209 results about "Long-term prediction" patented technology

The present invention teaches a new audio coding system that can code both general audio and speech signals well at low bit rates. A proposed audio coding system comprises a linear prediction unit for filtering an input signal based on an adaptive filter; a transformation unit for transforming a frame of the filtered input signal into a transform domain; a quantization unit for quantizing a transform domain signal; a long term prediction unit for determining an estimation of the frame of the filtered input signal based on a reconstruction of a previous segment of the filtered input signal; and a transform domain signal combination unit for combining, in the transform domain, the long term prediction estimation and the transformed input signal to generate the transform domain signal.

Real-time high-fidelity spatiotemporal data on transportation networks can be used to learn about traffic behavior at different times and locations, potentially resulting in major savings in time and fuel. Real-world data collected from transportation networks can be used to incorporate the data's intrinsic behavior into a time-series mining technique to enhance its accuracy for traffic prediction. For example, the spatiotemporal behaviors of rush hours and events can be used to perform a more accurate prediction of both short-term and long-term average speed on road-segments, even in the presence of infrequent events (e.g., accidents). Taking historical rush-hour behavior into account can improve the accuracy of traditional predictors by up to 67% and 78% in short-term and long-term predictions, respectively. Moreover, the impact of an accident can be incorporated to improve the prediction accuracy by up to 91%.

A multi-rate speech codec supports a plurality of encoding bit rate modes by adaptively selecting encoding bit rate modes to match communication channel restrictions. In higher bit rate encoding modes, an accurate representation of speech through CELP (code excited linear prediction) and other associated modeling parameters are generated for higher quality decoding and reproduction. A speech encoder employing various encoding schemes based upon parameters including an available transmission bit rate. In addition, the speech encoder is operable to identify and apply an optimal encoding scheme for a given speech signal. The speech encoder may be applied code-excited linear prediction when the available bit rate is above a predetermined upper threshold. Pitch preprocessing, including continuous warping, may be applied when it is below a predetermined lower threshold. The encoder considers varying characteristics of the speech signal including the long term prediction mode of a previous frame, and a spectral difference between the line spectral frequencies of a current and a previous frame, a predicted pitch lag, an open loop pitch lag, a closed loop pitch lag, a pitch gain, and a pitch correlation.

A multi-rate speech codec supports a plurality of encoding bit rate modes by adaptively selecting encoding bit rate modes to match communication channel restrictions. In higher bit rate encoding modes, an accurate representation of speech through CELP (code excited linear prediction) and other associated modeling parameters are generated for higher quality decoding and reproduction. A speech encoder employing various encoding schemes based upon parameters including an available transmission bit rate. In addition, the speech encoder is operable to identify and apply an optimal encoding scheme for a given speech signal. The speech encoder may be applied code-excited linear prediction when the available bit rate is above a predetermined upper threshold. Pitch preprocessing, including continuous warping, may be applied when it is below a predetermined lower threshold. The encoder considers varying characteristics of the speech signal including the long term prediction mode of a previous frame, and a spectral difference between the line spectral frequencies of a current and a previous frame, a predicted pitch lag, an open loop pitch lag, a closed loop pitch lag, a pitch gain, and a pitch correlation.

A method of concealing transmission error in a digital audio signal in which a signal that has been decoded after transmission is received, the samples decoded while the transmitted data is valid are stored, at least one short-term prediction operator and one long-term prediction operator are estimated as a function of stored valid samples, and any missing or erroneous samples in the decoder signal are generated using the operators estimated in this way, the method being characterized in that the energy of the synthesized signal as generated in this way is controlled by means of a gain that is computed and adapted sample by sample.

