阅读说明：本技术 一种基于dsp的水听器阵测向系统及其doa估计方法 (Hydrophone array direction finding system based on DSP and DOA estimation method thereof ) 是由 王绪虎 白浩东 田雨 于 2021-08-27 设计创作，主要内容包括：本发明公开了一种基于DSP的水听器阵测向系统及其DOA估计方法。本发明的一种基于DSP的水听器阵列测向系统由水听器基阵,前端放大模块,滤波模块,多通道数据采集传输模块,数据处理模块、显示模块,控制模块等部分组成。水听器基阵由6个水听器构成的均匀线阵,前端放大模块和滤波模块的工作频率范围为200赫兹至50000赫兹,数据采集传输模块可以实现6通道同步采样传输,数据处理模块的核心处理器为ADSP-21562,显示模块用于显示DSP模块处理后的结果,控制模块用于设置DSP处理器的运算参数。为了提升本发明测向系统的稳定性和测向精度,针对水听器基阵阵元间存在互耦,且入射信号方向与离散网格存在误差的情况,在本发明系统的平台上,设计了一种基于贝叶斯学习的稳健、高精度DOA估计信号处理方法,该方法在水听器基阵阵元间存在未知互耦,且入射信号方向与离散网格存在误差的情况下,可以稳健的实现入射信号的方位估计,且估计精度优于常规处理方法。(The invention discloses a hydrophone array direction finding system based on a DSP and a DOA estimation method thereof. The invention relates to a hydrophone array direction finding system based on a DSP (digital signal processor), which comprises a hydrophone array, a front-end amplification module, a filtering module, a multi-channel data acquisition and transmission module, a data processing module, a display module, a control module and the like. The hydrophone array is an even linear array formed by 6 hydrophones, the working frequency range of the front-end amplification module and the filtering module is 200 Hz to 50000 Hz, the data acquisition and transmission module can realize 6-channel synchronous sampling and transmission, the core processor of the data processing module is ADSP-21562, the display module is used for displaying the result processed by the DSP module, and the control module is used for setting the operation parameters of the DSP processor. In order to improve the stability and the direction-finding precision of the direction-finding system, aiming at the conditions that mutual coupling exists among hydrophone array elements and errors exist between the direction of an incident signal and a discrete grid, a stable and high-precision DOA estimation signal processing method based on Bayesian learning is designed on a platform of the direction-finding system.)

1. A hydrophone array direction-finding system based on DSP is characterized in that: the system comprises a hydrophone array, a front-end amplification module, a filtering module, a multi-channel data acquisition and transmission module, a data processing module, a display module, a control module and the like, wherein the hydrophone array comprises 6 hydrophones which are arranged in a linear shape and are arranged at equal intervals; the working frequency range of the front-end amplification module is 200 Hz to 50000 Hz; the working range of the filtering module is 200 Hz to 50000 Hz, and the working bandwidth can be adjusted and set as required; the data acquisition and transmission module can realize 6-channel synchronous sampling, each sampling data is converted into a 16-bit binary number, then 6-channel parallel acquisition data is converted into serial data, and the serial data is transmitted to the data processing module; the core processor of the data processing module is ADSP-21562, the processor transforms the received data according to the set program, calculates the space spectrum, estimates the direction of arrival, and transmits the operation result to the display module, the processing process is adjusted according to the control information transmitted by the control module; the display module and the control module are placed together in parallel by two display screens, the display module is used for displaying results processed by the DSP module, the results comprise the direction of incident sound wave signals, the mutual coupling coefficient among hydrophone array elements and the space spectrum of output signals of the hydrophone array, the control module is used for setting the data processing precision, the processing data length and the operation times of the DSP processor, and the display module and the control module further comprise a switch button for starting the whole system and a distribution and recovery button for the wet-end hydrophone array.

