209 results about "Least mean square error" patented technology

A method of computing a square root or a reciprocal square root of a number in a computing device uses a piece-wise quadratic approximation of the number. The square root computation uses the piece-wise quadratic approximation in the form:in each interval i. The reciprocal square root computation uses the piece-wise quadratic approximation in the form:in each interval i. The coefficients {overscore (A)}i, {overscore (B)}i, and {overscore (C)}i, and Ai, Bi, and Ci are derived for the square root operation and for the reciprocal square root operation to reduce the least mean square error using a least squares approximation of a plurality of equally-spaced points within an interval. In one embodiment, 256 equally-spaced intervals are defined to represent the 23 bits of the mantissa. The coefficients are stored in a storage and accessed during execution of the square root or reciprocal square root computation instruction.

The invention discloses a travel time prediction method for optimizing an LSTM neural network through a particle swarm optimization algorithm, and the method comprises the following steps: S1, collecting travel time data, performing data normalization, and dividing the data into a training set and a test set proportionally; S2, optimizing each parameter of an LSTM neural network prediction model by using the particle swarm optimization algorithm; S3, inputting the parameters, optimized through the particle swarm optimization algorithm, and the training set, and performing the iterative optimization of the LSTM neural network prediction model; S4, predicting the test set through the trained LSTM neural network model, and evaluating a model error. The method is quick in optimization. Compared with a random forest, SVM and KNN in the traditional prediction algorithm, the method of the invention has the least mean square error and square error for the data prediction, and the model reducesthe calculation burden, so the method shows better prediction performance.

A radar receiver system includes a receiver, a processor including a Doppler Compensated Adaptive Pulse Compressor (DCAPC) algorithm, possible other intermediate processing and a target detector. The DCAPC algorithm processes samples of a radar return signal, applies Minimum Mean Square Error (MMSE), or alternatively matched filtering, to the radar return signal to obtain initial radar impulse response estimates, computes power estimates, estimates a range cell Doppler shift for each range cell, computes range-dependent filters, applies the MMSE filters, and then repeats the cycle for subsequent reiterative stages until a desired length−L range window is reached, thereby resolving the scatterer from noise and other scatterers.

A computation unit computes a division operation Y/X by determining the value of a divisor reciprocal 1/X and multiplying the reciprocal by a numerator Y. The reciprocal 1/X value is determined using a quadratic approximation having a form:where coefficients A, B, and C are constants that are stored in a storage or memory such as a read-only memory (ROM). The bit length of the coefficients determines the error in a final result. Storage size is reduced through use of "least mean square error"techniques in the determination of the coefficients that are stored in the coefficient storage. During the generation of partial products x2, Ax2, and Bx, the process of rounding is eliminated, thereby reducing the computational logic to implement the division functionality.

The invention discloses a target tracking method based on measurement conversion Kalman filtering state fusion, which belongs to the field of radar target tracking, and particularly relates to a statistical property calculation method utilizing Doppler radar measurement conversion errors. The state fusion target tracking method comprises the steps of: calculating statistical properties of converted measurement errors based on a predicted value measurement conversion method, regarding a position predicted value in a rectangular coordinate system as a condition to obtain a mean value and a covariance of the converted measurement errors, and eliminating correlation between the covariance of the converted measurement errors and measurement errors; and then performing state estimation on a position and Doppler measurement, and finally using a least mean square error criterion to fuse estimation results of the position and Doppler measurement, so as to obtain a final state estimation. The tracking method can be further extended to CV and CA motion models of a 3D radar.

The invention discloses a motion blurred image blind restoration method using an improved generative adversarial network. A clear image is obtained through adversarial training of a generative networkand a judgment network, which is characterized in that a blurring kernel does not need to be estimated, and end-to-end image blind restoration can be realized. The generative network is a full convolutional neural network combined with a residual error network, can deepen a network layer, and shortens the training time. The judgment network is a binary classification network consisting of a convolutional layer, a pooling layer and a full connection layer, and is used for judging the image restored by the generative network or the original clear image. A loss function adopts a least mean square error of smooth and non-saturation gradients, can optimize network training, and avoids gradient disappearance. An image fidelity term is added in the loss function of the generative network and used for constraining distribution of the restored image to be closer to the clear image.

