number N of sensors is equal to or greater than the number of sources μ, i.e., NgM, thereby filling the mixing matrix. In addition, each sensor provides a single input to the mixing matrix, resulting in a larger volume of portable communication devices. SUMMARY OF THE INVENTION In view of the above background, it is an object of the present invention to provide a communication device that includes a small antenna array that receives mixed source signals to provide blind signal separation (BSS) technology applications. By separating the desired source signal from the mixed source signal. The above and other objects, features, and advantages of the present invention can be provided by a communication device via 1284465, thereby separating the source signals provided by the [V source], wherein the communication device includes an antenna array for receiving M The sum of the different sources of signals. A receiver or receiver component is coupled to the antenna array and a blind signal separation (BSS) processor is coupled to the receiver to form a mixing matrix. The hybrid matrix includes different sums of the source signals received by the antenna array. Subsequently, the blind signal separation (BSS) processor separates the desired source signal from the mixing matrix. A small antenna array can be applied, relative to the spatially separated sensors providing a different sum of signals to form the hybrid matrix. In view of this, Blind Signal Separation (BSS) technology can be applied to portable communication devices because the antenna array can provide more than one set of inputs to the hybrid matrix and can simultaneously maintain a small size. In particular, the antenna array can include a plurality of off-speed antenna elements to receive at least one different sum of the plurality of source signals, wherein the Ν and Μ are greater than one. The blind signal separation (BSS) processor can form the hybrid matrix, wherein the hybrid matrix includes at least one different sum of the plurality of source signals. The hybrid matrix may have a rank equal to at least one of the ranks. The number of antenna elements may be equal to the number of source signals, i.e., . Or 'the number of antenna elements can be greater than the number of source signals, ie: N> Μ. In another architecture, the rank of the hybrid matrix is equal to κ, where K<N, and the blind signal separation (BSS) processor separates the source signals from the hybrid matrix.値 Source signal. 1284465 N connected antenna elements can include N active antenna elements, whereby the antenna array can form a phase array. Alternatively, the N connected antenna elements may include at least one active antenna element, and up to (n-D passive antenna elements, whereby the antenna array may form a switched beam antenna. When receiving different sums of the source signals, The field type and the inter-beam system may make a difference. In this example, the antenna array may form at least one antenna beam to receive at least one different sum of the two source signals, wherein each antenna beam has a maximum gain point. The lower point further provides a signal rejection of the arrival signal in at least one direction. In another example, the antenna array forms at least one antenna field type to receive at least one sum of the different sums of the source signals of the country, wherein At least one antenna field type generally does not have a 3dB point below the maximum gain point, and thus cannot provide a signal rejection of any arrival signal in any direction. The sum of each source signal is linear. The blind signal separation (BSS) Processing of fasting can be based on principal component analysis (pCA), independent component analysis (ICA), and single values At least one technique of solving (SVD), and then separating the desired source signal from the mixing matrix. Another feature of the present description relates to a method for operating a previously defined communication device to separate M The method of providing a signal source, the original nickname, may include: applying at least one different sum of the source signals by the antenna array, wherein the antenna array includes one antenna element, and Ν and Μ In the case of - source signals, the different sums are provided to the receiver. The blind signal separation (BSS) processor processes at least N different 1284465 \ sums of the M source signals received by the receiver. The processing of the separation (BSS) processor may include a formation-mixing matrix comprising up to at least N different sums of the M source signals, and separating the desired source signals from the mixing matrix. The matrix system may have a rank number equal to at least one of N. [Embodiment] Now, the present invention will be described in detail with reference to the preferred embodiments and the accompanying drawings. However, the invention may be embodied in many different forms and should not be limited to the preferred embodiments described herein. In contrast, the preferred embodiments are provided so that this disclosure will be thorough and complete and will be fully conveyed by those skilled in the art. In addition, in the preferred embodiment of the invention, the same number denotes the same elements, and the 'numbers' denotes similar elements of the preferred embodiment. In a communication network, the desired source signal of a particular communication device may be present and the desired source of other communication devices operating in the same frequency band may also exist. In addition, the sources of noise will also exist at the same time. Although they will not generate signals for communication, they will still be received by individual communication devices. To facilitate the decoding of important source signals, the Blind Signal Separation (BSS) technology can be applied to separate the source signals that a particular communication device is intended to receive. As previously mentioned, the term, blind, means that, in the ideal case, the source signal can be separated without knowing the characteristics of the source signal or based on the interaction between the source signal and the communication channel. In practical applications, any knowledge that can be provided can be applied. In view of this, in this example, the separation of the signal 11 1284465 is not blind (semi-blind). Three common techniques for Blind Signal Separation (BSS) technology include: principal component analysis (PCA), independent component analysis (ICA), and single numerical decomposition (S VD ). As long as the measurable features of the source signal are independent, and 'as long as the sum of the source signals is linearly independent of each other, single or multiple blind signal separation (BSS) techniques can be applied to separate from the mixed source signal. Source signal. In general, the feature can be measured as the first, second, third, or fourth momentum of the source signal. The principal component analysis (PCA) is a bleach source signal, applying the first and second motion differences, and rotating the data set based on the related characteristics. If the signal of the source signal is mixed, the signal separation process can stop at the principal component analysis (PCA). If the signal noise of the source signal is low, the independent component analysis (ICA) can be based on the third source signal. And the statistical properties of the fourth motion difference, and then the desired source signal is separated. Since the source signal exhibits a Gaussian distribution, the third and fourth motion differences are dependent on the first and second motion differences. In addition to principal component analysis (PCA) and independent component analysis (ICA), a single numerical decomposition (SVD) can also be based on the eigenvalues, and the desired source signal can be separated from the mixed source signal. Figure 1 shows a typical scenario in which the source of multiple sources can be applied to transmit the source signal. The source signal 2^ is transmitted along the direction of the connected antenna beam 24 of the individual signal source 20. The plural signal source 20 series includes f-service source 2G(1) to the first signal source 20 (M). Similarly, the 'individual source signal' is expressed as the first source signal ^ 12 1284465 (1) to the third source signal 22 (Μ). In the communication network, a more intuitive implementation with an omnidirectional antenna field and a directional antenna field form can also be applied. In communication device 30, antenna array 32 receives a linear combination of source signals 22 (i.e., mixed source signals) via signal source 20. Antenna array 32 includes a plurality of antenna elements 34, wherein each antenna element provides at least one linear combination of source signals 22 via signal source 20. The antenna element 34 has a first antenna element 34(1) to a νth antenna element 34(Ν). First, the source signals 22(1) to 22(Μ) are received to form a mixing matrix 36. Next, the communication device 30 applies blind signal separation (BSS) techniques to determine the separation matrix 38, thereby separating the desired source signals from the mixing matrix. The separation signal is expressed as the number 39. The communication device 30 collects or mixes the received source signals (without knowing the characteristics of the source signals), and collectively captures the mixed source signals received by the antenna array 32. The input of each antenna element 34 can be modeled as the sum of the source signal 22 after convolved by the channel impulse response (the channel represents the transmission path between the output of the signal source 20 and the output of the antenna element 34) plus additional Gaussian (Guassian) noise. The communication device 30 that separates the source signals provided by the IV signal sources 20 (1) to 2 (M) will be further described in detail below with reference to FIG. The antenna array 32 includes N antenna elements 34 (work) to % (N), thereby receiving up to a different sum of the source signals, wherein the μ system is greater than one. Antenna array 32 need not be limited to any particular architecture. The antenna array 13 , 1284465 3^2 may comprise a single or multiple antenna element %. The individual antenna elements % are shot in a suitable architecture, whereby the scales 32 can be formed, for example, a phase array or a switched beam antenna, which will be described in further detail below. The 0 transceivers are connected to the antenna array 32. Streaming, so as to receive a source of 峨22 up to a different sum. Processor 42 is coupled to the downstream stream of transceiver 40. In FIG. 2, processor 42 may be included within transceiver 40, even though processor 42 is independent of transceiving $40. The different sums of the M source signals 22 received by the transceiver 4 can be applied to fill the mixing matrix 36. Hybrid matrix 36 can then be processed by single or complex blind signal separation (BSS) processing modules 44, 46, and 48 within processor 42. These single or multiple blind signal separation (BSS) processing modules include a primary component analysis (PCA) module 44, an independent component analysis (ICA) module 46, and a single numerical decomposition (SVD) module 48. These single or multiple blind signal separation (BSS) processing modules 44, 46, and 48 can be suitably architected to form part of a blind signal separation (BSS) processor 49. The principal component analysis (PCA) module 44 is based on the first and second motion differences of the different sums of the received source signals. In contrast, the independent component analysis (ICA) module 46 is based on the different sums of the received source signals. Three and fourth motion operations. In addition, the single numerical decomposition (SVD) module 48 performs signal separation based on the eigenvalues of the different sums of the received source numbers. First, the related component of the 'Principal Component Analysis (PCA) module 44 determines the start separation matrix 38 (1) that receives the different sums of the source signals, and the 'Independent Component Analysis (ICA) module 46 after 14 ^ 284465 k It is decided to strengthen the separation matrix 38 (2) by which the desired source signal is separated from the mixing matrix 36. If the source signal is separated by a single numerical decomposition (SVD) module 48, the separation matrix 38 (3) may also determine the different sums of the received source signals separated by the self-mixing matrix 36. The separation signal is represented by the number 39 via the individual separation matrices 38 (1) through 38 (3). Subsequently, the separation signal 39 can be applied to the signal analysis module to perform signal analysis to determine which signals are important signals and those signals are interference signals. The application-related processing module 52 processes the signals that the signal analysis module 50 rotates. The decision that the signal is an important signal may not always contain the last signal to be decoded. For example, individual applications may need to identify the source of the interference and remove it from the different sums of the received source signals, and then feed the signal from the source of the interference to the waveform decoder. In this case, the important signal is the signal that is ultimately rejected. • The information fed into the Principal Component Analysis (PCA) module 44 is the unique signal sum Xj. Suppose that N linear mixtures of independent components, Xl,..., are expressed as follows: XI (t) ^ansi (t) +...aiksk (t) +*..aiMSM (t) xj (0 =^lsl (t) +...