2. The DSP-based hydrophone array direction-finding system of claim 1, wherein: the DOA estimation processing method comprises the following steps:

the method comprises the following steps: setting hyper-parameters a, c, e, b, d, f, setting precision epsilon and maximum iteration number iNum, initializing parameters X, delta, c, alpha needing to be updated_x,α_c,α_nμ, the initial value of the sparse extension matrix of the signal matrix is set to X ═ O_N×TThe initial value of the off-grid error vector is set to δ ═ O_N×1The initial value of the array element mutual coupling coefficient vector is set as c ═ 1, O_(M-1)×1]^TThe initial value of the precision vector of the signal is set to alpha_x＝O_N×1The initial value of the precision vector of the mutual coupling coefficient is set as alpha_c＝O_M×1The initial value of the noise accuracy is set to alpha_n0.01, where N represents spatial dispersionThe number of grids, T the fast beat number, M the number of elements of the hydrophone array, mu₀＝O_N×T；

Step two: calculating a cross covariance matrix sigma of the sparse signal according to the initial values of the parameters in the step one_x＝[α_nΥ^H(δ,c)Υ(δ,c)+diag(α_x)]^-1In the formulaReconstructed off-grid array manifold matrixReconstruction matrix corresponding to array manifold matrix in formulaWherein

Sub-matrix Represents the azimuth angle of the kth incident signal;

step three: preprocessing the acquired signals of the hydrophone arrayConversion of a Scherberg converter into complex array dataCovariance matrix sigma calculated by step two_xMean matrix for calculating how fast and afraid of sparse signals from array receiving data Y

Step four: updating the signal accuracy alpha_xBy using a formulaWhere N is 1,2, …, N, updating alpha by calculating N values_xThe value of the element(s);

step five: updating the noise accuracy alpha_nIs calculated byWherein y is_tT column, μ, representing array received data Y_tThe t-th column of the mean matrix mu representing the sparse signal is used to update the noise accuracy with the calculated value, i.e.

Step six: updating the cross-coupling coefficient vector c, and calculating by formulaThen orderUpdating the mutual coupling coefficient vector; using formulasCalculating a precision value and then updating the precision vector alpha using the calculated M values_c；

Step seven: calculating the parameter lambda＝||μ₀-μ_i||/||μ_iI, |, wherein μ_iCalculating the mean matrix of the sparse signals for the third step, then judging whether the parameter lambda is less than epsilon or whether the cycle number reaches the maximum iteration number iNum, and if the two conditions are not met, making mu₀＝μ_iThen returning to the step two, and if any one of the two conditions is met, performing the step eight;

step eight: using formulasAn off-grid error vector is calculated, wherein,n-th behavior of PThe nth element of v is

Step nine: computing spatial spectraWherein the content of the first and second substances,is mu row mean vector, is equal to Schur product, and then uses the off-grid error vector calculated in step eight to update the uniform discrete grid vector, i.e. command

Step ten: performing spectrum peak search on the space spectrum SP calculated in the step nine to find out the corresponding spectrum peak valueI.e. K signal directions of arrival are estimated.

Technical Field

The invention relates to a hydrophone array direction-finding system based on DSP and a DOA (direction of arrival) estimation method thereof, in particular to a hydrophone array direction-finding system based on DSP, which utilizes a hydrophone array to receive and collect radiation signals of an underwater target, processes the signals, estimates the direction of the target and visually presents an estimation result and related parameters through a display system; based on the system, the invention provides a steady DOA estimation method based on sparse Bayesian learning, in particular to a steady DOA estimation method based on sparse Bayesian learning under the condition that mutual coupling exists between array elements of a hydrophone array and errors exist between an incident direction and a discrete grid.

Background

The problem of direction of arrival estimation is of great interest in many fields, such as radar, sonar, biomedicine, and the like. Many high-resolution DOA estimation algorithms have been proposed in the past decades, such as Multiple Signal Classification (MUSIC), Root-finding MUSIC (Root-MUSIC), rotation invariant subspace (estimation) and so on, but the above-mentioned high-resolution algorithms have a serious degradation and even failure in estimation performance under low Signal-to-noise ratio and few fast beat number.

With the continuous development and maturity of Compressed Sensing (CS) theory and sparse reconstruction technology, many scholars link it with DOA estimation, which makes DOA estimation technology enter into a new development stage. Compared with the traditional high-resolution algorithm, the DOA estimation method based on the CS and the sparse reconstruction technology shows good estimation performance under the conditions of low signal-to-noise ratio and few snapshots. The method is roughly divided into three categories: convex optimization method, greedy method and sparse Bayes learning method. Malioutov studied the sparse signal representation and DOA estimation problem, put forward L₁Norm singular value decomposition (L)₁norm-Singular Value Decomposition,L₁-SVD) method, converting the direction-finding problem into solving L₁The norm problem, but the optimization problem is influenced by the model regularization parameters, and the estimation precision is influenced. DOA estimation algorithm based on sparse signal power iterative compensation proposed by Wangweason, Zhang Quei et al, which is based on the compensation principle to approximate the sparse representation of the signalIn L₀Norm, converting DOA estimation problem into solution L₀Norm problem. Compared with L₁SVD algorithm, which does not need to set regularization parameters and has higher estimation accuracy for incoherent sources. However, the computational efficiency of the convex optimization method limits its further development.