The invention discloses a dynamic cooperative relay transmission method based on power domain non-orthogonal multiple access technology, and relates to a dynamic cooperative relay transmission method.The purpose of the invention is to solve the problems of base station slot resource waste of the existing cooperative transmission scheme based on the power domain non-orthogonal multiple access technology and low edge user reliability of the direct relay cooperative transmission scheme based on the power domain non-orthogonal multiple access technology. The process is: in a direct transmission phase, a base station S broadcasts a superposition coding signal xs of information x1 required by a central user U1 and information x2 required by an edge user U2, performing: the U1, a U3, and a relayR respectively decode x2 received signals by adopting a least mean square error algorithm, and then U1 decodes x1 using a serial interference cancellation algorithm; in a cooperative transmission phase, the S broadcasts x2 information required by the U3, the R broadcasts x2, and the U1 broadcasts x2 according to a dynamic protocol, respectively: U3 decodes x3 the received signals using MMSE; andthe U2 decodes x2 the received signals using maximum ratio combining and MMSE. The method is used in the field of wireless cooperative relay transmission.

The invention provides a mixing automatic re-transfer request method based on the multi-antenna input and output channel decomposition, which comprises a receiving end used to a do geometric average decomposition to the channel matrix to obtain the pre-code matrix and the adjusting zero weight generating matrix; it feedbacks the pre-code matrix to the sending end and uses the adjusting zero weight generating matrix to generate the least mean square error adjusting zero weight vector; it dose forward error correction coding to the sending data package, adds the cyclic redundancy check and uses pre-coding matrix to pre-code the sending data package at the sending end; it uses each sub channel receiving signal and the corresponding adjusting zero weight vector and sub channel gain to obtain each sub channel receiving signal and then dose forward error correction decoding, it uses cyclic redundancy check to check the sectionalization quality at the receiving end; if the data sectionalization has error, retransfer it and stores the error sectionalization at the receiving end, groups the error retransfer data and decodes after amalgamating it with the buffer storage group.

The invention discloses a low-order recurrence least mean square error estimation for an MIMO-OPDM channel, which relates to the wireless transmission technical field. After a pilot frequency is used to insert in to obtain a recurrence least square estimation of a time-varying channel fading, the channel fading is decomposed into a signal subspace and a noise subspace by adopting a subspace tracking method of being capable of tracking singular values and singular vectors under the non-stationary complicated noise, then an order-reduction is made according to the quantity of main singular values to obtain an auto-correlation matrix of the channel fading, and a least mean square error estimation with higher precision is obtained through the recurrence. The invention has the characteristics of having computation complexity of decreasing algorithm, higher estimation accuracy as well as good robustness and applicability, and being capable of providing channel estimation and self-adaptive equalization proposals of systems such as third generation (3G) cell mobile communication, beyond third generation (B3G) cell mobile communication, fourth generation (4G) cell mobile communication and digital TVs, wireless local area networks (WLAN), wireless wide area networks (WWAN) and so on, with an important theoretical evidence and a concrete realization method and so on.

The invention discloses a channel estimation method for a passive intelligent reflection surface based on deep learning. The channel estimation method is realized by an offline channel estimation stage and an online channel prediction stage. In the off-line channel estimation stage, in an uplink, a user side sends a pilot signal, a base station side controls an IRS to sequentially open passive elements to reflect an incident pilot signal, and the base station side receives the pilot signal and estimates corresponding cascade channel information through adoption of a minimum mean square error method. An equal probability uniform sampling method is adopted to select a small amount of sampling cascade channel information from the estimated cascade channel information, and the small amount of sampling cascade channel information and complete cascade channel information are adopted to construct a new data set; and in the online channel prediction stage, the base station side estimates a small amount of sampling cascade channel information online and inputs the sampling cascade channel information to the trained ResNet network to recover complete cascade channel information. According to the invention, the number of passive elements can be flexibly selected and the residual units of the residual neural network can be set so as to meet the service quality characteristics of different systems and users.