^kSk (t) (t) XN (t) = aNisi (t) (t) +...aNMSM (t) In general, the transceiver 40 does not know the channel coefficient 〇 Jk and the original signal sk. Applying the matrix representation, the above equation can be simplified and rewritten as x 15 1284465 = As, where A is a mixed matrix. Statistically, the model x = As can also be called independent component analysis (ICA). Model. The conventional technique attempts to find the inverse of the channel, ie: s = A — The Independent Component Analysis (ICA) module 46 can determine the separation matrix w, and y = W(As) = Wx. Vector y is a subset of s with blind order and scaling change. If all signals are inseparable, the general form of vector y can be expressed as y = W (As) + Wn = Wx + Wn, where extra | The η term indicates the residual noise based on the unrecognized source. The Independent Component Analysis (ICA) model generates the model, that is, the Independent Component Analysis (ICA) model describes how the observed data is mixed by the component %. The combination process is generated. The independent component is a potential variable, that is, the independent component cannot be directly observed. In addition, the hybrid matrix A is assumed to be a blind matrix. From the above, only the random vector x can be directly observed. In contrast, both A and s are based on the predicted value of X. The starting point of independent component analysis (ICA) is assumed to be independent of the independent component to the system. In addition, the independent component % also assumes at most Gaussian distribution. (Guassian distribution) one of the signals., Gaussian distribution ((9) move as diSt_tion) one signal "limit is based on the fact that the second motion difference of the Gaussian signal is equal to zero, and the fourth motion difference is not recognized by the Gaussian signal. For the sake of simplicity, the Blind Mixing Matrix A is assumed to be a square matrix. By this, the number of independent components is scaled to observe the number of mixed signals. However, the limitations of this assumption can sometimes be lifted. The specific measurable features are statistically independent, and the separation matrix w can be determined. 16 1284465 The rank of the mixed matrix A is determined by the actual decision. For example, the mixed matrix with a rank of four means that four source signals can be separated. In theory, the rank of the mixed matrix A should be at least the number of signal sources. The larger the rank, the more source signals can be separated. As the number of signal sources increases, the number of antenna elements needs to increase. In the background of the invention, U.S. Patent Nos. 6,799,170 and 6,693,362 disclose that the number of antenna elements should be equal to or greater than the number of signal sources Μ 'that is, otherwise signal separation should use other than blind signal separation (BSS) technology. technology. An industry standard that produces a linear independent sum of source signals is the application of non-contact sensors, i.e., each sensor is spaced apart from each other by at least one wavelength. This wavelength is based on the operating frequency of the communication device 30. ν sensors are unrelated in space, but there is a correlation between polarity and angle. A non-connected sensor provides a sum of linear independent signals, where each sensor provides a single item of the hybrid matrix. Figure 3 is a diagram showing a plan or overview of different sources in accordance with the present invention to produce a linear independent sum of the source signals of the mixing matrix A. After a brief reminder, the various means will be further elaborated below. The first section of the plan map proposes an antenna architecture. Block 1 represents a non-connected sensor' wherein each sensor provides a single input to the hybrid matrix A. Block 102 represents a non-connected antenna array in which a non-connected antenna array provides a complex input to fill the mixing matrix A. In addition, the square bar 104 also represents an antenna array in which portions of the antenna elements are related, and individual antenna elements have different polarities, #to fill the mixing matrix 1284465 A. The different combinations of the inductors and antenna arrays proposed by blocks 1〇〇, 102, and 104 can be combined with block 106 to further fill the mixed matrix of block 〇6, and the second segment of the plan is presented first. The enhancement of the antenna architecture provided by the block. These enhancements can be implemented to further fill the hybrid matrix A by collecting additional or alternate sums of source signals. Block 108 includes an array deflection operation in which the height of the antenna pattern can be varied to receive an additional sum of source signals. Any combination of blocks 1-6 can be applied to block 108 for performing array deflection operations. In block 110, path selection may be performed such that all sums of the source signals applied to fill the hybrid matrix A may be related (first and second motion differences) and/or statistically independent (third and third) Four motions). In other words, the incoming signal can be selectively selected to replace the absence of a speed and/or a statistically independent sum. Block 110 can be fed into any combination of block 106 and block 108. In addition, blocks 1〇8 and ι1〇 can be fed directly into the mixing matrix block 116. The third section of the plan map proposes signal segmentation to further fill the mix matrix A of block 116. For example, block 112 applies a spreading code to split the different sums. If the sum signal has k spreading codes, this particular sum signal can be processed, and then < k for the associated sum signal. Spread spectrum (Sprea (jing) and jing (7) and) can be combined with the output of blocks 106, 108, and no. Block 114 can divide the different sum signals into in-phase (j) components and orthogonal (q) 18 J284465 components. Further filling the mixing matrix A. Thereby, the in-phase (〗) component and the positive parent (Q) component can be used as the multiplier of the missing matrix, and the in-phase (I) component and the orthogonal (q) component can be matched. The outputs of blocks 1〇6, 1〇8, 110, and 112 are combined and applied. The last segment of the plan graph represents the blending matrix a formed by block 116. As shown in the plan diagram, 'based on any of the previously described blocks, the blending matrix The A system can fill different sums of the source signals. The antenna array architecture of the first segment has the advantage that the small antenna array can be formed to fill the hybrid matrix A. The second segment and the third segment are The advantage of the antenna array architecture is that N antenna elements (where the number of antenna elements N is smaller than the number of source signals M) can be applied, and then the M or more different sums of the source signals are used to fill the hybrid matrix A. An antenna array as shown in the drawing, an antenna array will be discussed in further detail below, wherein the antenna array includes N connected antenna elements to receive at least a different sum of the source signals and wherein The Ν and Μ are greater than one. In a preferred embodiment, the antenna array is a switched beam antenna 14 第 as shown in Figure 4. The switched beam antenna 140 can generate a complex antenna pattern, including: directionality Antenna field type and omnidirectional antenna pattern. The switching beam antenna 14 has an active antenna element 142 and a pair of passive antenna elements 144. The actual number of active antenna elements 142 and passive antenna elements 144 can be applied according to the desired application. For a more detailed description, please refer to U.S. Patent Application Serial No. The content of this U.S. Patent Application is hereby incorporated by reference. 44a and lower half 144b. The upper half 144a of the passive antenna element 144 is coupled to the ground 146 via a reactive load 148. The reactive load 148 is a variable reactance that is applied to a varactor, transmission line, or switching. To change the capacitance-to-inductance ratio, the radiation field system can be changed via the varying reactive load 148. Since the switching antenna element 140 has two passive antenna elements, four different antenna field types can be formed. Before the four different antenna patterns The three antenna field types can be applied to receive the unique sum of the sfL numbers. The fourth antenna field of the four different antenna patterns is a linear combination of the other three antenna patterns. Therefore, the fourth antenna pattern is Cannot be used as a hybrid matrix A project. It can be seen that with three antenna elements, the three unique signal sums Xj can be input to the mixing matrix A. The advantage of switching the beam antenna is that three antenna elements 142 and 144 are applied, and a mixed matrix system having a rank of three can be supported. In another preferred embodiment, the antenna array includes N associated active antenna elements to enable the antenna array to form a phase array 160, as shown in FIG. Phase array 160 includes a plurality of active antenna elements 162 and a complex weight control element 164 coupled to the active antenna elements. Weight control component 164 adjusts the amplitude and/or phase of the received signal to form a hybrid beam. Splitter/combiner 166 and controller 168 are coupled to weight control element 164. Please refer to US Patent No. US6473036, which is further detailed 20 1284465 ί 种 = active phase array 16°. This U.S. Patent and the present patent application are hereby incorporated by reference. The number of active elements 162 is supported by the same rank number (the mixed matrix A of the service is even if the number of sources M is equal to the number of active elements n, and also the active phase array (10) can still be kept small because In the conventional technique of flipping the antenna elements (the non-connected antennas are more than one wavelength apart), the active elements 162 are related in terms of space and polarity. In other preferred embodiments, the rank of the mixing matrix Α (rank) can be equal to κ (where 'K<N), whereby the blind signal separation (bss) processing family 49 can separate the κ source signals from the source signal from the mixing matrix a. In the preferred embodiment, it will be further described in detail below, and Ν may also be larger than μ. Regardless of the switching beam antenna Mo or the phase antenna 16 ,, the distance between the individual antenna elements 142, 144, and 146 may be appropriately set, thereby Rong Hui ideals the back to front ratio. This is based on the typical application of these antenna arrays to reject unwanted signals (ie, back arrival) and to strengthen the desired message. (ie, the front arrives.) However, in order to establish a hybrid matrix, the primary goal of the present invention is to generate a sum of different signals. In fact, important signals may always be below the source of interference for such applications and still be separated. Based on the significant difference in purpose, the distance between the antenna elements does not need to have a specific spacing. The antenna elements can be farther apart or closer to each other, and the antenna element 21 1284465 is tied to the "front" ratio of the application technology. (迕(10)t t〇back ratio) produces field patterns that are still suitable for hybrid matrix applications. In fact, in blind signal separation (BSS) applications, this type of field is usually very large. The front t〇back ratio is required to trace the direction to maintain the front pointing to the desired signal and/or to maintain the rear pointing to the source of the interference. In contrast, there are differences in all directions. This type of signal tracking system is no longer needed. The antenna beam system can be defined as having a lower than maximum gain. A point of 3 dB, whereby the signal rejection is provided to the arrival signal of at least one direction. Similarly, the antenna beam can also be defined as having substantially no point below the maximum gain point by 3 dB, so that the signal rejection cannot be provided to any direction. In many applications, the deviation of the specific distance between the antenna elements can greatly reduce the size of the overall antenna array. In other applications, the antenna element can be increased, and (10) and tracking In some other preferred embodiments, the antenna array 18 includes N antenna elements for receiving a plurality of source signals. To a different sum, as in the sixth. At least two antenna elements 182 &, grades of the N antenna elements are associated and have different polarities, thereby receiving at least two sums of different sums of M source signals, wherein m is greater than one. In antenna array 180, other antenna elements 18, such as 18 servants, may or may not be associated with antenna elements 182a, 182b. Even if another pair of polar antennas τa 184a, 184b are added, these antenna elements can have the same polarity as 22 1284465. In addition to these, these antenna elements may also be disconnected from each other. The different polarities of the antenna elements 182a, 182b may be orthogonal to one another. In another preferred embodiment, in addition to antenna elements 182a, 182b, antenna array 180 has a third antenna element 182c for supporting three polarities and receiving three different sums of VI source signals. The following further discussion discusses the application polarity to fill the mixing matrix | A. Three different polarity antenna elements 182a, 182b, 182c receive three linear and independent signal sums. The definitions of ^, 乂, and 2 axes shown in Figure 7 can be applied. For example, the following correlation exists: x = Scos (Θ) sin (φ) y = Ssin (Θ) sin (φ) z = Scos (φ) In order to simplify the mathematical operation, this preferred embodiment assumes: signal It has a linear polarity, the signal is linearly independent, and three linear antenna element components are present in each orthogonal axis. For example, antenna elements 182 & are present on the X-axis, antenna element 182b is present on the y-axis, and antenna element 182c is present on the z-axis. By locating the three linear antenna elements 182a, 182b, 182c to the respective orthogonal axes, the mathematical operating system can be significantly simplified. In actual deployment, antenna elements 182a, 182b, 182c do not need to be strictly orthogonal, and antenna elements 182a, 182b, 182c need not intersect at some common point. Removing this assumption does not invalidate the general conclusion. In contrast, removing this hypothesis will only change the situation where the rank is insufficient. 23 1284465 v ** ·· The following definitions can be applied, where the digital subscript indicates the connection of signal i, signal 2, and signal 3: S!, S2, S3: the signal incident on the antenna element; Θ3: χ γ plane electric field angle of the signal; Φι, Φ2, Φ3: the electric field angle of the Ζ axis of the signal; and Χχ, Xy, Χζ: the product of the sum of the signals incident on the antenna element. In this case, the vector component is expressed as: X y Z component "X" ·· 1 0 0 component "y": 0 1 0 component "Z": 0 0 1 Si coefficient: cos (θι) sin (φΟ sin (θι) sin (φι) cos (Φ0 s2 coefficient: cos (θ2) sin (()3⁄4) sin (Θ2) sin (Φ2) COS (Φ2) S3 coefficient: cos (θ3) sin (deduction) sin (θ3) Sin (Φ3) COS (for example) Calculate the point product of each antenna element and signal. (X.Ysx^+y^+Zi:^) determines the relative E-field component of each antenna element. These values can be applied. Generate a hybrid matrix:

Xx ^"cos (θι) sin (φι) cos (θ2) sin (Φ2) Xy sin (θι) sin (φι) sin (θ2) sin (Φ2) XzJ ^cos (φι) cos (Φ2) cos (θ3 ) sin (such as wide sin (θ3) sin (φ3) cos (such as) xy

Lv 24 , 1284465 where det Xy , XZ′′ =cos (θι) sin (φι) sin (Θ2) sin (eg) cos (Φ3) +cos (Θ2) sin (eg) sin (Θ3) sin (eg) cos (φι) +cos (Θ3) sin (eg) sin (Θ1) sin (φι) cos (eg) a cos (φι) sin (Θ2) sin (Φ2) cos (θ3) sin (eg) a cos (Φ2) Sin (θ3) sin (Φ3) cos (θι) sin (φι) —cos (deduction) sin (θι) sin (φΐ) cos (θ2) sin (Φ2) =cos (θι) sin (Θ2) sin (φι) Sin (eg) cos (eg) +cos (Θ2) sin (Θ3) cos (φι) sin (eg) sin (Φ3) +sin (Θ1) cos (Θ3) sin (φΐ) cos (eg) sin (eg) a sin (Θ2) cos (Θ3) cos (φι) sin (φ2) sin (eg) a cos (θι) sin (Θ3) sin (φι) cos (eg) sin (eg)-sin (θι) cos (Θ2 Sin

(φΐ) sin (Φ2) cos (cfe) = [cos (θι) sin (Θ2) sin (φι) sin (Φ2) cos (eg) a sin (Θ1) cos (Θ2) sin (φΐ) sin (Φ2) Cos (Φ3)]+[cos (Θ2) sin (Θ3) cos (φ))sin (eg) sin (eg) a sin (Θ2) cos (Θ3) cos (φι) sin (φ2) sin (eg) +[sin (θι) cos (03) sin (φΐ) cos (eg) sin (eg) a cos (θι) sin (03) sin (φι) cos (eg) sin (eg)] =sin (φι) sin (eg) cos (eg) [cos (Θ)) sin (θ2) — sin (θι) cos (θ2)] +cos (φι) sin (cfe) sin (φ3) [cos (θ2) sin (θ3) - Sin (θ2) cos (θ3)] +sin (φι) cos (eg) sin (()3⁄4)[sin (θι) cos (03) —cos (θι) sin (03)] =sin (φι) sin ( Φ2) cos (Φ3) sin (Θ1—Θ2) +cos (φι) sin (Φ2) sin (cfe) sin (θ2—θ3) +ώι (φι) cos (eg) sin (eg) sin (θ“θ3) Next, the case where the rank is insufficient will be discussed in further detail as follows. When the determinant is equal to zero, the rank of the mixed matrix is insufficient. The rank deficiency is caused by the following cases: 1) 01=02=03 '' X” Member and "y" are all identical elements provided for receiving part of the contribution of three signals. 25 Image 1284465 (2 c 9090 G g90°g90° such as 0) 0°0 0) 0° Increase by 180. For any combination of the table items, another example of a less-ranked rank is available. Examples of such a lack of rank are those in which the signal is not applied to a sufficient combination of antenna elements. (3) According to (1) and (2), all individual sums are not equal to zero, but: sin (φ!) sin (eg) cos (deduction) blood (θι—02) +cos (φι) sin (Φ2) sin (扣)(in) (s) (s) A signal on the opposite side, or a signal incident with a very low probability, which may result in the same energy level of the two components. The first section of the plan diagram as previously described proposes an antenna architecture. The previously described antenna architecture, including non-connected sensors, can be combined into a variety of different architectures to provide a sum signal of the source signals to the mixing matrix. Please refer to Fig. 8, which shows a communication device 200 for separating source signals provided by one source of signals. The antenna array 202 includes a plurality of antenna elements for receiving at least one different sum of the plurality of source signals, wherein the Ν and Μ are greater than one. 26 I284465 N antenna elements comprising at least one antenna element 2〇4, thereby receiving at least one sum of N different sums formed by M source signals, and N antenna elements comprising at least two connected antenna elements 2〇 6. At least two sums of n different sums formed by receiving one source nickname. There is no connection between the two connected antenna elements 206 and the antenna elements 204. The antenna array can have additional antenna elements in any combination, wherein each antenna element can be associated, non-existent, and polar. • Receiver 210 is coupled to antenna array 202 to receive at least one different sum formed by the source apostrophes. A blind signal separation (BSS) processor 212 is coupled to the receiver 21A to form a mixing matrix 214, wherein the mixing matrix column 214 includes at least N different sums formed by the μ source signals. The mixing matrix 214 has a rank equal to at least one of N, and the blind signal separation (BSS) processor 212 separates the desired source signal 216 from the blending matrix A. The second section of the plan map proposes the reinforcement of the antenna architecture provided by the first section. These enhancements can be implemented to further fill the hybrid matrix A by collecting additional or alternate sums of source signals. A reinforcement system includes an array deflection operation to receive an additional sum of source signals to provide a hybrid matrix A application without the need to add additional antenna elements. The array biasing operation includes an antenna pattern that controls the azimuth direction and/or the elevation (elevati〇n) direction. Referring to Figure 9, there is shown a communication device 24, wherein the communication device 24 applies an array bias operation, by which 27 1284465 separates the source signals provided by the source of the signals. The antenna array 242 includes a plurality of antenna elements 244' to generate ν initial antenna patterns and thereby receive at least one different sum of the one source signals. In addition, the antenna array 242 also includes an elevation controller 246 for selectively changing the height of at least one antenna pattern of the N starting antenna patterns, whereby at least one of the μ source signals is additionally different. The sum system can be received. The receiver 248 is coupled to the antenna array 242 to receive a different sum of the plurality of source signals using the ν starter antenna field patterns, and to receive at least one additional antenna pattern to receive at least one of the source signals. An extra different sum. A directory separation (BSS) processor 250 is coupled to the receiver 248 for 幵 > into a mixing matrix 252, wherein the mixing matrix column 252 includes jy [a different sum of the source signals and the μ source signals. At least one extra different sum. The rank of the mixing matrix 252 is equal to Ν plus the additional different sums of the source signals received by the additional antenna field type, and the 'Blind Signal Separation (BSS) processor 250 is self-mixing: matrix 252 The desired source signal 254 is separated. In general, any antenna array device that provides a summation of signals to increase the rank of the mixing matrix 252 can apply an array deflection mechanism. The array biasing mechanism can generate a sum of _ (four) and age scale W signals for each antenna_device. In view of this, applying this technique will produce a double multiplier effect. . If the array deflection is divided into the adjacent antennas of the associated antennas, each side of the unique area can provide two separate 28.1284465 vertical deflection areas and items to the mixing matrix. For example, if the antenna array can provide N total sums and there are κ unique deflection regions, the total number of signals of the hybrid matrix can be equal to 2 * K * N 〇 for convenience, please refer to the first In the figure, the switching beam antenna 100' shown in FIG. 4 is appropriately adjusted so that the antenna pattern can be tilted upward or downward. In particular, each of the upper half i〇4a' of the passive antenna element 104' is connected to the ground 106' via a reactive load 1〇8. The passive antenna element 104, each of the lower half 1〇41?, is also connected to the ground 1〇6 via a reactive load 108'. The passive antenna element 1〇4 has a reactance system that has the effect of lengthening or shortening the passive antenna element. The inductive load lengthens the electrical length of the passive antenna element 104, whereas the capacitive load shortens the electrical length of the driven antenna element 10#. The antenna beam system can be tilted upward or downward according to the ratio of the reactive load 1 (^ and the reactive load 1〇8 of the lower half 104b' in the upper half. For example, the reaction load 108' and the lower half of the lion have a reactive load of 1〇8, and the ratio 'antenna field type can be aligned upwards 97 or down 99, as shown in Figure U. § § Antenna field type elevation angle (elevati〇n angle) is adjusted to connect (4) the signal, at least - the extra rank (mnk) is added to the 匕 s matrix A application array bias (deflecti〇n) technology The hybrid matrix a can receive good signals without increasing the number of antenna elements. The specific real financial formula (4) has a _ unique bias (defleeti〇n) region and 'two unique biases (deflecti〇n) region, It is possible to apply the electric 29 • 1284465 anti-118' individual control. The field generation capability of the antenna array is three independent field types. The number of sums of the signals that generate the mixing matrix is equal to ten—(2*2*3=12 Please refer again to the previously mentioned US patent application Case η/〇65752, which further reveals how the height of the antenna beam should be adjusted. The array deflection technique can be applied to any of the preferred embodiments of the previously described antenna array, or any sensitive to ground interaction. Other antenna arrays. Another preferred embodiment of the elevation controller is based on a controllable radio frequency (RF) anti-flow 270 coupled to ground 272 of antenna element 274, as shown in Fig. 12. Antenna element 274 The connected antenna field type controls the radio frequency (RF) anti-flow 270 to move its height, as is known to those skilled in the art. Referring to Figure 13, there is shown a communication device 3, wherein The communication device 300 is based on path selection to separate the source signals provided by the M signal sources. The communication device 3 is another enhancement of the antenna architecture of the first segment provided by the plan, and previously described The enhancement of the array deflection technique. The communication device 3 includes an antenna array 3 〇 2 ' wherein the antenna array 302 includes n antenna elements to form at least N days a beam for receiving at least a different sum of the river source signals, and wherein the N&M system is greater than two. The controller 306 is coupled to the antenna array 3〇2 to selectively form at least N antenna beams. The components 3〇8 are connected to the array of wires 302 for receiving at least N different sums of the source signals. The blind 30 1284465 separation (BSS) processor 31 is connected to the receiver component 3〇8, thereby A hybrid matrix 312' is formed in which the mixing matrix 312 includes at least N different sums of the M source signals. The blind signal separation (BSS) processor 310 also determines whether the different source signals, the heart, and the heart are related or statistically independent. If not, the blind signal separation (BSS) processor 31 and the controller 3 〇6 cooperate to form different beams, so as to receive the different sums of the [V source signals], and then replace the difference between the source signals of the connected matrix 312 that are not related or statistically independent, ', heart and back' The source signal Μ# can be separated from the mixing matrix 312. The rake receiver is a wireless receiver that offsets the effects of multiple path fading. The raj^ receiver uses several independent delays with a slight delay to tune each multipath component to offset the effects of multiple path attenuation. Rake receivers can be used in most types of wireless access networks. Spreading type modulation is known to be particularly advantageous. The ability to select a particular incident signal path, the spreading type is particularly suitable as a means of changing the feed path for blind signal separation (BSS) processing. The selective formation of the previously described N antenna beam systems can be applied to a variety of wireless access network sounds as would be appreciated by those skilled in the art. For a code division multiple access (CDMA) system, the receiver component 308 includes n rake receivers 316. Each rake receiver 316 includes k correlators (fmger), thereby selecting k different sums for each of the n different sums of the source signals received by the respective connected antenna elements. 31 1284465 夕重路#Component . In this hybrid, the f-signal separation (BSS) processor 310 is coupled to a rake receiver 316 to form a hybrid matrix 312. The mixing matrix 312 includes at least k*N different multipath components of at least different sums of M source signals, and the mixing matrix 312 has a rank equal to up to k*N. In particular, when a code division multiple access (CDMA) waveform is transmitted, a code division multiple access (CDMA) waveform typically encounters a complex path between source and destination. In view of this, rake receivers 316 can be specifically designed to capture __路彳i and combine them for more robust signal decoding. Although the original signal is transmitted along an individual path, the characteristics of the original signal may also be adjusted based on the unique characteristics of the individual paths. In some cases, the adjustment of the correlation and/or statistical characteristics of the received signal will be significant enough to enable the received signal to be treated as a separate signal stream. In this case, an adjustment type rake receiver 316 can be applied to capture each of the adjusted signal streams and feed each of the adjusted signal streams to the mixing matrix 312 as a unique item. Although such a device for increasing the rank does not always exist, however, a highly multipath environment, because of its continuous need, usually has such a device for increasing the rank. Although a different path can be applied to the rake receiver 316, a more general method for any modulation technique is the beamforming technique, as shown in FIG. The beamforming technique is different from the rake receiver 316 because beamforming techniques are used to enhance the signal and reject the desired signal. However, the difference between the two is that, in fact, the rejection signal of the beamforming technique can form another version of the receiver wanted signal. However, 32 1284465 and the 'receiver member 308 need to detect a plurality of unique pass-through versions of the same signal, thereby establishing a third stage of the mixed matrix 312 〇 plan map with a sufficient rank number to signal splitting, thereby further Filling the Mixing Matrix A° In one approach, the sum signal can be split using spreading code. In another approach, the sum signal can be applied to the in-phase (I) module and the quadrature (Q) module segmentation. Refer to Figure 14, which shows how to apply the in-phase (I) module and the quadrature (Q) module for signal segmentation. The communication device 4 includes an antenna array 402, wherein the antenna array 4〇2 includes N antenna elements 〇4 to receive at least N different sums of the one source signals. A digital despreader 406 is coupled to ν antenna elements 404 to decode at least n different sums of the one source signals. Each of the n different sums includes k digits to provide a different sum of the related sources of the elbow. • Receiver component 408 is coupled to a digital despreader 406 to receive at least k*N different sums of the ]VI source signals. A blind signal separation (BSS) processor 410 is coupled to the receiver component 〇8 to form a mixing matrix 412, wherein the mixing matrix 412 includes at least k*N different sums of the source numbers. In addition to this, the matrix μ is mixed. The system has a rank number equal to up to k氺N (rank>. The blind signal separation (BSS) processor 410 can separate the desired source signal from the mixing matrix 412. According to the modulation type of the received signal, previously described The signal division can be applied to 33 1284465 to increase the rank of the hybrid matrix A without increasing the number of antenna elements N. IS - 95 code division multiple access (CDMA IS - 95 ), code division multiple access Two thousand (CDMA2000) and wideband code division multiple access (WCDMA) are examples of spread spectrum communication systems, in which spreading code can be applied. Common thread (comm〇n thread) indicates The unique digital system is applied to each signal processing to spread the data over a larger frequency band. φ The same spreading code is used to process the sum of received signals (desired signals, unwanted signals, blind sources). In this way, the wanted signal can be reconstructed and restored to the original frequency band, and the interference source can also be spread to the broadband frequency band. In the event of a shovel, the code division multiple access (CDMA) implementer is first shoveled. The system has a plurality of signal streams that simultaneously apply the same frequency band. In the case, each signal stream applies a set of numbers, and, theoretically, the set of numbers is orthogonal to all other numbers. If the decoder side satisfies this Conditions, which indicate that only important signals will be despread. If the total number of K signals is to be recorded in the county, the system will receive the majority and xk will increase the amplitude of the sk term and will remain unchanged. Or reduce the (k-丨) component of the value. The code division multiple access (CDMA) message usually has a certain degree of correlation. In view of this, more or less, the interference signal will also be merged with the desired signal. Rhyme establishment. In this case, the county has a delay of __ and multiple miscellaneous miscellaneous. Part of * want to shout (turn over, code division multiple access (CDMA) signal) may increase the value. Although this increase is not It will be as significant as the signal, but this increase will still increase the overall value of 34.1284465, and in view of this, the signal-to-noise ratio (SNR) is reduced. Despreading §fL equation form And the signal itself is a blind signal Separation (BSS) processing criteria. In fact, if a despreading code is applied to each of the known signals received by the communication device, the individual sum of the components of the Independent Component Analysis (ICA) model can be met. In view of this, the hybrid matrix system can provide the same number of available column items as the known digits. 'Assume that each column item is separately sufficient to generate a secret value. In the case of a sales situation, the mixed bribe may increase to more than the number. The number of values. For example, N antenna elements and M her code system can provide N*M matrix columns. For convenience of explanation, it is assumed that three numbers are known, and it is assumed that the three digital systems maintain their orthogonality. In the digital despreader lion: the hybrid matrix A is divided into the upper three columns and the lower three columns of the antenna stream, and is transported away from the diagonal after each antenna stream has been spread using three known digital despreading frequencies. The zero value is based on the orthogonality of individual numbers. Line Items Four, five, and six lines can be applied to the general example of blind signals of the same index.