For determining a long-term-prediction delay parameter characterizing a long term prediction in a technique using signal modification for digitally encoding a sound signal, the sound signal is divided into a series of successive frames, a feature of the sound signal is located in a previous frame, a corresponding feature of the sound signal is located in a current frame, and the long-term-prediction delay parameter is determined for the current frame while mapping, with the long term prediction, the signal feature of the previous frame with the corresponding signal feature of the current frame. In a signal modification method for implementation into a technique for digitally encoding a sound signal, the sound signal is divided into a series of successive frames, each frame of the sound signal is partitioned into a plurality of signal segments, and at least a part of the signal segments of the frame are warped while constraining the warped signal segments inside the frame. For searching pitch pulses in a sound signal, a residual signal is produced by filtering the sound signal through a linear prediction analysis filter, a weighted sound signal is produced by processing the sound signal through a weighting filter, the weighted sound signal being indicative of signal periodicity, a synthesized weighted sound signal is produced by filtering a synthesized speech signal produced during a last subframe of a previous frame of the sound signal through the weighting filter, a last pitch pulse of the sound signal of the previous frame is located from the residual signal, a pitch pulse prototype of given length is extracted around the position of the last pitch pulse of the sound signal of the previous frame using the synthesized weighted sound signal, and the pitch pulses are located in a current frame using the pitch pulse prototype.

Base layer coding section 101 encodes an input signal to obtain base layer coded information. Base layer decoding section 102 decodes the base layer coded information to obtain a base layer decoded signal and long term prediction information (pitch lag). Adding section 103 inverts the polarity of the base layer decoded signal to add to the input signal, and obtains a residual signal. Enhancement layer coding section 104 encodes a long term prediction coefficient calculated using the long term prediction information and the residual signal to obtain enhancement layer coded information. Base layer decoding section 152 decodes the base layer coded information to obtain the base layer decoded signal and long term prediction information. Using the long term prediction information, enhancement layer decoding section 153 decodes the enhancement layer coded information to obtain an enhancement layer decoded signal. Adding section 154 adds the base layer decoded signal and enhancement layer decoded signal to obtain a speech / sound signal. It is thereby possible to implement scalable coding with small amounts of calculation and coded information.

In accordance with one aspect of the invention, a selector supports the selection of a first encoding scheme or the second encoding scheme based upon the detection or absence of the triggering characteristic in the interval of the input speech signal. The first encoding scheme has a pitch pre-processing procedure for processing the input speech signal to form a revised speech signal biased toward an ideal voiced and stationary characteristic. The pre-processing procedure allows the encoder to fully capture the benefits of a bandwidth-efficient, long-term predictive procedure for a greater amount of speech components of an input speech signal than would otherwise be possible. In accordance with another aspect of the invention, the second encoding scheme entails a long-term prediction mode for encoding the pitch on a sub-frame by sub-frame basis. The long-term prediction mode is tailored to where the generally periodic component of the speech is generally not stationary or less than completely periodic and requires greater frequency of updates from the adaptive codebook to achieve a desired perceptual quality of the reproduced speech under a long-term predictive procedure.

A three-dimensional urban traffic road network global state prediction method under a large data environment is disclosed. Aiming at a coupling multi-layer large-scale urban traffic road network, a global traffic mode containing coupling information is defined, a method of extracting a main road network characteristic is provided, a high dimension state time sequence is converted into a discrete state type sequence, and difficulties of high dimension data in processing and prediction aspects are simplified. From aspects of traffic periodicity, delay performance and other characteristics, a historical state transfer information database containing a multielement traffic characteristic is constructed, mass data information and a data driving idea are fully used and a Markov transition probability theory is applied so as to realize multi-step global traffic state prediction. Through the global traffic prediction, a traffic state of each area can be acquired one time. Compared to an existing traffic prediction method, by using the method of the invention, a long-term prediction result is acquired rapidly and a good prediction output effect is possessed.

Techniques for performing adaptive and robust prediction. Prediction techniques are adaptive in that they use a minimal amount of historical data to make predictions, the amount of data being selectable. The techniques are able to learn quickly about changes in the workload traffic pattern and make predictions, based on such learning, that are useful for proactive response to workload changes. To counter the increased variability in the prediction as a result of using minimal history, robustness is improved by checking model stability at every time interval and revising the model structure as needed to meet designated stability criteria. Furthermore, the short term prediction techniques can be used in conjunction with a long term forecaster.

