Showing papers on "Encoding (memory) published in 1990"

Two coding strategies for bidirectional associative memory

Abstract: Enhancements of the encoding strategy of a discrete bidirectional associative memory (BAM) reported by B. Kosko (1987) are presented. There are two major concepts in this work: multiple training, which can be guaranteed to achieve recall of a single trained pair under suitable initial conditions of data, and dummy augmentation, which can be guaranteed to achieve recall of all trained pairs if attaching dummy data to the training pairs is allowable. In representative computer simulations, multiple training has been shown to lead to an improvement over the original Kosko strategy for recall of multiple pairs as well. A sufficient condition for a correlation matrix to make the energies of the training pairs be local minima is discussed. The use of multiple training and dummy augmentation concepts are illustrated, and theorems underlying the results are presented. >

Memory Mechanisms: A Tribute To G.v. Goddard

Abstract: Contents: Part I:Models of Neural Associativity. G. Lynch, R. Granger, U. Staubli, Long-Term Potentiation and the Structure of Memory. T.J. Teyler, A Model of Associative Interactions Mediated by N-Methyl-D-Aspartate Activation. Part II:Long-Term Potentiation in the Hippocampus. P. Andersen, Parameters Controlling the Induction of Long-Term Potentiation. W.C. Abraham, D.K. Bilkey, E.W. Kairiss, Long-Term Potentiation and Local Circuits in the Hippocampus. G. Buzsaki, F.H. Gage, Long-Term Potentiation: Does It Happen in the Normal Brain? When and How? Part III:EEG States and Memory. R.J. Racine, The Kindling Phenomenon: Epilepsy Model or Memory Model? (and Other Areas of Controversy). N. McNaughton, The Role of Hippocampal RSA: Spatial Memory Versus Anxiety. Part IV:Memory Disorders. R.G. Knight, B.E. Longmore, What Is an Amnesic? H.J. Markowitsch, Memory Disorders After Diencephalic Damage. J.A. Ogden, S. Corkin, Memories of H.M. Part V:Memory Models and Processes. W.E. Hockley, Interrogating Memory: A Decision Model for Recognition and Judgment of Frequency. A.A. Wright, Concept Learning by Monkeys and Pigeons. D.C. McCarthy, J.A. Nevin, The Consequences of Remembering. Part VI:Mechanisms of Nonhuman Memory. W.K. Honig, Structure and Function in the Spatial Memory of Animals. P.J. Urcuioli, A Search for Anticipatory Codes in Pigeon Working Memory.

Is memory for spatial location automatically encoded

Abstract: Naveh-Benjamin (1987, 1988) has shown that memory for spatial location does not meet the criteria for automatic encoding as claimed by Hasher and Zacks(1979). Age, intention, concurrent processing demands, practice, strategies, and individual differences affected memory for location. These variables should have affected effortful but not automatic processing. The experiments reported in the present paper, in which a different task was used, showed that intention, practice, and concurrent processing demands did not affect memory for location. I concluded that (1) the location task used by Naveh-Benjamin included effortful subtasks and also incidental, cover or concurrent processing tasks that interfered directly with performance, and (2) the variables that he manipulated may not have affected the encoding of location. The need to differentiate processes from task performance in analyzing the automaticity issue is discussed. The dominant mode for remembering location is automatic, but such information may also be remembered voluntarily.

Signal processing device

Abstract: A signal processing device in which compressed encoded information obtained by encoding an input information signal based on use of correlation between components thereof is written in a memory capable of simultaneously performing write and read operations, and the encoded information read out from the memory is decoded to restore the information signal while signal processing is effected by using the information signal prior to the encoding and the decoded information signal.

Chapter Three Aging and Schematic Influences on Memory

Abstract: Summary Variations in content knowledge have been shown to be related to both individual and developmental differences in memory performance. The availability of relevant knowledge appears to influence the efficiency of encoding and retrieval operations and to provide a meaningful interpretive structure for storing information in memory. The present chapter explores the possibility that adult age differences exist in the extent to which knowledge influences memory performance. One specific approach to the study of knowledge effects on memory–schema theory–is discussed and used as a framework for examining the relevant literature on memory and aging. It is concluded that the relation between the individual's knowledge and the nature of the memory task is more important in determining performance in older than in younger adults. Although uncertain at present, it appears as if the age differences may be related to both the relative efficiency of knowledge-based memory processes and age-related variations in memory styles.

