DC freeing of the write waveform is an important technique in digital magnetic recording. The higher the recording density is, the stronger the effect of high frequency noise becomes. We use the DC-freeing technique of write equalization to boost high frequency on the write side, and read equalization with which we can reduce high-frequency noise with little degradation of the eye opening, is adopted on the read side. In addition, we adopt Viterbi decoding as a decoding method for further improvement. In this paper, we report the synthesis of this write and read equalization method. Computer simulation shows that our system has an excellent error rate performance compared with the well-known combination of NRZ recording code and integrated detection.

View full abstract

A new block-adaptive bit assignment technique is proposed for the efficient compression of digital images. In picture coding technology for storage media, the data rate, which depends on picture contents, must be controlled. In this method, a proper parameter is selected that gives changes in the compressed data rate and controls it accurately. The method has been implemented with a relatively simple hardware by quantizing the block information which represents the activity of the block. Application to A²PS-DCT (Adaptive Assignment Coding with Patterns and Step-Sizes for DCT) which is a compression technique for digital VTRs is also described. Computer simulations show that the reconstructed picture quality is acceptable.

View full abstract

In many digital VTR systems, the playback signal is equalized to the Nyquist criterion and then detected using integrated detection. We digitized the playback signal after equalization at four times the synchronous clock frequency. Using computer analysis, we then investigated the error occurring waveform and the number of errors. From this investigation, we developed a new signal detection method that uses the second difference between each bit. The threshold level is controlled with the value of the second difference to increase the signal-noise ratio. Detection erros near the threshold level in integrated detection are significantly reduced. Some new parameters are required for this method, but they can be automatically adjusted using the zero forcing algorithm.

View full abstract

In variable speed playback of a composite video tape recorder (VTR), the reproduced video must be converted to a standard TV signal because fields in the reproduced video are skipped or repeated. For field conversion processing, a Y/C separation circuit and a line-interpolation circuit are used. In a conventional processor, the picture quality deteriorates due to vertical displacement of the image and lower resolution. Our new motion-adaptive field-conversion circuit that uses inter-field signals can remove these deficiencies. The focus of this paper is on a motion-adaptive line-interpolation technique. It also describes a new line-interpolation filter with a motion detection function that compares the amplitude of inter-field signals. Based on these techniques, three kinds of gate arrays were developed. The arrays were employed in D-2 VTR for NTSC signals. This VTR exhibits none of the problems mentioned above and has picture quality comparable to that at standard play-back speed.

View full abstract

We have obtained an SNR evaluation equation from experimental results for some wideband HD-VCRs that use an FM recording method. This equation shows a good agreement between experimental results and SNR calculated with recording parameters. We have attempted to improve SNR by using hyperbolic tangent functions to emphasize the dynamic linear-phase. SNR improvement of about 6 dB has been attained, and there is no notable distortion in the reproduced waveforms. This method is also effective in eliminating inverted white peaks. With respect to the RF system, we show that the head-oscillation compensation method, which can suppress waveform distortions, is also quite effective in preventing inverted white peaks and that SNR can be improved by about 2 dB by employing linearly descending dynamic equalization.

View full abstract

Compact, light weight Hi-Vision recorders with low power consumption are very much desired. Our investigation of design concepts shows that a tape transport system with a small scanner, non-segmented format and high recording density would be suitable for reducing size, weight and power consumption.
With respect to high recording density where a minimum recorded wavelength is less than 0.4, um, metal evaporated (ME) tapes with a high performance head have the potential to achieve such a short recorded wavelength. We have studied these tapes from the viewpoint of both magnetic recording characteristics and durability.
Based on these studies, we have built an experimental Hi-Vision recorder. The diameter of the scanner in the transport system is 76 mm. This machine also has baseband bandwidth luminance of 20 MHz and baseband bandwidth chrominance of 7 MHz, a minimum recorded wavelength of 0.37 gm which is the shortest ever reported, and S/N luminance of 41 dB and S/N chrominance of 45 dB.

View full abstract

Editing interchangeability is one of the most important factors in narrow-track digital VTRs for broadcast use. The worst degradation of the recording foot print occurs during insert editing. This work investigates all factors that can lead to format errors. Several new technologies, such as a head pair configuration, a distinctive audio sector arrangement, using a guard band only at the editing point and so on, are proposed as ways to reduce the influences of these errors. Moreover, a new definition for the tolerance zones of helical tracks is recommended. As an example of an application, the interchangeability of the D-3 format is identified.