The invention discloses a MIMO (Multiple Input Multiple Output) relay selection method on the basis of a MMSE (Minimum Mean Square Error), which comprises the following steps that in the step 1, an information sink D selects the optimal relay and a corresponding relay processing matrix by an iterative algorithm according to all known channel matrices, i.e. a forward channel and a backward channel are subjected to singular value decomposition (SVD) to obtain a corresponding characteristic mode of each channel matrix, a mean square error (MSE) detected by a signal at the position of the information sink is calculated by using all the channel characteristic modes as equivalent channel parameters, obtaining the channel characteristic mode which meets the MMSE by an iterative greedy algorithm, and then notifying information to each node in a communication network; in the step 2, a signal source S sends a signal sequence, the relay selected in the step 1 receives the signal sequence, and the received signal is processed according to the processing matrix returned by the information sink in the step 1 and then is forwarded to the information sink D; and in the step 3, the information sink D decodes by a Wiener filter. An experimental result shows that the bit error performance of the scheme disclosed by the invention is superior to the existing scheme.

A channel calculation method in a multi-antenna system is related to a pilot design and channel calculation method of a multi-antenna system under a double-selected decline channel environment. At the sending end, the discrete Fourier transformation matrix is used to set up the optimum circulation orthogonal pilot sequence of the least square to be inserted into the sending data interleave to constitute double circulate time slot structure, at the receiving end, the circulation orthogonal property is utilized to carry out the least square channel calculation of the least mean square error, the resolving of the received pilot matrix is used to realize the channel calculation quickly and discrete cosine transformation is used to carry out more accurate channel calculation and noise variance calculation of the pilot band to obtain the calculation of the data band channel parameters with the interpolates of its domain.

The invention discloses a maneuvering target tracking system capable of processing Doppler measurement on the basis of a predicted value and a maneuvering target tracking method capable of processingDoppler measurement on the basis of a predicted value, which belong to the field of target tracking by radars. The invention adopts a measurement conversion technique, and uses a subtractive debiasingmethod for Doppler pseudo-measurement; when statistic characteristics of converted measurement errors are calculated, with a predicted position value under a rectangular coordinate system as a condition, a mean value and covariance of the converted measurement errors are obtained, and the correlation between the converted measurement error covariance and measurement errors is eliminated; then, onthe basis of converted position measurement and Doppler pseudo-measurement, state estimation is respectively carried out on the position state and Doppler pseudo-state of each motion model; a least mean square error criterion is used for merging position and Doppler pseudo-state estimation results of each motion model, so that a state estimation result of the model is obtained; a model probability mean is calculated according to the position and Doppler pseudo-state information of each model and a covariance matrix thereof; and the state estimation result of each model is weighted, so that afinal state estimation result of the target is obtained.

A calculation method for a time domain adaptive channel of MIMO system based on pilot matrix is to combine the pilot matrix with blocking circulation character with the least mean square error(LMS) adaptive algorithm. First of all, a basic pilot matrix block and a spread matrix are grown on its actual model and necessary estimation length then to acquire their cross product to get the pilot matrix to be transmitted by emitting antenna as the pilot emission signal and the receiving antenna received signal. Then the LMS adaptive algorithm is applied by picking up the internal relation between the two signals to calculate the actual channel adaptive time domain.

A method for forming an intelligent antenna up-line beam includes: matching and filtering the signals received by each element of the array antenna, to carry out digit beam shaping, proceeding in Rake synthesis to the multiroute output signals of desired users according to the estimation got from the pilot symbols, to evaluate an error signal and judge if the smallest square error is in the limit of the convergence threshold to evaluate an optimize weight value applying the algorithm of combining simplified matrix inversion and the smallest square error to process the special data channel information . The device includes an antenna formed by N antenna arrays, N RF channels, N matched filter modules, a beam form module, a channel estimation and compensation and a Rake merge module, a parameter module feed back estimation modules and a weight.

The invention discloses a binaural sound source positioning method based on delay compensation and binaural coincidence. The binaural sound source positioning method includes the steps of respectively deriving the binaural energy difference and the binaural time difference in a time domain and a frequency domain based on the least mean square error criterion, improving the delay estimation accuracy through a binaural coincidence function, after binaural positioning characteristics are obtained, determining a turning angle and a pitching angle of a sound source through a hierarchical Bayesian search strategy, and achieving the positioning aim. Firstly, it is guaranteed that signal frames applied to positioning are reliable through the binaural coincidence function; secondly, limitation of the traditional mode that two independent algorithms are required to estimate the binaural time difference and the binaural energy difference is broken through delay compensation, and means for extracting the binaural characteristics have diversity and university and are suitable for multiple environments with noise in different degrees; finally, according to a hierarchical positioning system, the number of times of characteristic matching can be effectively decreased, the time complexity of the algorithm is reduced, and the real-time performance requirements for the sound source positioning system are met.