A method, device, and apparatus provide the ability to predict a portion of a polyphonic audio signal for compression and networking applications. The solution involves a framework of a cascade of long term prediction filters, which by design is tailored to account for all periodic components present in a polyphonic signal. This framework is complemented with a design method to optimize the system parameters. Specialization may include specific techniques for coding and networking scenarios, where the potential of each enhanced prediction is realized to considerably improve the overall system performance for that application. One specific technique provides enhanced inter-frame prediction for the compression of polyphonic audio signals, particularly at low delay. Another specific technique provides improved frame loss concealment capabilities to combat packet loss in audio communications.

A speech coding method of significantly reducing error propagation due to voice packet loss, while still greatly profiting from a pitch prediction or Long-Term Prediction (LTP), is achieved by limiting or reducing a pitch gain only for the first subframe or the first two subframes within a speech frame. The method is used for a voiced speech class; a pitch cycle length is compared to a subframe size to decide to reduce the pitch gain for the first subframe or the first two subframes within the frame. Speech coding quality loss due to the pitch gain reduction is compensated by increasing a bit rate of a second excitation component or adding one more stage of excitation component only for the first subframe or the first two subframes within the speech frame.

The invention discloses a prediction method based on a time sequence. The prediction method comprises the following steps of: continuing an original time sequence by using a mean generating function (MGF) method to obtain continuation sequences; screening the continuation sequences by using an optimal subset regression (OSR) method to obtain an optimal subset; by using the optimal subset which is obtained in the preceding step and using the original time sequence as output, training a back propagation (BP) neural network to obtain a BP neural network prediction model; and predicting by using the BP neural network prediction model. By combining the MGF method, the OSR method and the BP neural network, a novel MGF-OSR-BP prediction model is established. By focusing on the prediction model and construction of a learning matrix, the accuracy rate of prediction is higher, and reference can be provided for medium-and long-term prediction research of similar time sequence elements.

A method of determining the reliability of long-term predicted orbit data, includes: determining the reliability of long-term predicted orbit data, which is acquired by predicting a satellite orbit in a target period of at least one day, using predicted position data including predicted positions of a positioning satellite in time series and actual position data including actual positions of the positioning satellite corresponding to the predicted positions.

The invention discloses a fault prediction method and device for industrial equipment based on an LSTM circulating neural network, wherein the method comprises the following steps of: acquiring a state monitoring data set of a plurality of sensors at the periphery of target equipment, wherein the state monitoring data set comprises monitoring data from 0 moment to a current moment; selecting a prediction characteristics containing preset fault information from the state monitoring data set by utilizing a characteristic selection standard, wherein the characteristic selection standard comprisesa correlation index and a monotonicity index; performing characteristic conversion on the prediction characteristics to obtain a prediction characteristic vector; and performing single-step fault prediction, long-term fault prediction and residual life prediction on the target equipment according to the prediction characteristic vector and a fault prediction network model. The method can effectively avoid insufficient prediction precision caused by unreasonable preset fault threshold, can give a confidence interval under the occasion of single-step performance prediction, and can achieve long-term prediction of performance and residual service life of the equipment.

A method and apparatus for prediction in a speech-coding system is provided herein. The method of a 1st order long-term predictor (LTP) filter, using a sub-sample resolution delay, is extended to a multi-tap LTP filter, or, viewed from another vantage point, the conventional integer-sample resolution multi-tap LTP filter is extended to use sub-sample resolution delay. This novel formulation of a multi-tap LTP filter offers a number of advantages over the prior-art LTP filter configurations. Particularly, defining the lag with sub-sample resolution makes it possible to explicitly model the delay values that have a fractional component, within the limits of resolution of the over-sampling factor used by the interpolation filter. The coefficients of such a multi-tap LTP filter are thus largely freed from modeling the effect of delays that have a fractional component. Consequently their main function is to maximize the prediction gain of the LTP filter via modeling the degree of periodicity that is present and by imposing spectral shaping.