Holter function data encoder for implantable device

Abstract: A encoding/decoding algorithm for storing cardiac cycle length data in a memory of a body implantable device. The data encoding is based on a Holter function for sequentially storing in addresses in memory data representative of cardiac cycle length intervals and data representative of changes in the cardiac cycle length intervals. The sequence of storing is dictated by the value of the time interval change data. If the value of the time interval change data is greater than the storage capacity of a single memory location, then the time interval data is stored in memory at two consecutive memory locations. On the other hand, if the magnitude of the timer interval data is within the storage capacity of a single memory cell, it is stored in memory at a single memory cell. When decoding the stored data, retrieval proceeds through the memory according to the value of the data stored so that the time interval data is reconstructed from the originally stored time interval data or change in time interval data.

Apparatus for mix-run encoding of image data

Abstract: An encoder (12) in an image-display system converts explicitly represented pixel values from an image source (14) into mix-run-encoded representations thereof and stores them into the locations of a display memory (16). A display mechanism (18) draws the resultant stored data from the display memory and interprets them in accordance with a mix-run-encoding scheme of a type previously used for anti-aliasing purposes. As a consequence, the system is able to provide a wide range of color shades with only modest-sized display and palette memories (16 and 36).

Multiprocessor type time varying image encoding system and image processor

Abstract: A multiprocessor type time varying image encoding system having a plurality of digital signal processor (DSP) modules (DMM's) connected in parallel, each DMM having a DSP, a local memory and an interrupt control unit, a plural-ity of common memories for storing data which is being processed, parameters, etc., an input frame memory which enables reading and writing operations to be executed asyn-chronously, a combination of a task control unit and a task table for distributing tasks to the DMM's, a plurality of independent common buses, and a combination of a bus control unit and a bus control table for bus sharing control.

Bidirectional associative memory with several patterns

Abstract: The bidirectional associative memory (BAM2), so far limited to two input/output patterns, is generalized to several input/output patterns (BAMq). Earlier results on the memory capacity of BAM2 are generalized to BAMq. The consequences of such schemes for data fusion and pattern splitting are evaluated, and the encoding of several temporal patterns (ordered sets of vectors) in temporal associative memories (TAMq) is sketched

A recording medium library management system

Abstract: A digital reproduction system in particular an audio, video or audio-video digital system includes a tracking and reading means (2, 4, 6) for reading first digital data recorded on a recording medium (3), a memory (12) for storing second digital data, an encoder (11) for encoding a part of the digital data read by the tracking and reading means (2, 4, 6) and display means (9; 14; 17; 18) for displaying the second digital data read from the memory (12). The address location in the memory (12) of the second digital data which is associated with the first digital data is determined by encoding a part of the first digital data read from the recording means. In this way, bibliographical data stored in the memory (12) can be addressed and displayed for the relevant part of pre-recorded information on the recording medium. The second digital data can be input to the memory by a keyboard (10) or be read directly from the recording medium (3) or from another recording medium (3) which serves as a general catalogue. The control of the system may be carried out by a central processing unit (11).

Complex Maze Learning in Rodents

Abstract: An extensive body of research findings supports the view that normal aging in humans is associated with a defined dysfunction in memory processing. When sensorimotor performance requirements of tasks can be equated across age groups, the most salient age-related change appears in the ability to use secondary memory systems which represent the unlimited permanent store of newly acquired information (Poon, 1985). Age differences in sensory memory, which refers to the brief storage of sensory information; in primary memory, which refers to the temporary, limited capacity repository available for immediate recall; and in tertiary memory, which refers to memory for remote events, appear less robust (Poon, 1985). Much of the age-related impairment in retrieval of information from secondary memory can be attributed to deficient encoding processes, including organizational skills, visual imagination, depth of processing, and attention, as well as problems of equating motivational factors (Kausler, 1982; Poon, 1985). In terms of a storage-deficit hypothesis of age-related memory dysfunction, reviews of the literature have indicated that age differences in forgetting rates are minimal or nonexistent when effort is made to equate the original level of learning (Poon, 1985).

Multiprocessor type moving image encoder and bus control method

Abstract: PURPOSE: To utilize the throughput of a multiprocessor at a maximum by dividing a picture into plural blocks and sharing a processing task respectively equally for plural unit processor modules. CONSTITUTION: The picture is divided into the plural blocks by a task control part 7 and while referring a task table 8 storing information required for controlling the unit processors, a processing block and a processing task optimum to each unit processor module 11 are decided. Then, encoding is executed while sharing the processing task respectively equally for the plural unit processor modules 11. Plural shared memories 10 to store local decoding data or data under encoding and parameter are connected through plural memory buses, which are independently provided, to the respective plural unit processor modules 11 and the plural memory buses can be utilized for the access of the shared memory. Thus, the throughput of the multiprocessor can be utilized at a maximum. COPYRIGHT: (C)1991,JPO&Japio