The invention provides a weighted DV _ Hop node positioning method based on a minimum mean square error criterion, and the method comprises the steps: setting other anchor nodes to have the same weight under the minimum mean square error criterion, and obtaining a preliminary estimation value of the average distance of each hop of the anchor nodes; setting weights for other anchor nodes for the second time, and obtaining the corrected average hop-by-hop distance of the anchor nodes under the minimum mean square error criterion; calculating the connectivity of each adjacent anchor node and theaverage distance error of each hop; calculating the average hop distance of the unknown node and the estimated distance from the unknown node to the anchor node according to the average hop distance of the adjacent anchor nodes and the normalized weight value; and setting a linear solution of a weighted least square method as an iterative initial value, minimizing a weighted distance estimation error objective function through a nonlinear weighted iterative method, and taking an iterative result as a positioning coordinate of an unknown node. According to the method, the more accurate averagehop distance between the anchor node and the unknown node can be obtained, the calculated amount is small, and the positioning accuracy and stability are improved.

The invention provides a hybrid precoding design method for a wideband large-scale MIMO system. The method comprises the following steps: a simulation precoding matrix is initialized, and a vector form of the simulation precoding matrix is obtained by means of a matrix straightening operation; simulation precoding is updated by means of a Riemann manifold-based conjugate gradient method; digital precoding is updated according to a least-mean-square-error criterion; and the above steps are iterated to update simulation precoding and digital precoding until the iterative computing process is convergent. According to the method, frequency-flat simulation precoding is designed for a frequency selective channel, signal to interference plus noise ratios of various users in an actual system can be effectively increased, the transmission rate performance of the actual system can be improved, and compared with traditional pure digital precoding, the performance approximate to pure digital precoding is obtained with less hardware loss.

The present invention provides a radar angle estimation and error correction method and belongs to the target detection technological field. The method includes the following steps that: a radar arraymodel is established; the error matrixes of a receiving end and a transmitting end are obtained; the DOA (Direction of Arrival) and DOD (Direction of and Departure) of a target are obtained; the nonlinear least-mean-square error optimization models of the receiving end and the transmitting end are acquired, and the mutual coupling error matrix and amplitude-phase error matrix of the correspondingreceiving end optimization model and the mutual coupling error matrix and amplitude-phase error matrix of the corresponding transmitting end optimization model are obtained; and the mutual coupling error matrix of the receiving end optimization model is adopted to correct a receiving end mutual coupling error matrix, the amplitude-phase error matrix of the receiving end optimization model is adopted to correct a receiving end amplitude-phase error matrix, the mutual coupling error matrix of the transmitting end optimization model is adopted to correct a transmitting end mutual coupling errormatrix, and the amplitude-phase error matrix of the transmitting end optimization model is adopted to correct a transmitting end amplitude-phase error matrix. With the method of the invention adopted,the mutual coupling and amplitude-phase error of the transmitting-receiving array can be corrected, and the DOA (Direction of Arrival) and DOD (Direction of and Departure) of the target can be accurately calculated. The operation amount of the method is small.

The invention discloses a spectrum denoising method which comprises the following steps: acquiring a plurality of groups of spectrum signal samples; setting an order number and a regularization coefficient of a self-adaptive filter, selecting a minimum mean square error function as an optimal target function of the filter, and taking the samples as input signals of the filter so as to obtain output signals; based on a minimum mean square error function corresponding to a same position n of k samples, acquiring a weight coefficient vector W of the self-adaptive filter according to an Adam algorithm; calculating a signal to noise ratio of the self-adaptive filter; within a preset range of the order number and the regularization coefficient, updating the order number and the regularization coefficient of the self-adaptive filter, repeating the step of acquiring the signal to noise ratio of each self-adaptive filter, and selecting a self-adaptive filter corresponding to the maximum singleto noise ratio; performing filtering denoising on a same type of spectrum signals under a same environment condition by using the selected self-adaptive filter. Compared with a conventional standard LMS algorithm, the method disclosed by the invention is optimal in denoising effect, and rapid in convergence rate.