Xl ai1 〇〇aH ai5 a% si r X2 0 a22 0 a24 a25 a26 s2 U3]=[ 〇〇细/34 a35 a36][S3] ^ 341 〇^ 344 a45 Θ46 S4 X5 〇a51 〇^54 a55 a56 S5 ^ 0 0 As the 365 366 S6 line item, the corresponding reduction of the fourth, fifth, and sixth lines can be other roads and versions of the known number of browns, or other cell signals of known numbers. In addition, one pupil 35 1284465 is a Gaussian signal, and the other signal can be a group of code division multiple access (CDMA) signals that adheres to the central limit theory, so that it can present a single The Guassian signal, for example, releases four channels. In other words, a sufficient number of non-random signals will be added to the Guassian signal. The source of the interference may be a source of non-Gassian signals, or at least one Gaussian signal on the Internet. After despreading the known number to _ using a digital despreader (Spreader) 406, the blind signal separation (BSS) processor 410 receives a mixed matrix 412 having a rank equal to six. The rank equal to six is multiplied by two times based on two antenna components because the three digital systems are known. The six signals are applied to a blind signal separation (BSS) processor 41A, wherein a mixing matrix 412 having a rank equal to six can be formed. The blind signal separation (BSS) processor 410 determines the separation matrix w only by the received signal of the application channel adjustment, that is, X==AS. In the example shown in Figure 14, the six signals are separable. The blind signal separation (BSS) processor 410 selects the signal to be decoded. For example, the interference source signal can be discarded, and all versions of the desired signal can be selected. The selection signal is applied to the demodulation module to customize the demodulation operation. The demodulation module applies a known equalization technique, and it is known that the equalization technique combines multiple path versions of the same signal. In the general case, the value away from the diagonal (for convenience of explanation, the example is represented by zero) may also be a non-zero value. In the case where the digital signal is not 70 US dollars, the non-zero value away from the diagonal may be more common. The non-zero value of the diagonal line indicates the additional impurity of each separate signal 36 1284465 吼. However, as previously stated, the rank of the mixing matrix is sufficient to separate these signals, so that after performing a blind signal separation (BSS) operation, the non-zero value away from the diagonal will be significantly reduced. As a result, noise reduction, increased signal-to-noise ratio (SNR), and increased channel capability (as shown by Shannon’s Theorem) are available. Next, please refer to FIG. 15 , which shows another means for increasing the rank of the mixing matrix A without increasing the number of antenna elements N, wherein the means separates the received mixed signals into in-phase (j). Ingredients and the parent (Q) component. The in-phase (〗) component and the positive-father (Q) component of the associated radio frequency (RF) signal have the same amplitude but a phase separation of 9 degrees. The pass-through device 500 includes an antenna array 502, wherein the antenna array 502 includes N antenna elements 504 for receiving at least one different sum of the μ source signals. Individual in-phase 〇) and orthogonal ((^) module 5〇6 series are connected, and each antenna element 504 is streamed downstream, thereby separating the sum of the different sums of the one receiving source signals into the same phase (1) and The set of orthogonal components. The receiver component 508 is coupled to the respective in-phase (1) and quadrature (Q) modes, and 506 is downlinked to receive at least one of the n different sums of the two source signals. (1) and orthogonal (9) component sets. The blind signal sub-segment (BSS) processor 51 is connected to the receiver component 5〇8 under the line string capture to form a hybrid matrix 512, wherein the hybrid matrix 512 includes (4) solid source information. At least 2 different sums. The individual in-phase (1) and positive-father (Q) component sets provide two rounds to the mixing matrix 512. The mixing matrix 512 has a rank equal to one of up to 2 degrees, and 1284465 The blind signal separation (BSS) processor 51G can separate the desired source signal 514 from the mixing matrix 512. Figure 16 shows the individual in-phase (phantom and quadrature (Q) modules of the downlink connection of the antenna elements 5〇2. 506. Mixing of antenna elements 502 received The system uses a pair of hybrid II 52G divisions. The _(1) and quadrature (Q) components apply two simultaneous measurements H (where two identical reference signals with a phase difference of 9Q are applied to the two lang step! m (4) The radio frequency (IF) signal is translated to another frequency range and generated together. The in-phase (1) and quadrature (Q) are the same as the phase information of the marginal intermediate frequency (IF) nickname, so that the signal with positive frequency can The dog is distinguished from the signal with a negative frequency. b. By dividing the received mixed signal into in-phase (1) and orthogonal components, the size of the mixing matrix can be doubled. As long as the in-phase (Z) and orthogonal (q) components are With different data stream coding, the mixed signal received by any antenna element can be divided into two different mixed signals. In the case of differential coding, the modulation characteristics need to be analyzed to determine the in-phase (I) and the positive (Q) Whether the component satisfies the linear requirement. For example, for global machine communication (10) (deletion), it is known that the Gaussian minimum shift key GM (GMSK) can be assumed to be a line when appropriate. Raw, and 'Muse minimum displacement The control (GMSK) coding system can operate as a binary-shifted phase-keyed (BPSK) code. Since the binary-input (BPSK) (BSS) device's previously described in-phase (1) and quadrature (Q) processing systems can Application: The in-phase (1) and quadrature (Q) processing systems can be applied to the previously described antenna array, which is better than the full-age matrix A. When applying in-phase (1) 38 1284465 and orthogonal (Q) processing' The hybrid matrix A can be filled as if the number of double antenna elements is used. Another example is to apply two antenna elements (double), which are not related and have different polarities (2 * 2 = 4 times). Eight independent mixed signals are generated by combining the contract phase (I) and the quadrature (Q) component (2*2*2=8 times). This mechanism can also be used in conjunction with antenna array biasing to generate more signals. The sums can also be separated into in-phase (work) and quadrature (Q) components in sequence. Another feature of the invention relates to multiple input multiple output (MIMO) antenna technology by which the same radio frequency (Rp) channel is repeatedly applied. The receiver processing technique for removing interference sources can minimize the number of antennas required for field diversity, rather than applying antenna diversity, thereby achieving signal robustness and corresponding data rate increases. The receiving path of the antenna array has a changeable weight. When the weight changes 牯' the receiving antenna field type can be adjusted. By applying a similar technique to Blind Signal Separation (BSS) technology, the wanted signal can be extracted from the receiver data containing several sources of interference. —, how to form a Na-type ,, in the multi-input multiple-output (MlM〇) only: mode receiving structure, the application of field diversity instead of antenna diversity is 疋 仃 仃, as shown in Figure 17. In theory, the number of κ field patterns will be equal to the number of antenna elements. However, the κ field patterns can also be generated using L antenna elements of less than one antenna element required by the known technique. The application of the existing antenna array multiple input multiple output (μιμ〇) implementation of the similar hand I and 'IV [and κ will only be equal to each other in the case where all transmission spatial channels can be applied with 39 J284465 κ receiver field identification. Since this situation usually only occurs in the case of fixed transmitters and receivers, an additional receiver field or transmitter antenna will have its need to achieve a minimum of Κ or Μ spatial gain. In addition, multiple user detection (MUD) processing techniques can be applied to separate the data channels of the receiver system. All of the previously described methods of establishing an integrated matrix can be applied to portions of such an implementation.