With the continuous and deep research of the CS theory, the Sparse Bayesian Learning (SBL) is considered to have the same global convergence as the convex optimization method, and its computational efficiency is far better than the latter. Ji Shihao et al introduced SBL into the CS domain and proposed a Bayesian compressed sensing algorithm. The algorithm generally needs to satisfy that the incoming wave direction of the sound source is located on the grid point, so that high azimuth estimation accuracy can be achieved. When the direction of arrival of an incident sound source deviates from the pre-divided mesh, the accuracy of the azimuth estimation is degraded. Therefore, Yang proposes an Off-Grid Sparse Bayesian Learning (OGSBL) algorithm, and in the Off-Grid model, a first-order taylor expansion approximation is adopted at a true arrival angle of a signal, so that the estimation performance is further improved. Dai Jisheng et al propose a Root-off grid Sparse Bayesian Learning algorithm (Root-OGSBL), reduce the computational complexity of the OGSBL method, and ensure the estimation accuracy while improving the computational efficiency under the condition of small grid spacing. Yang Jie et al uses the variation theory in the mean field theory to estimate the posterior distribution of the parameters, and for providing three-layer signal prior distribution, the layering prior further improves the sparsity of the signals and the precision of the orientation estimation. Subsequently, the sparse signals are put forward a new prior distribution framework of the sparse signals, namely layered synthesis Lasso prior, compared with the assumed gamma prior distribution, the sparse signals have higher sparsity and lower reconstruction errors, and the azimuth estimation precision is improved.

In an actual sonar direction finding system, array errors inevitably exist. For example, coupling between hydrophone array elements, deviation of hydrophone positions, and inconsistency of amplitude-phase channels of hydrophone array elements. The position estimation precision of the conventional algorithm is reduced, and even the conventional algorithm is invalid. Dai Jisheng et al propose an estimation algorithm of the direction of arrival under the condition that unknown mutual coupling exists between array elements, and solve the problem of recovery of sparse signals under the condition that unknown mutual coupling exists between array elements, but the method does not consider the influence of errors existing between incident signals and discrete grids. If a direction-finding system can be designed, multi-channel hydrophone signals can be collected, the multi-channel signals can be processed in real time to estimate a target direction, meanwhile, errors existing between the direction of incident signals and discrete grids in a sparse model and unknown mutual coupling among hydrophone array elements can be considered in the signal processing process, the direction-finding system and the signal processing method thereof can improve the robustness and precision of hydrophone array direction estimation, and the engineering application value of the direction-finding system is greatly improved.

Disclosure of Invention

The invention aims to design a robust and high-precision hydrophone array direction-finding system.

In order to achieve the purpose, the invention designs a hydrophone array direction-finding system based on DSP, which is characterized in that: the system comprises a hydrophone array, a front-end amplification module, a filtering module, a multi-channel data acquisition and transmission module, a data processing module, a display module, a control module and the like, wherein the hydrophone array comprises 6 hydrophones which are arranged in a linear shape and are arranged at equal intervals; the working frequency range of the front-end amplification module is 200 Hz to 50000 Hz; the working range of the filtering module is 200 Hz to 50000 Hz, and the working bandwidth can be adjusted and set as required; the data acquisition module can realize 6-channel synchronous sampling, each sampling data is converted into 16-bit binary data, then the 6-channel parallel acquisition data is converted into serial data, and the serial data is transmitted to the data processing module; the core processor of the data processing module is ADSP-21562, the processor transforms the received data according to the set program, calculates the space spectrum, estimates the direction of arrival, and transmits the operation result to the display module, the processing process is adjusted according to the control information transmitted by the control module; the display module and the control module are placed together in parallel by two display screens, the display module is used for displaying results processed by the DSP module, the results comprise the direction of incident sound wave signals, the mutual coupling coefficient among hydrophone array elements and the space spectrum of output signals of the hydrophone array, the control module is used for setting the data processing precision, the processing data length and the operation times of the DSP processor, and the display module and the control module further comprise a switch button for starting the whole system and a distribution and recovery button for the wet-end hydrophone array.

