Resources & Downloads

Patent No./Name

Summary

6,420,279

Methods of Using Atomic Layer Deposition to Deposit a High Dielectric Constant Material on a Substrate

Many issues arise as dimensions of semiconductor devices shrink to 2 nm.  For example, in order to maintain the same capacitance between gate and channel regions, the thickness of the SiO₂ layer must also decrease.  Unfortunately, the dielectric constant of SiO₂ is typically 4, making thin layers of SiO₂ susceptible to tunneling with the use of high power devices common today.  This problem has been anticipated for some time and many solutions have been proposed.  For example, the efforts investigating TiO₂ and Ta₂O₅ have failed, as interfacial SiO₂ and high temperature annealing create insurmountable problems. 

Sharp has developed a method applying high dielectric materials (including hafnium oxide, zirconium oxide and nanolaminates of hafnium oxide and zirconium oxide) to hydrogen passivated surfaces.  The method has successfully applied a thin film high k dielectric material capable of supporting dimensions down to .7 nm.

This patented method uses atomic layer deposition with nitrogen based precursors to form ZrO₂ or HfO₂ films (and their nanolaminates) of any desired thickness.

Advantages:

• Works well with hydrogen passivated surfaces
• No high temperature annealing is required
• Substrate is kept below 200^o C
• No chlorine or carbon impurities

This technology is supported by US patent 6,420,279 and related to CN1396638 (A); JP2003068732 (A); and TW577130 (B).

6,303,502

MOCVD Metal Oxide for One Transistor Memory

Ferroelectric memory devices could outperform other semiconductor memory technologies by offering faster write times and greater endurance.

Though many attempts have been made to create a ferromagnetic, single transistor memory integrated circuit, such as MFMOS (Metal-Ferroelectric-Metal-Oxide-Semiconductor), MFOS (Metal-Ferroelectric-Oxide-Semiconductor), or MFMS, none have been successful.  The major issue has been depositing the correct amount of ferroelectric material.

This method, with a combination of spin and metal organic chemical vapor deposition (MOCVD) applications, elegantly resolves this issue to allow for the creation of these advanced devices.  Although the method does require the use of MOCVD, all other steps in the process rely on traditional semiconductor manufacturing methods.

This technology is supported by US patent 6,303,502 and related to EP1162657 (A2); JP2002026288 (A); and TW510088 (B).

6,429,952

Rather than develop proprietary applications to control a networked printer, scanner or image acquisition device, consider leveraging the browser already on your clients desktop as your scanner controller and viewer.

Although there have been some standards efforts for image acquisition devices, these have stopped at API layer (e.g. TWAIN and ISIS) and still require a separate driver to be installed on the PC for each new device. Some scanner interfaces have simplified the data transfer by leveraging EMail and shared directories, but they do not offer any control of the scanning hardware.  To provide the complete control and data transfer, scanners use proprietary applications and drivers.

This patent claims a novel method of controlling networked based scanners and copiers through a browser. Use of this method substantially simplifies the development necessary for controlling networked printers by reducing the development time for the PC based device control application.  In addition, it provides the end user a familiar interface (browser) capturing images directly or via network.

The implementation requires adding a web server capability to the imaging device, which will not add considerable additional resource requirements.

Supported by USPN 6,429,952 and working reference code, this solution is mature and ready for implementation. All controls typically found in these applications have been implemented, including cropping, rotating, resolution, and image enhancement. A demonstration CD is available upon request.

Benefits:

• Significant user interface simplification
• Network compatible
• Operating system independent

This technology is supported by 1 US patent: 6,429,952.

6,603,799

Method for detecting the hopping sequence of an interfering wireless system

The presence of multiple wireless devices using ISM band, including HomeRF Bluetooth, and various cordless phones, often creates suboptimal network bandwidth for HomeRF or any hopping protocol using the same frequency band.  This typically is a problem for the wireless networks transmitting time sensitive information (e.g. voice) where collisions cause noticeable delays. This solution proposes an elegant resolution to this problem by anticipating the frequencies of likely collisions and minimizing their impact by sending only minimal data on suspect frequencies

Wireless protocols that conform to the standard use one of limited number of hopping sequences. The patent covers the identification of which sequence is being used by evaluating the time interval between several frequencies.  Research has shown that a hopping sequence can be uniquely identified by making two measurements of the time interval between three frequencies.