Another feature of the invention relates to loose inter-symbol interference (ISI). Please refer to Figure 18, which provides an architecture that can be applied to reduce the mutual symbol interference (ISI) by applying a Fourier transform method. The following blocks are added to the transmission side to improve the Fourier transform method for reducing mutual symbol interference (ISI), ie: Viterbi coding, repetition or puncture, and block redundancy interleaving are added to Transfer the sides. In addition, the following blocks are added to the receiving side, ie: blind signal separation (BSS) interference source removal, block deinterlacing, de-repeating/de-puncturing, and Viterbi decoding are added to receive. Side. "Viterbi, coding, has the advantage of overcoming the redundancy of the data decoding process. In addition, alternative coding forms, such as: turbo (=.) coding ' can also be applied.,, repeat or puncture "The system can reach the source of the material rate and the rate of the material to match the money. ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, Viterbi solves .1284465 code 5^^) and, compared to the block error, the Viterbi solution efficiently returns the data stream from the random distribution error in the transition = 2: i time domain The previous 'Bss' interference source removal system can reduce the signal to the desired signal.

The signal system has a known statistical characteristic (not very rhyme), and the non-distributive distribution (the peak-to-average ratio (PAR) bit, the set-wise mode will be on the output side of the fast Fourier transform (FFT) $ (4) Linear mapper (mapper) (by equalizing the signal reverse conversion of each frequency and touching the _ (IFFT) reader side increase. In addition, in actual situations, this signal usually modulates and adjusts the frequency. To the transmission frequency range, @ increase the viewer, the up converter and the lower converter, and the demodulation variable H system can complete the picture. The transmission of the boundary between the skins can be discontinuous. The system can be mitigated in several ways. One way is to increase the guard band between waveforms (interpolated between waveforms) to minimize the frequency component. Establish all the previously described methodologies for the hybrid matrix. It can be applied to parts of this embodiment. Another feature of the invention relates to field diversity to support hierarchical spatial communication. Referring to Figure 19, in a preferred embodiment, the transmitter is timed. Basically changing the power level of each hierarchical space stream. In view of this, 'each hierarchical space stream system can use different power levels to reach the receiver' to provide appropriate differences in received signals to fill the appropriate blind signals. Separate (BSS) processed matrix. Since all power adjustments are done on the side of the transmission 41, 1284465, the number of L antenna elements on the receiver side is equal to one, and the receiver side does not need the field type to generate hard Body or software component. This approach also overcomes the problems of the prior art because it is no longer a problem to have a different angular profile between the signals such that there is a small angular difference between the arriving signals. In another preferred embodiment, In addition to the source of interference from the transmitter, other significant sources of interference may also exist. If there is only a single source of interference, the source of the interference and the difference between the wanted and the wavefront of the wanted transmitter will be suitable for blind signal separation (BSS). Processing to separate all signals. If there is more than one single source of interference, the rank of the matrix may not be appropriate. Additional field type changes are generated on the side of the device, and the system performance can be improved. Although this method is only the deviation of the preferred embodiment, the means can still be significantly better than the previously described embodiments. There is a small field type, and therefore, this method is also less relevant to the implementation of the receiver side. • In another preferred embodiment, the sum of the plurality of data streams is transmitted via a single antenna 70 piece power amplifier. Based on the time slot, the relative power level of the sum tiger can be changed, which is suitable for decoding on the side of the receiver. The advantage of this method is that the individual signal streams of the mixed signal can undergo individual transmission paths. , which means that the relative signal relationship between the transmitter and the receiving state can be maintained, whereby the non-durable decoding system on the side of the receiver can be obtained. The concept can be scaled, whereby a plurality of individual sum signals can be transmitted by different antenna elements. In view of this, the separation of the secret number can be obtained in 42 J284465 with multiple loop diversity gain and/or space capacity gain. To solve the problem of fixed peak-to-average signal power ratio, the power of the sum signal can be adjusted appropriately to maintain a near-fixed power level. All of the previously described methods of establishing this S-matrix can be applied to portions of this implementation. Another feature of this description relates to the volatility field to support complex sync transmitters. Referring to Figure 20, the multiple devices transmitted to the wireless network base station (AP) are tuned to their radio frequency (RF) field types. In view of this, it is desirable to have a wireless base station (AP) and an undesired wireless network base station (Ap) that can receive different power level versions of the transmission signal. In this way, Blind Signal Separation (BSS) technology will get enough information to separate the signals. This modulation can be as simple as changing the transmission power. This modulation can be done independently of the field profile, so omnidirectional, segmented, or wire-formed fields can be applied. In addition to this, other techniques such as changing the transmission beam aligner can also be applied. The most effective means is to align the transmitter application with the time slot. The timing can be set by the internal clock settings of the application or synchronized with the common time stamp transmitted by the wireless network base station (Ap). If the time at which the signal arrives at the receiver is misaligned, the ability of the blind signal separation (BSS) to separate the signals is mixed. Subsequent 'time-leading or delay technology' can provide storage applications. ', assuming that the signal reception gain change is applied to the blind signal separation (BSS) that is considered as the target, and the interference source to be equipped with the wireless network base station (Ap) and other examples = 43 1284465, the appropriate receiver system to be aligned Can be changed. If there is no overall network coordination, the receiver system should be aligned. If there is overall network coordination, the measurement system can show that the best means is to enable the signal to be easily removed as the source of the interference, but still provide proper alignment on the side of the receiver for separation. Traditional signal rejection techniques can be applied if there are other sources of signals that do not use radio frequency (RF) power level modulation techniques. Alternatively, the receiver can apply % or other means to increase the rank of the appropriate matrix of blind signal separation (Bss). Even with the latter approach, deriving the matrix information level will significantly reduce the burden on the side of the wireless network base station (Ap) receiver. All of the previously described methods of establishing a hybrid matrix can be applied to portions of such an embodiment. Another feature of the invention relates to adjusting blind signal separation (BSS) radio frequency (RF) decoding for optimal processing and power drain. The number of signals that need to be separated to decode the important stream can be reduced. In general, the rank of the decoding matrix determines the number of most important signals that will be separated, while other signals are treated as noise. In view of this, this value needs to contain the minimum value of the signal to be decoded. In addition, reducing the noise component may require a higher minimum, so that the signal to noise ratio (SNR) can achieve an acceptable decoding error rate. Figure 21 shows an embodiment in which only receiver operation is performed. Figure 22 is a superset of Figure 21, which also includes the transmitter-to-receiver data and, optionally, the receiver-to-transmitter data. If the selection of the filled matrix exceeds the rank required for the operation, the antenna selects 44, 1284465, and the control can reduce the number of application choices. The combination of the choices is better than the other ones, and the best choice is the lowest. This collection can be determined by comparing the results of various selections and other selections of signal checks. The error technique _, the application of the whistle and the choice of ==, or, can be applied to the history of the application of conditions and results. Which type of f or prescription is used can be based on given conditions and the effectiveness of the test. *When the known device is located in the important signal range of several sources, such as when it appears in the overlapping area, the maximum power signal can be pre-exposed to the direction of the *. In view of this, these choices can be determined, and thus provide these The important signal difference of direction. For coding, the error correction coding determines that the error rate of the original decoded stream can be tolerated. Since the original error rate is also a function of the matrix filling the selected subset, there is a trade-off between these settings. The feedback and control loop between the programmer and the decoder # can be applied to select the optimal mutual setting. If the receiver is known not to be in the power limit (for example, the line power supply), the decoder will increase. Its matrix rank (rank). This method can be applied to several purposes. A higher rank (rank) meter can reduce the noise, which can increase the signal-to-noise ratio (SNR) and thus reduce the error rate. The noise system can be applied to the increase of the transmission data rate, the error correction coding ^ reduction, or the improvement of the overall connection reliability. In addition, the matrix is filled up. Transferring the load to the receiver can also reduce the load on the side of the transmitter, which can be used to control the loop between the two. There is a lack of control. 45 1284465: The device using the battery will try to achieve the rank (magnetic) production wide (three) port, thereby It becomes a more durable power supply device. The most economical operation system requires the decoding matrix to be 2,000 kilograms every other symbol. However, in general, the connection time is more than the rate connection. The measurement only needs to be slightly slightly faster than the connection. The processor load of time can be reduced. Decreasing the decoding matrix determines the appearance of the system can save power and the calculation ^ see ^ car ^ change the appearance can determine the decoding matrix needs to be re-interval. Each sub-frequency t broadband system, system, sub-channel usually It has an individual correlation time rate. By means of its own solution matrix and related measure speed =: two 'sub-solutions and one test sum will be: tune, the chest system can be used to provide the appropriate rank number (called News, receiving string; 魬民's standing on the relevant information of the transmission characteristics. The right blood knows the measurement of greed, or negotiates with the source to set the base ==::::: Source of resources Restricted "One side can be shouldered to bear a higher burden, and in another case, the decoding matrix will not be determined by the solution: the type of intrusion: when starting to solve large matrices, repeat 46 1284465 blind state begins to solve Situation. This method is a conventional solution to fill a matrix with a large rank. In general, the combination of all the methods described above is based on the available components, the modified digital level, the appropriate equipment, and the (4) It is possible that all of the previously described methods of establishing a hybrid matrix can be applied to portions of such an embodiment.