In order to improve the stability and the precision of the direction finding of the system, under the conditions that mutual coupling exists among hydrophone array elements and errors exist between the direction of an incident signal and a discrete grid, a stable and high-precision DOA estimation signal processing method based on Bayesian learning is designed and realized on a platform of the system. The stable and high-precision DOA estimation signal processing process comprises the following steps:

the method comprises the following steps: setting hyper-parameters a, c, e, b, d, f, setting precision epsilon and maximum iteration number iNum, initializing parameters X, delta, c, alpha needing to be updated_x,α_c,α_nμ, the initial value of the sparse extension matrix of the signal matrix is set to X ═ O_N×TThe initial value of the off-grid error vector is set to δ ═ O_N×1The initial value of the array element mutual coupling coefficient vector is set as c ═ 1, O_(M-1)×1]^TThe initial value of the precision vector of the signal is set to alpha_x＝O_N×1The initial value of the precision vector of the mutual coupling coefficient is set as alpha_c＝O_M×1The initial value of the noise accuracy is set to alpha_n0.01, where N denotes the number of grids discretized in the spatial domain, T denotes the number of fast beats, M denotes the number of elements of the hydrophone array, μ₀＝O_N×T；

Step two: calculating a cross covariance matrix sigma of the sparse signal according to the initial values of the parameters in the step one_x＝[α_nΥ^H(δ,c)Υ(δ,c)+diag(α_x)]^-1In the formulaReconstructed off-grid array manifold matrixReconstruction matrix corresponding to array manifold matrix in formulaWherein

Sub-matrixRepresents the azimuth angle of the kth incident signal;

step three: preprocessing the acquired signals of the hydrophone array, and converting the signals into complex array data through a Hilbert converterCovariance matrix sigma calculated by step two_xMean matrix for calculating how fast and afraid of sparse signals from array receiving data Y

Step four: updating the signal accuracy alpha_xBy using a formulaWhere N is 1,2, …, N, updating alpha by calculating N values_xThe value of the element(s);

step five: updating the noise accuracy alpha_nIs calculated byWherein y is_tT column, μ, representing array received data Y_tThe t-th column of the mean matrix mu representing the sparse signal is used to update the noise accuracy with the calculated value, i.e.

Step six: updating the cross-coupling coefficient vector c, and calculating by formula

Then orderUpdating the mutual coupling coefficient vector; using formulasM is 1,2, …, M, calculates the precision value, and then updates the precision vector α using the calculated M values_c；

Step seven: calculating parameter lambda | | | mu₀-μ_i||/||μ_iI, |, wherein μ_iCalculating the mean matrix of the sparse signals for the third step, then judging whether the parameter lambda is less than epsilon or whether the cycle number reaches the maximum iteration number iNum, and if the two conditions are not met, making mu₀＝μ_iThen returning to the step two, and if any one of the two conditions is met, performing the step eight;

step eight: using formulasAn off-grid error vector is calculated, wherein,n-th behavior of P

The nth element of v is

Step nine: computing spatial spectraWherein the content of the first and second substances,is mu row mean vector, is equal to Schur product, and then uses the off-grid error vector calculated in step eight to update the uniform discrete grid vector, i.e. commandn＝1,2,…,N；

Step ten: performing spectrum peak search on the space spectrum SP calculated in the step nine to find out the corresponding spectrum peak valueK is 1,2, …, K, i.e. K signal directions of arrival are estimated.

Drawings

FIG. 1 is a diagram of the practical application of the system

FIG. 2 is a block diagram of a hydrophone array direction-finding system based on DSP

FIG. 3 is a diagram of a control interface in a display control module

FIG. 4 is a display interface in the display control module

FIG. 5 is a spatial spectrum of the DOA estimation signal processing method of the present patent

FIG. 6 is a resolution curve diagram of DOA estimation signal processing method of the present patent

FIG. 7 is a graph showing the relationship between the estimation performance and the data length of the DOA estimation signal processing method of the present invention

Detailed Description

The invention will now be further described with reference to the embodiments and the accompanying drawings.