Method and device for pre-processing moving image signal and encoding device using them

Abstract: PURPOSE:To suppress the generation of excessive information quantity accompanied with sudden motion, and to remove the discontinuity of the motion due to the stopping of encoding by switching the characteristic of a processing filter in conformity to the situation of the moving image signal of an input CONSTITUTION:A pre-processing circuit is provided with a frame memory 1, a control circuit 2 and a pre-processing filter 100 Then, the moving image signal of the input supplied to the frame memory 1 and the control circuit 2 The frame memory 1 delays the supplied moving image signal by one frame time, and supplies it to the control circuit 2 The control circuit 2 generates a switching signal to select the passing characteristic of the pre-processing filter 100 based on the supplied moving image signal of a present frame and the moving image signal of the previous frame supplied by the frame memory 1

Three-dimensional orthogonal transformation adaptive coding system

Abstract: PURPOSE:To perform the encoding of image data with high efficiency by a moving image by performing three-dimensional orthogonal transformation by a three-dimensional block in which a time base direction is considered. CONSTITUTION:An image memory 1 stores digital image data corresponding to L continuous images on a time base supplied as input. The segmentation circuit 2 of the image segments the data as the block having size of (MXN), and supplies it in a three-dimensional orthogonal transformation circuit 3. As for segmented digital image data, the orthogonal transformation is applied on the L images (P1-P(L-1)) at every block group at a corresponding position, respectively in the three-dimensional orthogonal transformation circuit 3, and a required orthogonal transform coefficient is supplied to a motion decision part 4 and a definition decision part 5. By performing bit allocation for adaptive quantization by utilizing the information of the motion of the image and that of the definition of the image, the encoding with high efficiency can be attained.

Clinical Memory Assessment: Matching the Method to the Aim

Abstract: In 1977 Erickson and Scott published a review of methods of memory assessment in clinical settings. Their conclusion was that the procedures then in common use, such as the Wechsler Memory Scale (Wechsler, 1945) and the Benton Visual Retention Test (Benton, 1974), were of limited clinical value. In particular, they found that undue emphasis had been placed on the ability of tests to discriminate between normal subjects and ill-defined groups of patients with various types of acquired brain damage. Other more clinically relevant issues, such as the exact nature of a patient’s memory deficit and the implications for rehabilitation, had been ignored. They suggested that future work be directed by two considerations. First, techniques from experimental studies of human memory should be incorporated into the clinical field so that clinical assessments could provide information on the location of the deficit within the memory process (encoding, retention, or retrieval). Second, the relevance of memory assessment of rehabilitation would be increased if new memory tests could be developed to predict the patient’s concurrent or future learning ability in everyday life.

Speed selective communication processor

Abstract: PURPOSE: To secure communication quality by providing a function for detecting the line quality of a communication network and a function for rewriting a speed display part in a negotiation signal and converting a communication speed corresponding to the line quality. CONSTITUTION: A line quality measuring circuit 18 detects the line quality in the communication network during connecting operation and an encoding rate selecting circuit 19 calculates an encoding rate required to obtain prescribed communication quality and the communication speed in the communication network when a signal is sent at the encoding rate according to detection information on the line quality. An encoding rate variation error correcting circuit 22 sets the encoding rate on receiving the calculation results and a binary signal terminating circuit 21 changes the signal transmission speed display area in a binary control signal so that the communication speed between terminals equals the communication speed in the communication network. Consequently, the communication of high quality between the terminals becomes possible and no buffer memory is required. COPYRIGHT: (C)1992,JPO&Japio

Encoding of videoconferencing signals using VDPCM

Abstract: The techniques of motion detection, interframe linear block prediction and vector quantization have been incorporated in this paper, in a scheme for encoding monochrome image sequences for videoconferencing application. Data transmission rate reduction is accomplished by identifying and processing only those regions that exhibit noticeable changes between successive frames, by estimating the magnitude of the change through linear block or vector prediction and quantizing the residual vectors through a vector quantizer. The motion detector uses a modified block matching algorithm to detect the moving blocks. Perceptually-based edge detectors are used to design vector quantizer (VQ) codebooks for different classes of image blocks to achieve better visual quality. Encoding rates under 60 kbps are achieved with acceptable visual quality at nominal computational complexity.