(rank), quite similar to the invention, is related to the volatility field type to support effective area coverage. For field-type transmissions, the basic idea is to apply segmentation coverage patterns on the side of the infrastructure. The actual number of segments can vary with capacity needs and associated cost factors. The actual implementation may be (four) - the number of segments varies to a random number of segments. The segment itself can be distinguished as an azimuth or elevation or a herm and elevation plane. The main advantage of the Leo section is that the section tracking can be buffered relative to the beamforming method section. In view of this, retaining the coverage area of one section to another section can be reduced to the traditional handoff situation. Conventional techniques enable the receiver to generate field type changes to provide blind signal separation (BSS) signal separation processing applications. In contrast, the transmitter is an application technique whereby a suitable blind signal separation (BSS) decoder environment is present at least four. In the partial implementation mode, this means that the receiver does not need to generate any wave pattern. In other embodiments, this means that the number of fluctuating field types can be significantly reduced. A preferred embodiment is applied to the early transmission point. This preferred embodiment solves the problem of the following situation 'When it is not known whether other transmission sources 47.1284465 in the area are also operating. Please refer to Figure 23 for the wheel-type wheel _ fluctuations at the known timing of the receiver. The change in transmission pattern is the timing by which the segmentation of the transmitted symbols is aligned. The field profile changes moderately with respect to the bore sight movement and remains fixed at each time slot. In view of this, the coverage area will not change significantly, and there is no such problem as before. Based on the changed transmission profile, the receiver will experience a change in the wavefront power level. In view of this, the Blind Signal Separation (BSS) matrix will be filled with various signal stream differences for different relative gain values. If the main signal received is from a single or multiple transmitters that use the heartbeat signal, the receiver only needs to sample during each field change and apply the data to fill the mix_ to provide blind signal separation (Bss). Separation. If there is a hybrid transmitter with and without wave signals, the receiver can apply traditional signal separation techniques to compensate. For example, methods such as • Beamforming and Multiple User Detection (MUD) can be applied. However, the blind signal separation (BSS) method is generally more durable. In practice, the receiver can perform field-type deformation and generate enough additional field patterns to increase the rank of the blind signal separation (BSS) matrix to more than the number of signals to be separated. For example, for a blind signal separation (BSS) decoder, if two types of sigma-holes are used to transmit the transmitter, and there are two types of pedestal reception numbers, the receiving system needs to generate at least two types. Contours to separate sources of interference and other signals. This method is more necessary to reduce the three contours. If the transmission of the 48.1284465 transmitter does not produce its own set, the means for this receiver is always low. /Right transmissions carry a single stream along a single path, and the field-type gallery collection does not need to be rotated or different. This is because the receiver detection signal has changed relative to all other received signals. In view of this, the transmitter can be applied simply: the power is changed to the overall field type without changing the contour shape. If only a single other stream is summed to the receiver, the blind signal separation (Bss) process will be able to separate even if the amplitude of a single stream remains fixed. This is due to the fact that the power dithering source provides a change in operational requirements. If more than one other stream is received, these streams can be considered as a single source of interference for blind signal separation (BSS) processing, unless the receiver itself applies other separation devices, or unless the receiver itself has a wave field Type generation ability. The field mode transmitter of the receive mode will be discussed in further detail below. Since the blind signal separation (BSS) processing of the unequal field profile is an excellent method for signal separation, the same technique applied to separate the transmission field types can also be applied to generate multiple receiver values. In view of this, the only cost factor for blind signal separation (BSS) reception is the processing burden of blind signal separation (BSS) processing when transmission is already supported. Feedback from the User Equipment (UE) Receiver to Transmitter will be discussed in further detail below. Although not strictly required, the feedback information of the receiver (UE) receiver can be applied to improve the overall connection operation. For example, the receiver can determine the extent to which each field profile change provides useful information. This information is fed back to the transmitter. Subsequently, the transmitter system can

4Q • 1284465 adjustment operation to change the fish less interference 4 2; #= less power, her and symbol transmission wheel ^^=; ^ bracket · · the application sequence of each field type, the wheel performance of each symbol. Will be passed to the receiver for the most

The foregoing description of Γ2/Γ has been concerned with most transmission points, and it is known that the implementation of the side of the real heavy transmitter side of the application side has substantially the same receiver operation as the single transmitter mode. The difference between the two is that the field type generated by each transmitter can be counted in the receiving test to perform signal separation for blind age separation (BSS) processing.曰 = and durable operation can also be obtained from the network receiving the negative correlation of the parameters of the coordination parameters. For example, the car rank (&), which then dominates the need, the number of models can be adjusted. In view of this, the type of receiver is produced: when it is available, it can be measured according to this (4). The Network Broad Wireless Resource Management System can apply information fed back to User Equipment (UE) to establish network wide field applications, trends, power levels, and timing. All of the previously described methods of establishing a hybrid matrix can be applied to portions of such an embodiment. Another type of signature of the present invention relates to blind signal separation (BSS) and field fluctuations to assist in code division multiple access (CDMA) signal separation. Blind Signals _ (BSS) Derivatives are intended to effectively separate signals. ※ The receiving signal must be a collection of received signals from antennas whose individual signals are related to different weighting factors. This means can be done at the transmitter, receiver, or both. Regardless of whether the weighting factor is changed to the transmission side or the receiving side, the weight of the .1284465 value factor can be changed according to the segment or the set of consecutive segments. The basic thing is that the set signal is adjusted during the period of each symbol to adjust the number of signals to be separated ^ times. Fig. 24 shows a case in which the symbol is changed ten times in frequency (twelve segments). The variable parameters are fixed to four slices. Three variations of each symbol indicate that three different signals can be separated from the received signal. # If the transmitter transmits a single stream along a single-path, the field wheel set does not require rotation or difference. This is because the receiver detection signal changes relative to all other received signals. In this case, the transmitter can (10) use a simple, field-type power change without the need to change the shape of the porch. ^ Only a single other stream system sums up on the receiver side, and the blind? Tiger Separation (Cong) process will be able to separate it even if the amplitude of a single stream remains fixed. This is because the source of the power turbulence (the ering) source provides the operational needs of the modified m other types of streaming systems. These streams can be treated as a single-group interference source, unless received. The device itself applies other separate skirts, or unless the receiver itself has a fluctuating field type generating capability. Although rarely required, the User Equipment (UE) receiver's feedback smuggling system can determine the extent to which each field profile change provides useful information. This Baibei (10) is fed back to the transmission ||. Subsequently, transmission (4) can be adjusted to operate, borrowing (4) good links, less power, or less interference with other connections. Although there are many ways to change the power shape, some adjustments can be 51 1284465 including: the order of each field type, the symbol and the modulation or dither power to the individual link = the number of changes, the rim change adjustment will Pass to the yoke. The actual power amplification of each symbol «small enough material mm ° large value average material, the secret operation of the operation ^ ^ ° ° utilization resulting in a reduction in the linear dynamic control range of the power amplifier == this and the operating distance between the device Reduced. When the power system applies the transmission parameter ¥, this consideration can be mitigated in several ways. These means can include: · When more than a single source (s to provide power to the same amplifier, the blind signal separation (BSS) can be changed by two: so that the sum of the powers of all signals can be fixed. In other words, The increase of the sub-transmission is the reduction of other transmissions. (4) The offset is 2. The power is modulated to a value close to the fragment rate. The extra power is usually used to store the component, and the storage component can only be generated. A small amount of fluctuations. The extra power can be transferred to the heat dissipation load. The 2D or 3D field type is generated by several means to provide the transmitting antenna and the receiving antenna application. These methods include: · Adjusting the delay and power of the phase array antenna Level, adjust parasitic antenna elements that can switch loads, adjust polarity changes, adjust power plane load changes that cause field deflection, adjust mechanical movement of components or reflectors, and then combine the previously described adjustments. All of the previously described methods of the matrix can be applied to portions of such an embodiment. There is a single receiver for multiple spatial independent channels. The switching parasitic antenna can be used for high-speed digitizers and down conversion 52 • 1284465 Early low noise amplifier (lna), mixer, ^ m skin (LPF), And the analog digital conversion state (ADC) is provided. The use of this technology to get the township weight (four) alone (four) secret can be applied to various evenings: processing, for example, related combinations, blind signal separation (win), or evening input multiple output (MIM0) Receive processing.