View full abstract

We incorporate a magnetization model into the finite element method to calculate the recording characteristics for a VTR signal. The fact that calculated results agree with measured ones shows that this method is effective in the analysis of VTR signal recording. In addition we use the Fourier transfer to analyze magnetization distribution. This analysis clearly shows that the appearance of a spurious signal is caused by a nonlinear magnetization process. We also show that a video luminance signal can be recorded on a medium surface and that a video chrominance signal can be recorded on the middle layer of a medium. Therefore, a double-layered medium having a high coercive force in the surface layer and a low coercive force in the underlayer, is effective for the recording of a signal with a wide frequency region. Moreover, a double-layered medium with a perpendicularly oriented surface is excellent for high density recording.

View full abstract

We use the Fourier Transform Grid Method as the basis for a new approach to analyzing the displacement distribution of the track on which signals are recorded in magnetic recording systems. It is important to make the track width narrower for high density recording on magnetic tape. However, as the track becomes narrower, track linearity must be improved. To improve track linearity, it is necessary to clarify the displacement distribution of the track accurately. Our new measuring method can obtain a two-dimensional distribution of the displacement accurately and can clarify the relation between the displacement distribution and the factors that deteriorate track linearity.

View full abstract

This paper describes a tape driving system for magnetic recording or reproducing apparatus. First, we clarified the parameters that influence tape driving force. Experimental results verify that the parameters which has the greatest influence on tape driving force are the pressure force of the pinch roller (P), the loss torque of the pinch roller (Φ), and the coefficient of the friction between the tape and capstan shaft (μ). Next, we formulated each of these parameters. These formulas well explain the results of the experiment.

View full abstract

Factors affecting the threshold level of tape-originated erorrs in D2 and D3 systems were studied.
Dropouts that are too small to be perecived as such in conventional analog systems turned out to cause errors in very high density digital recording systems. Erorrs caused by these small dropouts appeared randomly and, in some cases, irreproducibly. These facts reveal that a considerable number of the random erorrs result from small dropouts rather than to random noise.

View full abstract

This paper describes a software servo system for a VCR. We developed a software servo microcomputer called the M37770M4A to improve servo performance. It has a 16-bit CPU, a timing-measurement-timer with a 16 MHz clock that is reset by a drum PG (Pulse Generator) signal, a 14-bit PWM and so on. The M37770M4A can make the programming of outputting pulses easy and can reduce quantize noise.
We applied this microcomputer to a camcorder and adapted a disturbance observer to the drum/capstan servo system to suppress fluctuations in the motors' rotations at low frequencies. Moreover, inserting a digital filter into the observer to compensate for the LPF characteristic reduced the increase in rotational fluctuation at high frequencies caused by model mismatching of observer.

View full abstract

In the area of magneto optical disk memory, we have focused our efforts on achieving high density recording, recording readout and high speed access. We have also been looking into the potential benefits of magneto optical disk memory. CONCURRENT ROM-RAM optical disk memory, which comprises phase pit information and magneto optical information, has very good potential because the memory capacity is ideally twice as large as that of conventional memories and both ROM and RAM are reproduced simultaneously by means of laser feedback method and so-called differential detection readout. We have confirmed the feasibility of CONCURRENT ROM-RAM through basic experiments. However, we have also found that the polarization noise generated by the edge of the phase pit is quite dominant.

View full abstract

An LD-compatible, write-once type video disk called LD-WO has been developed. An organic dye film was spin-coated on a 2P glass substrate. Subsequently, the Au reflective layer was formed by sputtering on the organic dye layer. Then spin-coating was again used to form the UV hardening resin layer. We achieved a long recording time (about 50 minutes per side) by using the constant linear velocity (CLV) mode (11.4 m/s). A carrier to noise ratio (CNR) of 62.5 dB at 8.1 MHz, was obtained for a duty-controlled (46%) recording when the groove width was 0.70μm.
The signal to noise ratio (SNR) for a NTSC direct FM signal recording was more than 43 dB.

View full abstract

A doubled disc recording density is used to produce a high definition videodisc with 120 minutes of playback time on both sides. A MUSE video signal and two kinds of PCM sounds are recorded. A narrower track pitch (from 1.6 μm to 1.1 μm) and a lower linear velocity (from 20 m/s to 14 m/s) is used to produce a higher recording density. Improved optical pick-up resolution is achieved by the employment of a visible laser diode with a wavelength of 670 nm and an objective lens with a numerical aperture value of 0.55. Signal parameter examination is used to increase the signal to noise ratio of playback signals.
These technologies applied an injection-molded disc with a fine playback image.

View full abstract