Although originally designed with HomeRF and Bluetooth in mind the system will work with any two hopping protocols sharing the same frequencies.

Benefits:

• Increased actual bandwidth in the presence of multiple wireless devices
• No knowledge of the content of the system is necessary
• Reduction in packet collisions
• No additional hardware requirements

This technology is supported by 1 US patent: 6,603,799.

5,839,000

Automatic zoom magnification control using detection of eyelid condition

Automatic focusing and aperture cameras are now well established.  However, zooming or changing the focal length is still a manual process.  In situations where your hands are occupied with other activities, it would be convenient to have a hands-off approach to controlling magnification or zoom.  While some hands off approaches exist, this is the only solution that allows the viewer to maintain his gaze while changing the zoom.

Sharp has developed a solution that measures characteristics of viewer to determine the viewers intent.  The patent covers two main implementations:

1)    A calculation of the amount of the inside corner of the eye (conjunctiva) that is visible.

2)    A bioelectric measurement of muscle contraction.

This solution has application in:

• Still cameras
• Video cameras
• Computer or other video displays
• Medical optical devices
• Surveillance or military devices

This technology is supported by 1 US patent: 5,839,000.

6,044,282

Dual clock power conservation system and method for timing synchronous communications

In many systems, high and low speed clocks are used to perform various functions. In wireless devices it can be optimal to reduce the use of high-speed clock to save battery life.  However, these high-speed clocks are used to synchronize time for outside event, such as when to listen for messages from a base station, so turning the clock off can cause the device to fail.

This patent discloses a method to provide the timing resolution of the high-speed clock while utilizing to a low speed clock to time long periods of inactivity.  It includes methods to train both clocks to achieve the accuracy needed to be synchronized to outside events. This technique would be especially useful for cell phones.  Cell phones are synchronized with their base stations and only receive messages during a short window of time. Use of this solution would allow the phone to conserve power by de-energizing the high resolution clock during the periods of inactivity between messages.

While this technology was developed with a cellular network in mind, it covers any communication network and would be useful to extend the battery life of any wireless device that relies on power intensive clocks to maintain synchronized time.

Keywords:  communications networks, wireless, cellular, battery optimization, high resolution clocks, variable computing

This technology is supported by 1 US patent: 6,044,282.

6,078,307

Method for increasing luminance resolution of color panel display systems

Sharpness is one of the primary measures of projection televisions.  By using an untapped physiological characteristic of the eye, it may be possible to increase sharpness by 100%.

Great lengths are made to achieve a 20% improvement in sharpness.  Typically this is done through advanced image processing, however this often leads to the creation of other artifacts.  The best solution would be to increase the resolution by generating more samples from the image, not by creating them from whole cloth with signal processing.

This patented technique achieves this result by offsetting the color pixels by one-half to double the resolution in both the horizontal and vertical directions. The physiological source of this effect is color sensed by comparing the frequencies that converge on a single cone cell in a subtractive manner. Luminance, however, is essentially additive.  While this technique does sacrifice color resolution, the subtractive technique of the human eye makes us 1/2 to 1/4 as sensitive to color as luminance, so the overall picture is viewed as better. This physiological fact is the essence of this solution.

With the increased perceived sharpness, the image will have better texture rendition to provide a more realistic image, without the common downsides of other sharpness enhancing techniques (luminance aliasing, edge halos or ringing).

Benefits:

• Increased sharpness with the same image density
• No signal processing artifacts

This technology is supported by 1 US patent: 6,078,307.

Keywords:  image processing, sharpness, display, resolution

6,351,523

Method and apparatus for management of Email originated by thin client devices

Even with the prevalence of wireless email devices, such as the BlackBerry¨, the most effective method for conveying sensitive information is voice.  Voicemail provides the user the ability to add inflection and tone to enhance delivery and responsiveness. Indeed, voicemail is often given a higher priority than email.

This patent covers a method of allowing the user to create and automatically deliver voicemail in response to an email message.  The technique requires the use of 1) a handheld voice capable email device, 2) a method of identifying the emailed voice data file and converting it to a voice stream, and 3) an email-to-voicemail gateway capable placing the stream in a voicemail box.

While this method involves the integration of several systems (PBX, mail system, etc.), as IT becomes more streamlined these products are commonly being sold as a package, making this implementation easily attainable.

This technology is supported by US patent 6,351,523 and related to EP1039733 (A2) and JP2000278439 (A).