Figure 25 is a further detailed description of various system principles. This preferred implementation includes a single-antenna _, wherein the antenna array has switching elements that are switched to switch to inductance H or capacitance ϋ. The bandpass filter and the waver system can simultaneously limit the frequency band and overall radio frequency (RF) power exhibited by the low-frequency magnetic amplifier (LNA). The hybrid, amplifier (LNA) is not only a low noise amplifier that receives signals. The w and the regional oscillator (L〇) adjust the RF signal to the intermediate frequency (IF) or the fundamental DC (Dc). Any implementation is compatible with backend processing. The antenna switching, the selective regional oscillator (L〇) switching and the demultiplexer switching are all the phase-separated sequence generator generator axes, so that the N signal channels can be generated by the N rural antennas. Thereby, the single-channel radio frequency (RF) output of the mixer can be generated, which is presented in a low-pass filter (LPF) and an analog-to-digital converter (ADC) analog-to-digital converter (ADC), although not shown in detail in the drawings. It is synchronized with the same antenna sequence generator that drives the antenna mode, the selective regional oscillator (LO), and the multiplexed work. Considering the carrier frequency Fc and the signal with the 4 frequency X* B, the demultiplexer can implement the 1284465 sampling operation under the clock shape pulse. For arrays with N components, the sampling frequency of an analog digital converter (ADC) requires at least 2*N*B. The n-series is required because the N samples are only age-independent - a demodulator link is presented to the fresh processor. It takes 2 * B to satisfy the Nykster (Caf) sampling theory. In view of this, the signal bandwidth of this pure reception is also limited by the device switching rate. The demultiplexer is alternately sampled to the baseband processor (10) p) (iv) each of the demodulation transformer circuits of the N parallel demodulation transformer circuits. The sampling distribution method must not be a group distribution form, but a sequential distribution form. For example, if there are two antenna diversity options (left, right, omnidirectional), then ^^^. Samples i, 2, 3, 4, 5, 6, 7, 8, 91(), 11, 12 of analog-to-digital converters (ADCs) will likely have the following distributions, ie: 丨, 4, 7, and transmitted to The first demodulation transformer is connected; 2, 5, 8'11 are transmitted to the second demodulation transformer link; and 3, 6, 9, 12 are transmitted to the third demodulation transformer link. As mentioned previously, the demodulator may be a combination of multiple input multiple output (MIM〇) demodulation of the associated combination, blind signal separation (BSS) or two common multiple input multiple output (mim〇) demodulation techniques. The form of technology. Such an implementation may be N examples of a single demodulation circuit or a single package of N spatially independent channels. The associated portfolio can be a weighted or hard decision of the soft decision. Some embodiments are discussed in further detail below. These limitations include • Signal Noise & (SNR) considerations, noise data, impedance matching, and received signal power. If the false a and the antenna array have a bandwidth matching the received signal, the inner band number noise ratio (SNR) remains the same. However, the internal band signal energy of 54 1284465 is reduced to (1/N2) times compared to the conventional array. Since the low noise amplifier (LNA) is the signal path of the antenna array in the future, the noise data is not as significant as when the switching array starts with the photo detector (PIN) diode. Since each channel after the multiplexer receives the signal power (丨/N) times, the low noise amplifier (LNA) pitching requirement is increased (10*l〇gi()N) times, so that it is output at the mixer. Maintain a fairly signal amplitude. Switching to different antenna elements will result in a change in impedance matching characteristics. This is not the case for the following antenna implementations, which are always applied, active, and the antenna elements are antenna elements that are directly connected to the radio frequency (RJ7) path. Other, parasitic antenna elements have only an impact on the radio frequency (RF) path. Another preferred embodiment that can be compatible with partial multiple input multiple outputs (mim〇) and other parallel path transmission means integrate the following actions, namely: adjusting the regional oscillator (L0) to different carrier frequencies, and switching to Different diversity modes of the antenna array. Such an embodiment can be done synchronously, or such an implementation can be done independently of each other. In time, they still have to appear synchronously, but the individual states (array mode versus carrier frequency) do not need to have the same phase. Such an implementation would be a useful implementation for receiving IEEE 802 Jlg waveforms in which two rules IEEE 802.11g waveforms are transmitted in parallel = the same carrier. In this example, the upper and lower carrier frequencies can be substituted for the regional oscillator (L0), and then the upper and lower carrier frequencies can be substituted for different diversity modes of the antenna array in different field types. The mixer can be set appropriately to down convert the RF (Rp) wave to 3 to 55, 1284465 intermediate frequency (IF) or fundamental frequency DC (DC). This implementation can change the partial sampling requirements of an analog digital converter (ADC). Attentional or other techniques can perform intermediate frequency (top) downsampling: and still be able to respond to the desired information content. This approach also considers the repeated application of antennas for receiving and transmitting functions. For some applications such as satellite reception, the transmission function is not required. However, for time-division duplex (TDD) systems (such as wireless local area network _ (WLAN), wireless metropolitan area network (WiMAX), broadband code division multiple

Access time division duplex (WCDMA-TDD), time division synchronous code division multiple access (TD-SCDMA), etc., or, for time slot frequency division duplex (fdd) systems (such as · Global Action The communication system is a whole packet wireless service (GSM/GPRS), etc., in which reception and transmission are not synchronized, and when the transmission mode can be regarded as independent, the receiving antenna can also be multiplexed. For full-duplex frequency division duplex (FDD) systems (such as: code division multiple access two thousand (CDMA2〇〇〇) or wideband code division multiple access frequency division duplex (wcDMA. — FDP)) The wheel function can be done with a separate antenna. Any enabling interface technology can be applied to any of the enabling demodulation techniques (Connected Combination, Blind Separation (BSS), Multiple Input Multiple Rounds (ΜΙΜΟ)). Another feature of Ben Maoming is blind signal separation (BSS) for code division multiple access (CDMA) 2 receiver processing applications. An antenna array having a field spacing between antenna elements is suitable for feedback decoding connections. Available literature surveys show that, in general, those who are familiar with this technology are convinced. ~ Other documents discuss the so-called Single Antenna Interference Cancellation (SAIC) technology. Persons applying blind signal separation (BSS) require that the modulation have a correlation of 56, 1284465 in-phase (1) and orthogonal (9) channels, so as to produce a matrix of eight. In view of this, these decoders can both be early and interfere with the source and desire. A source of interference, the existing pre-existing SAIC technology is called application, virtual, and second antenna. η丁(10) Conventional improvement of shooting at the domain device to obtain independent signals

, and 'and, the unread financial law. Although Langxiang (1) = parent (Qj device is feasible for some wireless access networks, however, the in-phase and right-of-father (Q) devices are not suitable for coded multiple access (cdma) =. The silk system can be applied to this part of the implementation. Although this kind of lining increases the rank of the available matrix of the ICA, and makes the ICA analysis Paying is more likely, but this part is not guaranteed. Therefore, as previously stated