Fig. 1 shows an application diagram of a practical scene of a hydrophone array direction-finding system based on a DSP according to the present invention. The hydrophone array and the front end amplification module of the system are arranged underwater, the filtering, data acquisition, data processing, display and control modules are arranged on a ship on the water surface, the hydrophone array receives a target signal, the target signal is amplified and filtered and then transmitted to the data processing module, the control module transmits control parameters to the data processing module, the data processing module carries out corresponding processing transformation, output parameters and space spectrums are calculated and transmitted to the display module, and the display module displays the space spectrums, azimuth angles, mutual coupling coefficients and other parameters.

Fig. 2 shows a block diagram of a DSP-based hydrophone array direction-finding system. The system comprises a hydrophone array, a front-end amplification module, a filtering module, a multi-channel data acquisition module, a data processing module, a display module, a control module and the like, wherein the hydrophone array comprises 6 hydrophones which are arranged in a linear shape and are arranged at equal intervals; the working frequency range of the front-end amplification module is 200 Hz to 50000 Hz; the working range of the filtering module is 200 Hz to 50000 Hz, and the working bandwidth can be adjusted and set as required; the data acquisition module can realize 6-channel synchronous sampling, each sampling data is converted into 16-bit binary data, then the 6-channel parallel acquisition data is converted into serial data, and the serial data is transmitted to the data processing module; the core processor of the data processing module is ADSP-21562, the processor transforms the received data according to the set program, calculates the space spectrum, estimates the direction of arrival, and transmits the operation result to the display module, the processing process is adjusted according to the control information transmitted by the control module; the display module and the control module are placed together in parallel by two display screens, the display module is used for displaying the result processed by the DSP module, the result comprises the direction of incident sound wave signals, the mutual coupling coefficient between hydrophone array elements and the space spectrum of output signals of the hydrophone array, the control module is used for setting the data processing precision, the processing data length and the operation times of the DSP processor, and the display module and the control module further comprise a switch button of the system and a release and recovery control button of the wet-section hydrophone array.

Fig. 3 shows a control interface diagram in the display control module of the direction-finding system. The control interface is provided with a text input box for processing parameters such as data processing precision, processing data length, operation times and the like, the left side of the text box is provided with a parameter marking description, and a system startup and shutdown button and a control button for releasing and recovering a wet end hydrophone array are arranged below the interface.

Fig. 4 shows a display interface diagram in the display control module of the direction-finding system. The display interface has the input direction of the sound wave signal and the mutual coupling coefficient between the hydrophone array elements to output the text box, the output parameter description is marked above the text box, the space spectrogram shaped window output by the hydrophone array is positioned at the left side of the display screen, the left side and the lower side of the graph window are marked with coordinate axes, and the name of the graph window is marked above the graph window.

Fig. 5 shows a spatial spectrum of the DOA estimation signal processing method of the present patent. The result is obtained by simulation of matlab software on a computer, and the processing process is as follows:

the method comprises the following steps: setting the number M of hydrophones to be 10, the number N of signal processing snapshots to be 100, the number K of information sources to be 2, the signal-to-noise ratio SNR of array data to be 5dB, the distance r from the grid to be 2 degrees, and the initial direction of the signalThe array element spacing is equal to the half wavelength of the signal, and the background noise is additive white Gaussian noise;

step two: setting the super parameter b ═ d ═ f ═ 0.001, a ═ c ═ e ═ 1+ b, setting the precision epsilon ═ 10^-3And the maximum iteration number iNum is 1000, initializing the parameters X, delta, c, alpha needing to be updated_x,α_c,α_nμ, the initial value of the sparse extension matrix of the signal matrix is set to X ═ O_N×TThe initial value of the off-grid error vector is set to δ ═ O_N×1The initial value of the array element mutual coupling coefficient vector is set as c ═ 1, O_(M-1)×1]^TThe initial value of the precision vector of the signal is set to alpha_x＝O_N×1The initial value of the precision vector of the mutual coupling coefficient is set as alpha_c＝O_M×1The initial value of the noise accuracy is set to alpha_n0.01, where N91 denotes the number of meshes in the spatial domain discretization, T100 denotes the number of fast beats, and M10 denotes the array of the hydrophone arrayNumber of elements, μ₀＝O_N×T；