Three-dimensional orthogonal transformation encoding system for moving image signal

Abstract: PURPOSE:To highly efficiently compress picture data by supplying a read signal line to the prescribed terminal of the memory of a memory device according to the read sequence of a moving image signal. CONSTITUTION:In a memory read control circuit 13, by the signal of a decision result sent from a deciding circuit 20 of the picture content change between screens, to which part of continuous screens in an N number on a time base the signal of the decision result corresponds is checked. Further, read control signals are supplied to prescribed terminals 3 to 10 in the memories of the memory device in the N number through a line 22 according to the pattern of the read sequence of the moving image signal from the memories of the memory device in the N number fixed beforehand corresponding to the part. Thus, the deterioration of a picture quality is reduced, the discontinuous motion is not generated in the moving image, and the picture data can be highly efficiently compressed.

Moving image encoding processing system

Abstract: PURPOSE:To execute the high efficiency encoding which is suitable for accumulation media by executing the inter-frame encoding using a motion compensation difference in the reverse direction, as well by only adding some data to motion compensation difference inter-frame encoding data used for the forward encoding. CONSTITUTION:This system is constituted of a buffer memory 1, a vector processing part 2, a decoding part 3, a frame buffer 4, a decoding part 5, a block dividing part 6, a block moving part 7, a frame buffer 8, a block gap calculating part 9, an in-gap picture element predicting part 10, a difference device 11, an encoding part 12, and an accumulation processing part 13. In this state, a dynamic vector used for the motion compensation in the forward direction is reversed, and by using this vector as the dynamic vector, the inter- frame encoding using a motion compensation difference in the reverse direction is executed. In such a way, the high efficiency encoding of a moving image for accumulation media can be executed.

Generalised ENCAP-4N digital modulation technique

Abstract: A class of encoding schemes for generating offset-QPSK-type signals is introduced. Each scheme supposes the specification of 4N functions directly related to the eye-diagrams of the in-phase and quadrature signal components. The digital implementation of the encoding rules, by table look-up techniques, is straightforward, and leads to a flexible and cost-effective transmitter that allows high spectral efficiencies to be achieved without any selective linear filtering. The main features of the proposed modulation format are described, namely its connections with well known modulation formats. It is shown that the new signal design method provides a unified representation of all OQPSK-type (MSK-type) signals recommended so far (IJF-OQPSK, TFM, GMSK, XPSK, 12PM3X, etc.), and allows the generation of a large variety of new signals. The conditions under which a linear filter and a nonlinear amplifier preserve the proposed format are discussed. It is also pointed out that, in most practical situations, the encoding rules can be modified so as to allow the transmission of a nonlinearly amplified signal of the proposed class, having a variable envelope.

Deterministic simulation of PRAMs on hypercube networks without look-up tables

Abstract: A deterministic simulation of an N processor EREW PRAM on an N processor hypercube network is given, where the size of the PRAM's shared memory is significantly larger than N. Although the simulation is asymptotically less efficient and uses smaller shared memory than previous deterministic simulations, it is more practical in that it is based on an available interconnection network, its memory organization scheme (MOS) is easily constructible, and it does not require that large look-up tables encoding the entire MOS be stored in each processor. >

Memory card and signal recording device or signal reproducing device to use memory card

Abstract: PURPOSE:To attain the recording and keeping of a lot of data signals with a memory card by compressing and storing an input data signal. CONSTITUTION:When primary color picture signals R, G and B of 8 bits are respectively inputted to 3-R, 3-G and 3-B of an encoding and compressing circuit 3 to a memory card 1, the signals are compressed and converted to the code data of 2 bits (R), 4 bits (G) and 2 bits (B) in the circuit 3. The code data are respectively stored to 4-R, 4-G and 4-B of a picture storing memory 4. The respective code data to be read from the memory 4 are inputted to decoding circuits 5-R, 5-G and 5-B and converted to the primary color picture signals R, G and B to be same as the time of the input. After that, the signals are inputted to a signal reproducing device. Thus, even when the picture storing memory has small capacity, much data can be stored.

Background predictive frame generating method

Abstract: PURPOSE:To shorten a selection time for selecting a motion compensation predictive encoding system or background predictive encoding system by generating the prediction error at the time of the background predictive encoding system by using one background predictive frame memory. CONSTITUTION:The above method is equipped with frame memories 1 and 2, the background predictive frame memory 3, a comparator 4, and a control part 6. Here, real background information on the shadow of a moving body is written on the background predictive frame memory 3, so the prediction error of the same level with the background predictive encoding system which uses plural background predictive frame memories can be realized by the constitution where the background predictive frame memory 31 is mounted. Thus, the prediction error at the time of the background predictive encoding system can be reduced by the one background predictive frame memory 3, so the selection time for selecting the motion compensation predictive encoding system or background predictive encoding system can be shortened.