Technology still needs to be properly applied, and the code is broken. For example, it is still possible to return from independent component analysis (ICA) processing if it is extremely detrimental to the sum of the processed signals. In the second preferred embodiment, the 26th_ indicates the application of a different decoding link. At node A, an example of signal aggregation is shown in Figure 27. For convenience of explanation, Figure 27 shows the source of single-interference, and it is also possible to cope with multiple sources of interference and increase the rank of the matrix. The noise reference is made by a narrow-band dry (four) source that exceeds 'and' wants a code division multiple access (CDMA) signal to be below the noise reference. At node B of Figure 28, the source of the interference is taken out. "Selector" 57 ♦ 1284465 determines whether the signal is really the source of interference. If there is no interference source, no signal will be selected. If the signal has the characteristics of the desired signal, the $ signal will not be selected. If single or multiple sources of interference are selected, these sources of interference are presented in the "inverter" (node c). The Independent Component Analysis (ICA) acquisition system can receive signals in the reverse or non-reverse direction, and it is necessary to determine whether each signal needs to be reversed to match the received signal. The source of interference with the correct amplitude sign is presented as the negative input of the addition of node D. Those skilled in the art should understand that other alternative or equivalent embodiments are also possible. For example, a simple adder can be applied to this circuit stage, and the inverter will only be applied when the capture signal has a non-inverted waveform. The delayed version of the original received signal (Node A) is presented at another adder input. The delay value is equal to the delay caused by network connection sharing (ICS), selection, and "reverse" processing. Those skilled in the art should understand that other alternative or equivalent embodiments are also possible. For example, the 'Delay and Addition function block can apply the minimum block substitution, which translates and sums up the two signals until the minimum value is obtained. The node interference source in Figure 29 can be removed. At point E in Figure 3, the rake receiver can despread the signal and present it at the baseband decoding $. The further details of this preferred embodiment are: Antennas, and the collection of the tigers can be obtained through the selection of the previously described preferred embodiments to enhance the existing technology. The structure of the $26 ride is only one of the embodiments of the present invention. In addition to having the selector, 'there is no signal at the appropriate time, the conventional implementation of the & processing or post-processing location to select a different path can also be applied to 1284465. The trade-offs are: processing delays, implementation costs, overall operational durability, and some degree of designer choice. The basic idea of subtracting the source of interference from the signal stream only before presenting to the rake receiver needs to remain in all variations of the invention. While the previous explanations are ideal for removing sources of interference, it should be understood that not all sources of interference can be removed. However, in general, any removal of interference sources can provide better performance than conventional techniques, assuming that the rake receiver will handle improved signal aggregation. The characteristics of Code Multiple Access (CDMA) signals have a more prominent Guassian distribution than their despreading versions, and make Independent Component Analysis (ICA) more difficult to detect. However, it is also possible to remove some of the information related to the nickname, as the signal will retain some statistical significance. Again, it is often more important to remove the source of the interference, and the overall gain presented by the rake decoder can also be improved. Alternatively, the overall decoding process can be further enhanced by applying a means of scaling. This means that the addition or exclusion of the signal can be further checked and/or removed. The number of signals can be incremented or incremented, and the decoded signal integrity can be measured to obtain an improvement/deterioration of the result. The gist of this preferred embodiment is that • Independent Component Analysis (ICA) is applied to signals that may be identified and not applied to code division multiple access (CDMA) signals prior to rake processing. Multi-access (CDMA) signals are difficult to identify and/or capture. Another feature of the present invention relates to a hybrid minimum mean squared error matrix pen beam separation weight separated by a field type blind signal. Please refer to U.S. Patent No. 59,128,465, the patent number US6931362, in which a plurality of segments are required to provide a linear independent sum signal. The content of U.S. Patent No. 6,931,362 is hereby incorporated by reference. The previously described antenna array can be used in place of a plurality of sections, however, the processing disclosed in U.S. Patent No. 6,931,362 is still applicable. Although the present invention has been described in detail with reference to the preferred embodiments of the present invention, various modifications and variations of the present invention are possible. In view of the above, it should be understood that the present invention is not limited to the specific preferred embodiments described above, and that various modifications and changes of the present preferred embodiment are also made without departing from the scope of the present invention. It should be included in the scope of the following patent application. • 1284465 [Simple Description of the Drawings] Figure 1 is a block diagram showing a typical operating scenario in accordance with the present invention, wherein a communication device receives desired and unwanted signals via individual signal sources; 1 is a detailed block diagram of a communication device shown in FIG. 1; FIG. 3 is a plan view showing different means according to the present invention, thereby generating a linear independent sum of source signals of the mixing matrix; ❿ FIG. 4 is a view showing an antenna according to the present invention. a block diagram of an array, wherein the antenna array architecture is a switched beam antenna; FIG. 5 is a block diagram of an antenna array according to the present invention, wherein the antenna array architecture is a phase array; and FIG. 6 is a diagram showing the antenna array according to the present invention. a block diagram of an antenna array, wherein the antenna array is structured with a polar antenna element; FIG. 7 is a three-dimensional diagram showing a three-polarity application according to the present invention; and FIG. 8 is a block diagram showing a communication device according to the present invention. In the spring, the communication device has an antenna array, and the antenna array includes related and non-connected antenna elements. To provide a blind signal separation (BSS) processing application by providing different sums of source signals; FIG. 9 is a block diagram showing a communication device according to the present invention, wherein 'the communication device is based on array bias (such as丨〇11) operation for providing different sums of source signals to provide a blind signal separation (BSS) processing application; FIG. 10 is a block diagram S of a switched beam antenna according to the present invention, wherein the switching beam antenna The system has an elevation control 61, 1284465 device for selectively changing the height of an antenna field type (elevati〇n); FIG. 11 is a schematic diagram of an antenna field type antenna, wherein the antenna field The pattern is along the azimuth direction, and then, the antenna type is rotated according to the height (eievati〇n) controller shown in Fig. 9 in the elevation direction; Fig. 12 is based on A block diagram of an antenna element of the present invention, wherein the antenna element has a radio frequency (PJ7) anti-flow, and the radio frequency (rf) • anti-flow system is formed on the ground, thereby The antenna pattern is rotated in an elevation direction; Figure 13 is a block diagram showing a communication device according to the present invention, wherein the communication device is based on a path selection operation to provide different sums of source signals, thereby providing a blind signal. Separating (Bss) processing application; Figure 14 is a block diagram showing a communication device according to the present invention, wherein the communication device is based on a spreading digital operation to provide different sums of source signals. Further providing blind signal separation (BSS) • processing application, Fig. 15 is a block diagram showing a communication device according to the present invention, which: 'The spur is based on the in-phase (1) and quadrature (Q) signals. Staying in the hunt for the additional sum of the source 5 tigers, the rain provides blind signal separation (BSS) processing applications; the second map shows a detailed block diagram of the in-phase (I) and quadrature (Q) modules The in-phase (1) and quadrature (Q) modules are connected to the antenna element shown in FIG. 15; the seventh diagram shows the multiple input multiple output 62 1284465 (ΜΙΜΟ) according to the present invention. a block diagram of a system in which the multiple input multiple output (ΜΙΜΟ) system is based on field diversity operation; and Fig. 18 is a block diagram showing a Fourier transform communication system in accordance with the present invention, wherein the Fourier transform communication The system can deal with mutual symbol interference (ISI); Fig. 19 is a block diagram showing a communication system according to the present invention, wherein a transmitter is based on a time slot to change each layered space stream. Figure 20 is a block diagram showing a communication system in accordance with the present invention, wherein the fluctuation % type can support a plurality of transmitters transmitted to the same wireless network base station (AP); A block diagram of a receiver of the present invention, wherein the receiver is capable of optimizing processing and power absorption; and FIG. 22 is a block diagram showing the receiver shown in FIG. 21, wherein the receiver The operation is coordinated to a transmitter; the second diagram shows a schematic diagram of the transmission field profile according to the present invention as the receiver knows the timing fluctuations; The figure shows a time axis according to the present invention, wherein one symbol period has eleven different motions (ie, twelve notches), and the variation parameter is maintained fixed at four notches; A block diagram of a receiver of the invention, wherein 'the receiver is applied to a plurality of spatially independent channels; FIG. 26 is a block diagram showing a receiver decoding link according to the present invention; and 63 1284465 7 to 30 Indicates the amplitude relative to '~_, and then, the relationship between the amplitudes and the frequencies of the nodes shown in Fig. 26 is 2 A, B, D, & e 〇 [component symbol description] ADC analog digital converter ΒΒΡ Baseband processor BPF Bandpass filter C Capacitor ICA Independent component analysis L Inductor LNA Low noise amplifier LO Region oscillator LPF Low-pass filter 20 Complex signal source 20(1)~20(Μ) First to third signal source 22 plural source signals 22(1)~22 (Μ) first ~ first source signal 24 complex antenna beams 24(1)~24(5) first to pp antenna beams 30, 200, 240, 300, 400, 500 communication 32, 180, 202, 242, 3〇2, 402, 5〇2 antenna array 34 complex antenna elements 34(1) to 34(Ν) first to second antenna elements 36, 116, 214, 252, 312, 412, 512 Mixed Matrix 38, 38(1)~38(3) Separation Matrix 39, 216, 254, 314, 414, 514 Separation Signal 40 Transceiver, 42, 250, 310, 410, 510 Processor 64 1284465 44 PCA Principal Component Analysis Module 46 ICA Independent Component Analysis Module 48 SVD Signal Value Decomposition Module 49, 212 Blind Signal Separation Processor 50 Signal Analysis Module 52 Application Related Processing 1 100 Non-Connected Sensor 100, 140 Switch Beam Antenna 102 Related Antenna Array 104 polar antenna array 104f > 144 passive antenna element 104af, 144a upper half 140bf, 144b lower half 106 sensor and array combination 108 array bias 110 path selection 112 despreading digital 114 in-phase and quadrature component 140 switching beam antenna 142, 162 active antenna element 144 passive antenna element 144a upper half 144b lower half 146 ground plane 148, 108f, 118, anti J 160 phase array 164 weight control element 166 Controller/combiner 168 controller 182a, 182b, 182c, 184a, 184b, 204, 244, 274, 304, 404 504 antenna element 206 two associated antenna elements 210, 248 receiver 246 height controller 270 can control radio frequency Adjustment device 272 plane 65 1284465

308, 408, 508 Receiver Components 316 RR Target 耙 Receiver 406 Code Despreader 502 Antenna Array (Figure 15) 502 Antenna Components (Figure 16) 506 In-Phase and Orthogonal Modules 520 Mixer 66

Claims (1)

: 1284465 X. Patent application scope: 1. A communication device for separating source signals provided by one signal source, the communication device comprising: an antenna array having one connected antenna element for receiving the V [at least one different sum of the source dl numbers, wherein ν and Μ are greater than one; a receiver connected to the antenna array to receive the at least one different sum of the one source signals; and the turtle-blind signal Separating a processor, connecting to the receiver, to form a mixing matrix, wherein the mixing matrix comprises up to the at least one different sum of the one source signals, and the hybrid matrix has a rank equal to at least Ν The number of the blind signal separation processor separates the desired source signal from the mixing matrix. 2. The communication device according to item 1 of the patent application, wherein, Μ. 3. The communication device according to claim 1, wherein the rank of the hybrid matrix is Κ, where Κ <Ν, and the blind signal separation processing is separated from the mixing matrix κ source signals of the source signals. 4. The communication device according to the first aspect of the invention, wherein: the communication device of claim 1, wherein the one of the connected antenna elements comprises one active antenna The components are such that the 'antenna array' forms a phase array. The communication device of claim 1, wherein the n connected antenna elements comprise at least an active antenna element and up to N-1 passive antenna elements such that the antenna array forms a switching beam antenna. 7. The communication device of claim 1, wherein the antenna array forms at least N antenna beams, thereby receiving the % source signals, forming at least N different sums, and each antenna beam has A point below the maximum gain point of 3 dB to provide a signal rejection to at least one of the directions to reach the signal. 8. The communication device according to claim 1, wherein the antenna array forms at least an antenna field type, wherein at least one of the N different sums is formed by the _M source signals, the at least An antenna pattern does not substantially have a point 3 dB below the maximum gain point, so no signal rejects an arrival signal in any direction. 9. The communication device as described in the third paragraph of the patent application scope is in which the sum of the respective source signals is linear. 10. The communication device according to the scope of the patent application, wherein the blind signal Separating and processing the ϋ基社魏分分析 (PCA) to separate the desired source signal from the mixing matrix. 11. The communication device according to the scope of the patent application, wherein the blind signal separation processing _ Based on an independent component analysis (ICA), the desired source signal is separated from the hybrid matrix. The communication device according to claim 1, wherein the blind signal separation processor is based on a single numerical decomposition ( SVD) to separate the desired source signal from the mixing matrix. .1284465 13. The method for the communication device, the device separates the source signal provided by the M signal sources, and the communication device is supported by the rainbow And an antenna array, a receiver connected to the antenna array, and a blind signal separation processor connected to the receiver, the method comprising the steps of: receiving by the antenna array At least one different sum of the source signals, the antenna array includes one of the associated antenna elements, wherein = one; • providing the at least one different sum of the one source signals to the receiver; and the blind The signal separation processor processes the at least one different sum of the one source signals received by the receiver, the process comprising the steps of: forming a mixing matrix, wherein the mixing matrix comprises the source signals up to the at least混合 a different sum, the mixing matrix having a rank number equal to at least one of Ν, and separating the desired source signal from the mixing matrix. _14. The method of claim 13, wherein ν= The method of claim 13, wherein the connected antenna element comprises one active antenna element such that the antenna array forms a phase array. The method, wherein the one of the connected antenna elements comprises at least one active antenna element, and up to (Ν - /) passive antenna elements, such that the day The method of claim 13, wherein the antenna array forms at least one antenna beam, thereby receiving the μ source signal 69 1284465 formed to ^ is [ a different sum, wherein each antenna beam has a point that is less than 3 dB below the maximum thinning, and the Wei number is provided to one of the at least one direction to reach the signal. 18. The method of claim 13, wherein The antenna array is configured to receive an antenna field type to receive at least one sum of the different sums of the one source signals, and the at least one antenna pattern has no lower than maximum The gain point is 3dB, so no signal rejects an arrival signal in any direction. I9. The method of claim 13, wherein the respective sums of the sources §fl are linear. The method of claim 13, wherein the blind signal separation processor is based on principal component analysis (PCA) to separate the desired source signal from the mixing matrix. The method of claim 13, wherein the blind signal separation processor is based on an independent component analysis (ICA) to separate the desired source signal from the hybrid matrix. The method of claim 13, wherein the blind signal separation processor is based on a single numerical decomposition (SVD) to separate the desired source signal from the hybrid matrix. 1284465 VII. Designated representative map: (1) The representative representative of the case is: (2). (2) Brief description of the component symbols of the representative diagram: 30 Communication device 32 Antenna array 34(1) to 34(N) First to Nth antenna elements 36 Mixed matrix 38(1)~38(3) Separation matrix 39 Separation Signal 40 Transceiver 42 Processor 44 PCA Main Component Analysis 46 ICA Independent Component Analysis 48 SVD Single Value Decomposition 49 Blind Signal Separation Processor 50 Signal Analysis Module 52 Application Related Processing Module 8. If there is a chemical formula in this case, please reveal the most Chemical formula that shows the characteristics of the invention: 5