Step three: calculating a cross covariance matrix sigma of the sparse signal according to the initial values of the parameters in the step one_x＝[α_nΥ^H(δ,c)Υ(δ,c)+diag(α_x)]^-1In the formulaReconstructed off-grid array manifold matrixReconstruction matrix corresponding to array manifold matrix in formulaWherein

Sub-matrixRepresents the azimuth angle of the kth incident signal;

step four: generating acquisition signals of a hydrophone array, and converting the acquisition signals into complex array data through a Hilbert converterCovariance matrix sigma calculated by step two_xMean matrix for calculating how fast and afraid of sparse signals from array receiving data Y

Step five: updating the signal accuracy alpha_xBy using a formulaWhere N is 1,2, …, N, updating alpha by calculating N values_xThe value of the element(s);

step six: updating the noise accuracy alpha_nIs calculated byWherein y is_tT column, μ, representing array received data Y_tThe t-th column of the mean matrix mu representing the sparse signal is used to update the noise accuracy with the calculated value, i.e.

Step seven: updating the cross-coupling coefficient vector c, and calculating by formula

Then orderUpdating the mutual coupling coefficient vector; using formulasM is 1,2, …, M, calculates the precision value, and then updates the precision vector α using the calculated M values_c；

Step eight: calculating parameter lambda | | | mu₀-μ_i||/||μ_iI, |, wherein μ_iCalculating the mean matrix of the sparse signals for the third step, then judging whether the parameter lambda is less than epsilon or whether the cycle number reaches the maximum iteration number iNum, and if the two conditions are not met, making mu₀＝μ_iAnd then returning to step two if either of two conditions is satisfiedOne, carrying out the step eight;

step nine: using formulasAn off-grid error vector is calculated, wherein,n-th behavior of P

The nth element of v is

Step ten: computing spatial spectraWherein the content of the first and second substances,is mu row mean vector, is equal to Schur product, and then uses the off-grid error vector calculated in step eight to update the uniform discrete grid vector, i.e. commandn＝1,2,…,N；

Step eleven: performing spectrum peak search on the space spectrum SP calculated in the step nine to find out the corresponding spectrum peak valueK is 1,2, i.e. K signal directions of arrival are estimated.

FIG. 5 is a space spectrum obtained by calculation after the above process cycle is stopped, and in order to show the performance of the method of the present patent, the space spectrum of the classical MUSIC algorithm is given, L₁Spatial spectra of the SVD algorithm and the OGSBL algorithm, from which we see MUSIC and L₁The SVD algorithm estimates to be relatively poor; l is₁-SVD method is parametrized by regularizationThe influence is large, the width of a main lobe is large, and the number of side lobes is large; when the coupling coefficient is larger, the space spectrum processed by the OGSBL algorithm has more side lobes, and the performance is much worse than that of the method.

Fig. 6 shows a resolution graph of the DOA estimation signal processing method of the present patent. By using the monte carlo method, the success probabilities of the two targets are correctly estimated when the two targets are respectively 1 °,2 °, 3 °, and 12 °, and 200 statistical experiments are performed in each case, and the obtained result curve is shown in fig. 6. As can be seen from the figure, the angular resolution of the processing method proposed by the patent is obviously superior to that of the traditional MUSIC algorithm and OGSBL algorithm.

Fig. 7 shows a relation curve of the estimation performance of the DOA estimation signal processing method of the present patent to the data length. The number of snapshots for each type of experiment was changed at a step size of 25, increasing from 50 to 300, with a number of monte carlo experiments per type of experiment of 300. As can be seen from fig. 7, in the case of few snapshots, the orientation estimation accuracy of the algorithm proposed herein is poor, and when the number of snapshots increases to 100, the estimation accuracy is better than that of other algorithms. When the number of snapshots is increased to 250, the estimation error is basically unchanged, and the performance of the algorithm tends to be stable.

According to the analysis, under the conditions that mutual coupling exists between hydrophone array elements and errors exist between the direction of an incident signal and a discrete grid, the hydrophone array direction-finding system based on the DSP and the DOA estimation method thereof design a steady and high-precision direction-finding system of a hydrophone array based on the DSP. The above embodiments are preferred embodiments of the present invention, and all equivalent changes or modifications of the structure, features and processing methods described in the claims of the present invention are intended to fall within the scope of the claims of the present invention.