Keywords:  email, voicemail, thin client

6,556,825

Method and apparatus for automatic adaptation of communications systems to regional spectrum variations

Bluetooth and other wireless protocols using the industrial, scientific, and medical (ISM) band often must conform to slightly different geographic restrictions.  For example, while the band in the Americas and most of Europe is 2.400-2.4835 GHz, in France the band stops at 2.4465 GHz. In Japan there are similar differences.  These restrictions are in place to avoid interference with legacy allocations and are often severely enforced.

Today, devices handle these geographic differences through user input on installation of the device.  While sufficient for most applications, many devices (especially Bluetooth enabled cell phones) often change geography.

This patent covers a method of using a long-range wireless protocol (such as cellular or GPS) to identify a location and then automatically adjust the frequency band of the ISM device.

Implementation of a system would be straightforward as geographic information is readily available from within the cellular network or it could be calculated using GPS.   

Key words: Wireless devices, Cell Phone, PDA, Bluetooth, ISM, HomeRF, regulations, location

This technology is supported by 1 US patent 6,556,825.

6,085,104

Pilot aided, time-varying finite impulse response, adaptive channel matching receiving system and method

In a communications network, often several signals are received.  Often these are reflected off buildings or other objects, or they may derive from separate base stations.  It is advantageous to accurately combine these signals to create a higher signal to noise ratio from the sum than can be achieved from any of the signals alone.

While the concept of multipath is well understood and examined, Sharp has designed a solution that achieves an optimal increase of signal while efficiently using both hardware and power.

This solution discloses a method of reducing the number of traffic rake sections and efficiently summing the traffic channel data.  While the technique could cover all methods for performing this analysis in parallel, a preferred embodiment includes the use of a time variable finite impulse response (FIR) filter.  While FIR filters have been employed in the past, this is the first implementation to leverage the timing information in the pilot signal to analyze multiple signal paths and assign unique delays to these paths.

Benefits:

• Higher signal to noise ratio (better reception)
• Efficient use of chip real estate for RAKE receivers
• Decreased power consumption

This technology is supported by 1 US patent:  6,085,104.

6,208,632

System and method for CDMA channel estimation

In a communications network, several signals are often received.  These are reflected off buildings or other objects, or they may derive from separate base stations.  It is advantageous to accurately combine these signals to create a higher signal to noise ratio from the sum than can be achieved from any of the signals alone.

While the concept of multipath is well understood and examined, Sharp has designed a solution that achieves an increase of signal while efficiently using both hardware and power.  Although it could be used with any wireless technology, its method of calculating timing information makes it especially useful for wideband CMDA technologies.

This method uses the perch channel to communicate back from the receiver to the base stations to synchronize their transmissions and increase signal to noise ratios.

Benefits:

• Higher signal to noise ratio (better reception)

This solution is protected by USPN# 6,208,632 and related to DE69903926D (D1); DE69903926T (T2); EP1057279 (B1); JP3483852B2; and WO9939452 (A1).

6,259,828

Sprite-based video coding system with automatic segmentation integrated into coding and sprite building processes

When compressing video data for delivery over a limited bandwidth connection, it is often necessary to create a sprite or mosaic.  These areas are typically only sent once and then recreated on the far end when needed.  While highly efficient with their use of bandwidth, it is necessary to define the boundaries of the sprite at any given instant.  In the absence of blue screen or non-real time production routines, there is no system to automatically segment the image for developing sprites.

The Sharp patented method solves this problem.  Specifically designed for MPEG 4 it can be easily integrated into any video object-based coding framework.  It is especially suitable for applications where camera view may change frequently, such as video conferencing with multiple cameras, or a talk show captured with more than one camera.  In these applications, multiple sprites are built and used as needed.  Further this method anticipates the use of zooming and uses a high-resolution background sprite, which is used to generate lower resolution distance angles.

Benefits:

• Does not require a-priori knowledge of scene segmentation
• Automatically compensates for zooming
• Adaptable for very low bit rates

This solution is protected by USPN# 6,259,828 and related to US6205260 and JP2001507541T, WO9829834, and EP1042736.

6,441,867

Bit-depth extension of digital displays using noise

In displays having limited bit-depth, contour artifacts appear in smooth gradient regions.  For example, an image of the sky will show visible contour lines in the smooth gradient blue regions, somewhat like geographic contour lines on a topographical map. While techniques for reduction of these artifacts exist, these 1) are typically computationally complex, or 2) reduce the spatial resolution of the image.