To implement scalable coding, a base layer coding section encodes an input signal to obtain base layer coded information, which is decoded by a base layer decoding section to obtain a base layer decoded signal and long term prediction information (pitch lag). An adding section inverts the polarity of the base layer decoded signal to add to the input signal, and obtains a residual signal. An enhancement layer coding section encodes a long term prediction coefficient calculated using the long term prediction information and the residual signal to obtain enhancement layer coded information. Also using the long term prediction information, an enhancement layer decoding section decodes the enhancement layer coded information to obtain an enhancement layer decoded signal. An adding section adds the base layer decoded signal and enhancement layer decoded signal to obtain a speech / sound signal.

The invention discloses a spacecraft lithium ion battery cycle life prediction method. The method comprises the following steps: firstly collecting capacity data of a lithium ion battery; computing ahealth state time sequence SOHBAT of the lithium ion battery; decomposing the health state time sequence SOHBAT of the lithium ion battery by applying an empirical mode decomposition model; predictinga global degradation trend of the SOHBAT based on an ARIMA model; predicting the local regeneration and fluctuation of the SOHBAT based on the GPR model; integrating prediction results of the ARIMA model and the GPR model, and acquiring the spacecraft lithium ion battery cyclic life prediction. The global degradation trend and the local capacity regeneration and fluctuation phenomenon of the battery health state time sequence SOHBAT can be effectively extracted by utilizing the empirical mode decomposition method, the global trend and local fluctuation phenomenon of the battery health changecan be simulated by utilizing each of the ARIMA model and the GPR model, the capacity regeneration and fluctuation prediction problem in the battery performance degradation can be effectively solved,so that real health degradation trend of the lithium battery can be captured by integrating the models, and the accuracy of the lithium battery long-term prediction is improved.

Reference frames are selected for use in a frame reference buffer in a video encoding system, so as to provide long term prediction of motion in video sequences. Candidate reference frames are characterized by characteristic vectors. From the candidate reference frames, a number of reference frames are maintained in the frame buffer so as to obtain a specified distribution of the characteristic vectors over a vector space. Frames are selected for use in encoding a current frame based on the characteristic vector of the current frame and the reference frames in the frame buffer.

A method, device, and apparatus provide the ability to predict a portion of a polyphonic audio signal for compression and networking applications. The solution involves a framework of a cascade of long term prediction filters, which by design is tailored to account for all periodic components present in a polyphonic signal. This framework is complemented with a design method to optimize the system parameters. Specialization may include specific techniques for coding and networking scenarios, where the potential of each enhanced prediction is realized to considerably improve the overall system performance for that application. One specific technique provides enhanced inter-frame prediction for the compression of polyphonic audio signals, particularly at low delay. Another specific technique provides improved frame loss concealment capabilities to combat packet loss in audio communications.

The invention relates to the technical field of intelligent traffic, and discloses a parking guidance method and device, and a medium. The method comprises the following steps: receiving a parking lotberth prediction request sent by a client, wherein the berth prediction request comprises a starting point and a destination of travel; predicting an arrival moment of the destination according to the starting point and the destination of travel; determining a parking lot in a designated range with the destination as a benchmark; and according to a pre-built long-term prediction model, determining an unoccupied berth number of each parking lot within the designated range at the arrival moment, and sending the unoccupied berth number to the client for display, wherein the long-term predictionmodel is a model which is built according to historical parking data of the parking lot in advance and can predict the unoccupied berth number of each parking lot in different time periods. The futureunoccupied berths are predicted through the long-term prediction model, so that parking guidance is facilitated. In addition, the unoccupied berths can be viewed in real time through a short-term prediction model, so that the berth guidance efficiency is further improved.