Decoding system for 2-dimensional successively encoded data

Abstract: PURPOSE:To attain high speed processing by using a detection section so as to detect succeeding linear encoding code and applying decoding processing to a succeeding 2-dimension successive encoding data with other decoding section while the 2-dimension successive encoding data after the linear encoding code is subject to decoding processing by one decoding section CONSTITUTION:When a detection section 22 detects a linear encoding code MH among 2-dimension successive encoding data, the 2-dimension successive encoding data after the detected linear encoding code MH is decoded by decoding sections 14, 15 The detection section 22 detects the succeeding linear encoding code MH during that time, the 2 dimension successive encoding data is subject to decoding processing in parallel by other decoding sections 15, 14 from the detected linear encoding code MH Thus, the 2-dimension successive encoding data such as MR encoding data is subject to decoding processing at a fast speed

Matrix Computations and Neural Associative Memories

Abstract: In view of recent interest and applications of Neural Associative Memories, it becomes increasingly important to evaluate their capacity limits and saturation effects. This paper presents a unified mathematical approach to evaluation of saturation/capacity for a large class of associative memories based upon matrix operations. This class includes, among others, Correlation Matrix Memory, Higher Order Associative Memory, Generalized Inverse Memory and Hamming net. The general model is based on Linear Algebra and is applicable to both binary and continuous-valued memories, and also includes auto-associative, hetero-associative and classification modes of operation. It is argued and demonstrated that the well-understood Linear Algebra formalism can be effectively applied to evaluate saturation/capacity limits, scaling properties and various input/output encoding schemes, as well as to compare different supervised learning (memory construction) techniques. As a practical application of our approach, we present detailed comparative analysis of the Outer Product Learning and the Generalized Inverse Memory construction rules for the auto-associative memory and the unary classification modes of operation.

The storage and recall of auditory memory

Abstract: The architecture of the auditory memory is investigated. The auditory information is assumed to be represented by f−t patterns. With the help of a psycho-physical experiment it is demonstrated that the storage of these patterns is highly folded in the sense that a long signal is broken into many short stretches before being stored in the memory. Recognition takes place by correlating newly heard input in the short term memory to information previously stored in the long term memory. We show that this correlation is performed after the input is accumulated and held statically in the short term memory.

Data encoding system for memory card

Abstract: PURPOSE:To obtain a flexible system constitution which is set for the encoding method at each time by setting encoding data in a memory card file driver on an OS. CONSTITUTION:A register 22 where conversion data of encoding is held and a gate 21 which inverts the polarity of an address line for the purpose of encoding are provided to constitute hardware 2. Software 3 is provided with the memory card file driver which can sets encoding data on the file managing function on the operation system (OS), and hardware 2 and software 3 are combined to constitute an encoding system. Consequently, encoding data is set in the file on the OS to freely change and set mapping for encoding of data. Thus, the flexible data encoding system for memory card is obtained.

Test equipment for muse receiver

Abstract: PURPOSE:To facilitate the shipping test by detecting a fault of a digital signal processing circuit or the like and its faulty location including a memory based on the result of comparison between a test signal via a digital signal processing circuit or the like including the memory and a test signal inputted just after A/D conversion. CONSTITUTION:A prescribed digital test signal is inputted to a digital signal processing circuit including a memory 5 of a MUSE decoder provided to a MUSE (multiple subNyquist sampling encoding) receiver, and a test signal via the MUSE decoder and the test signal inputted to the digital signal processing circuit are compared with a control section 17 of a test equipment T. Then whether or not the memory 6 of the MUSE decoder and circuits 6-8 for still area interpolation and moving area interpolation or the like have a fault is discriminated depending on the result of comparison. Moreover, a display device displays corresponding to the fault location. Thus, a picture of the monitor has only to be observed visually at the shipping of the MUSE receiver provided with the monitor and the test at shipping is simplified.

Multilevel picture data compressor

Abstract: PURPOSE:To process a picture with high data compressibility by sharing one picture memory for bit plane data storage and for other data storage. CONSTITUTION:The picture memory is shared for the storage of bit plane data and for the storage of the other data. In the case of picture processing, the picture memory is equipped with idle capacity to store one page of the bit plane data, the picture is read for the unit of one page and obtained multilevel picture data are converted to the bit plane data and temporarily stored in the picture memory. Then, data compression is executed to the bit plane data for the unit of the page by a two-dimensional encoding system. When there is no idle capacity for one page in the picture memory, the picture is read for the unit of one line and the obtained multilevel picture data are converted to the bit plane data and temporarily stored in the picture memory. Then, the bit plane data for the unit of the line are data-compressed by a one- dimensional encoding system. Thus, the read picture can be processed with the high data compressibility.