Sharp has developed an elegant solution that extends bit-depth by using the static noise of the display. The static noise of the display is measured and then subtracted from continuous tone image data. The noise-compensated image data is then quantized and displayed. The use of the noise breaks up contouring artifacts and leaves little visible noise in the image or images displayed.

The invention also covers using the inherent noise in the human visual system (HVS) instead of (or in combination with) the static display noise as the subtraction or pseudo-random noise.

Benefits:

• Computationally simple
• Easy to implement
• Robust

This solution is protected by USPN# 6,441,867 and related to EP1094420; EP1322113; JP2001175236; and JP2003263143.

6,353,686 and 6,411,740

Method for non-uniform quantization in a resolution hierarchy by transmission of break points of a nonlinearity

This work takes advantage of the masking properties of the human visual system.  This approach is particularly useful for images that are not as affected by other methods (amplitude transformation and spatial frequency discrepancies). In cases where the frequency sensitivity of the visual system does not change much over the  spatial frequencies present in an image, this method is especially helpful (e.g. low resolution displays or close viewing distances).

Previous work in this area involved quantization schemes using local image variance. Discrete cosine transform (DCT) blocks were used unsuccessfully, as they do not map well to the HVS.  SharpÕs method takes advantage of compression techniques that decompose an image into frequency bands analogous to the visual system frequency channels, resulting in compression in excess of the 5-25% compression (depending on content) possible with DCT routines.

Benefits:

• Improves perceived image quality in applications where frequency-weighting techniques are ineffective
• Useful for low display pixel resolution and close viewing distances
• Image quality is uniform over a variety of image content
• Adaptive and not computationally intensive

6,411,740 offers similar coverage and benefits but instead of requiring a set of breakpoints it uses one of several known curves.

This solution is protected by USPN# 6,353,686 and related toUS 6,411,740 and JP2000152240 offer additional protection.

6,205,168

Sequential detection system and method with adaptive bias

Sequential detection techniques are very effective at selecting the correct code used to spread the transmission; however it relies on the bias value remaining relatively constant.  In a cell-phone environment, the bias is constantly changing with the position of the receiver, making this method less optimal.

Sharp has developed a method to determine an unknown spreading code in an environment where the signal to noise ratio of the received signal varies. A communication system, such as W-CDMA, provides a periodic reference signal that the receiver can demodulate without the spreading code. The reference signal is used to estimate and constantly update the signal to noise calculation. From this estimate a bias is updated for use in the sequential detection of the spreading code.

The patent also provides a method to update the CDMA sequential detector bias value.

Benefits

• Simplifies determining the long code being used by a base station.
• Calculates bias for use in the sequential detection of CDMA type signals in ever-varying signal to noise ratio environments.
• Determines spreading code through sequential detection with the use of the varying bias value.

This solution is protected by USPN# 6,205,168 and related to 6,208,632 System and method for CDMA channel estimationÓ EP1001550 (A2); EP1001550 (A3) and JP2000152312 (A) offer additional protection.

6,468,901

Nickel silicide including iridium for use in ultra-shallow junctions with high thermal stability and method of manufacturing the same

Currently Titanium silicide and cobalt silicide are used in silicide manufacturing processes to produce metal oxide semiconductor (MOS) transistors. Both have disadvantages.  When the polysilicon line width is reduced, Titanium silicide is difficult to transform to a low resistivity C54 phase. Cobalt silicide has a high silicon (Si) consumption rate to form cobalt disilicide, limiting its use directly on an ultra-shallow source/drain area. Moreover, a reduction in the junction depth requires a very flat interface between the silicide layer and the silicon active layer.

It would be advantageous to use Ni-silicide as it has low resitivity, lower Si consumption (50% of Co), and demonstrated low junction leakage.  However, Ni-silicide has previously not been stable in temperatures over 650¡C, which are necessary for further processing.

This patented technique developed by Sharp solves the drawbacks of Ni by using iridium in the CMOS process.  The method stabilizes the Ni-silicide film to temperatures in excess of 800¡C.

Benefits:

• Reduced Si consumption during processing
• Reduced junction leakage
• Useful in ultra shallow junctions of 400 Angstroms

This solution is protected by USPN# 6,468,901and related to CN1384531 (A); JP2002367929 (A); and TW543167 (B).