A speech coding method of significantly reducing error propagation due to voice packet loss, while still greatly profiting from a pitch prediction or Long-Term Prediction (LTP), is achieved by limiting or reducing a pitch gain only for the first subframe or the first two subframes within a speech frame. The method is used for a speech class decided by a classification algorithm; the classification algorithm is designed, depending on at least one pitch cycle length compared to one subframe size. Speech coding quality loss due to the pitch gain reduction is compensated by increasing a coded excitation codebook size or adding one more stage of excitation only for the first subframe or the first two subframes within the speech frame.

PCT No. PCT / GB91 / 02291 Sec. 371 Date Jul. 27, 1993 Sec. 102(e) Date Jul. 27, 1993 PCT Filed Dec. 20, 1991 PCT Pub. No. WO92 / 11627 PCT Pub. Date Jul. 9, 1992A codebook excited speech coder analyzes speech to produce coefficients of a synthesis filter, the parameters of a long-term prediction filter (LTP) and a codeword indication one of a set of excitations. The results are transmitted to a receiver where they can be used to resynthesize the speech. The LTP and excitation analysis involve generation of impulse response products by adding additional terms to products already formed and storing them in a store. Multiplication of these products by excitation terms is performed using a pointer table storing precalculated addresses of locations in the store. If some excitations are shifted versions of others, some addresses can be obtained by modifying other addresses. The LTP analysis may include selection between a simple delay prediction and a prediction including the sum of two differently delayed terms, to provide improved predictor delay resolution.

The invention belongs to the building energy consumption prediction field, and provides an energy utilization management method so as to achieve the purpose of energy saving and amend the energy saving weak links. Therefore, the adopted technical scheme of the method is that the office energy consumption prediction method comprises the following steps: 1) establishing a people-in-the-room rate model; 2) establishing a device hourly power model; 3) establishing a lamp hourly power model; 4) calculating hourly power of indoor consumed energy according to the hourly number of people obtained from the step 1) and the hourly power obtained from the step 2)- the step 4), and with the running time being combined, obtaining daily average power consumption amount of target offices in the long-term prediction; and 5) providing a power consumption optimization scheme. The method is mainly applied to the building energy consumption management.

For determining a long-term-prediction delay parameter characterizing a long term prediction in a technique using signal modification for digitally encoding a sound signal, the sound signal is divided into a series of successive frames, a feature of the sound signal is located in a previous frame, a corresponding feature of the sound signal is located in a current frame, and the long-term-prediction delay parameter is determined for the current frame while mapping, with the long term prediction, the signal feature of the previous frame with the corresponding signal feature of the current frame. In a signal modification method for implementation into a technique for digitally encoding a sound signal, the sound signal is divided into a series of successive frames, each frame of the sound signal is partitioned into a plurality of signal segments, and at least a part of the signal segments of the frame are warped while constraining the warped signal segments inside the frame. For searching pitch pulses in a sound signal, a residual signal is produced by filtering the sound signal through a linear prediction analysis filter, a weighted sound signal is produced by processing the sound signal through a weighting filter, the weighted sound signal being indicative of signal periodicity, a synthesized weighted sound signal is produced by filtering a synthesized speech signal produced during a last subframe of a previous frame of the sound signal through the weighting filter, a last pitch pulse of the sound signal of the previous frame is located from the residual signal, a pitch pulse prototype of given length is extracted around the position of the last pitch pulse of the sound signal of the previous frame using the synthesized weighted sound signal, and the pitch pulses are located in a current frame using the pitch pulse prototype.

The invention discloses a power load medium and long term prediction method based on an improved grey prediction model. In order to solve the problem that large errors are easily generated in a fixed model prediction time varying system, a model updating algorithm is provided, in other words, a method combining a sliding window algorithm with a traditional grey prediction model is adopted. The prediction model is recursively updated in a mode that a part of old data are deleted and a part of new data are introduced in a modeling sequence, key quantity a&b related to the mathematic model is decomposed, a recursive mathematical formula is simplified, and the influence of the length of a sliding window on modeling accuracy is discussed.