The invention provides a multi-feature and adaptive time delay estimation-based leakage detecting and locating method. The method comprises the steps of enhancing a water leakage signal by utilizing wavelet denoising; then taking a short-time energy feature, a short-time zero-crossing rate feature and a frequency spectrum variance of the water leakage signal as inputs of a neural network, training the neural network, and detecting whether leakage occurs or not; estimating time delay values of two sensors in receiving the leakage signal through an LMS (Least Mean Square Error) adaptive algorithm; and performing high-precision water leakage locating on pipeline leakage points through the time delay values. According to the method, time delay estimation is carried out by utilizing an LMS adaptive filter; the adaptive filter can perform adaptive dynamic adjustment, so that statistical features of signals and noises do not need to be pre-obtained; and the method is suitable for a water supply pipeline leakage detection environment.

The invention discloses a parameter wavelet threshold signal denoising method based on an improved artificial bee colony algorithm. The parameter wavelet threshold signal denoising method comprises the steps: firstly obtaining a to-be-denoised signal, carrying out wavelet transformation, and obtaining a wavelet coefficient; designing a new threshold function on the basis of a traditional thresholdfunction, proving the property of the new threshold function through mathematical derivation, and determining threshold parameters to be optimized; improving an original artificial bee colony algorithm; taking a mean square error between the to-be-denoised signal and the denoised signal as a fitness function of the improved artificial bee colony algorithm in S3, and obtaining an optimal thresholdparameter under the condition of obtaining a minimum mean square error; and applying the optimal threshold parameter obtained in the step S4 to the new threshold function in the step S2, performing shrinkage processing on the wavelet coefficient to obtain a new wavelet coefficient, and performing inverse wavelet transform to obtain a denoised signal. According to the parameter wavelet threshold signal denoising method, a smaller mean square error, a higher output signal-to-noise ratio and a larger noise rejection ratio can be obtained.

The invention relates to a signal detecting method for a multi-antenna two-way relay channel. The method is characterized in that a physical layer network coding technology is adopted in the multi-antenna two-way relay channel, and the bit error rate of a system is reduced by adopting a combined detection and interference elimination algorithm and a combined modulation and demodulation scheme at a relay node. According to the combined detection and interference elimination algorithm, the conventional least mean-square error detection algorithm is applied to the multi-antenna two-way relay system, and the estimated value of source node transmitting signals, which is obtained by detection, is subjected to parallel interference elimination, so that mutual interference between the source node transmitting signals can be furthest reduced. In addition, a new combined modulation constellation mapping scheme is proposed at the relay node according to the modulation modes of two source nodes and based on the constellation point design criteria of minimum Euclidean distance maximization. Compared with a relay node signal processing scheme for completing information combination by using xor operation, the method under the asymmetrical transmission mode that different modulation modes are adopted at the two source nodes has the advantages that the minimum Euclidean distance between relay node broadcasting signals is prolonged, the bit error rate of the system is reduced, and the transmission reliability of a two-way cooperative communication system is improved.

The technique of the present invention temporarily registers a tone error, which arises in each pixel due to determination of the dot on-off state in the pixel, in an intermediate buffer. The technique calculates diffusion errors or error divisions, which are diffused to each of non-processed pixels in a neighborhood of a preset number of pixels, from tone errors arising in the preset number of pixels, and writes the error divisions into an error buffer. One procedure diffuses each tone error into the intermediate buffer in response to every occurrence of the tone error, and subsequently writes the diffused error divisions into the error buffer. Another procedure registers tone errors arising in the preset number of pixels in the intermediate buffer and writes the error divisions, which are calculated from the registered tone errors with regard to the preset number of pixels, into the error buffer. Compared with the conventional technique that diffuses each tone error into the error buffer in response to every occurrence of the tone error, either of these procedures enables collective diffusion of the tone errors with regard to the preset number of pixels, thus ensuring higher-speed diffusion of the tone error. The error diffusion may be attained by utilizing a procedure mathematically equivalent to the least mean square error method.