6,496,051

Output sense amplifier for a multibit memory cell

The unsung workhorse of computers is memory. As flash memory is reaching its theoretical density limits, new technologies are envisioned to store more bits in less space.  One such solution uses a material, such as perovskite, which can store multiple bits of data in each memory cell through a change of resistance (each cell can have more than the standard 2 values). 

This Sharp developed patented method provides a solution to read a resistive bit of resistive information and convert it to binary data.  This particular method is unique in that this analog result may be converted to a large number of binary states, limited only by the accuracy of the resistance measurement.

Benefits:

• Numerous memory states (>2)
• Computationally simple

This solution is protected by USPN# 6,496,051 and related to EP1291881 (A2); JP2003178575 (A); and TW569241 (B).

6,506,637

Method to form thermally stable nickel germanosilicide on SiGe

Si--Ge based integrated circuit devices have created the need for stable, low thermal budget, low resistivity metal-silicide contacts to SiGe alloys. Most Metal/SiGe systems show poor phase segregation. In particular, during the thermal anneal in a Ti/SiGe system, excess germanium combines with silicon and nucleates as SiGe precipitates along the grain boundaries. In the Co/SiGe system, excess Ge diffuses to the grain boundary to form Ge-enriched clusters.  These byproducts of fabrication make these systems inadequate for small feature sized devices unless a buffer is used.

Nickel has shown promise with increased drain currents, however its thermal stability has previously been weak over 600¡ C.

Sharp has developed a process involving Cobalt or Iridium that stabilizes the Nickel keeping its sheet resistance low even after 800¡C anneals.

Benefits:

• Low sheet resistance
• Thermally stable to 800¡ C
• 10% increase of the saturated drain current in mm .15 pMOSFET
• Beneficial in the fabrication of ultra-shallow junctions (<40nm)

This solution is protected by USPN# 6,506,637 and related to JP2002289558 (A); TW540091 (B); and US6627919.

6,509,268

Thermal densification in the early stages of copper MOCVD for depositing high quality Cu films with good adhesion and trench filling characteristics

As device dimensions continue to shrink, the need increases for lower resistance materials for the interconnects.  Copper is the obvious choice, however its deposition has proven to be challenging.  The standard methods of deposition, physical vapor deposition (PVD) and electrochemical deposition (ECD), have their drawbacks. PVD has poor step coverage, and ECD requires the initial deposition of a copper seed layer, which of course, must be deposited by PVD or metal organic chemical vapor deposition (MOCVD). 

While MOCVD is more suited to deposition of copper in deep sub-micron trench/via copper fillings, the known MOCVD processes produce poor adhesion between the copper layer and a metal barrier layer, such as a nitride layer.

Sharp has developed a patented technique to adequately adhere copper to nitride layers using a series of simple replicable steps.

Benefits:

• Superior adhesion of Cu to Si with nitride layer
• Superior trench filling
• No interfacial carbon fluorine impurities

This solution is protected by USPN# 6,509,268 and related to JP2003163174 (A)

6,289,061

Wideband frequency tracking system and method

In many digital communication systems the signal is transmitted with the aid of a carrier signal with a certain frequency.  To receive this signal, a receiver must ensure that its local oscillator frequency (LOF) is operating very close to that of the transmitters. The control of the local oscillator is maintained through a mechanism called automatic frequency control (AFC). 

In wideband signal transmissions, the signal is received along a number of distinct paths, each of which might be used to set the LOF.  However, receiver clock inaccuracies, transmitter frequency drift, and Doppler effects introduce error and require computationally intensive calculations to adjust carrier frequency of each multipath signal.

To simplify this multipath calculation, Sharp has developed a wideband receiver AFC system that combines multipath or diverse antenna signals before the carrier frequency error is determined.  The frequency error is calculated after the multipath signals are summed together, requiring only a single frequency detector circuit. This system has similar accuracy to one that combines the independently calculated frequency errors, but with fewer computation steps.

The patent also describes a method for combining multipath signals to calculate an average AFC frequency error.

Benefits:

• Simplifies the AFC function while maintaining accuracy
• Cuts the number of parts
• Decreases the receiver's power consumption
• Reduces the number of arithmetic operations

This solution is protected by USPN# 6,289,061 and related to 6,208,632 System and method for CDMA channel estimation and 6,085,104 Pilot aided, time-varying finite impulse response, adaptive channel matching receiving system and method.  EP0989687 (A2); EP0989687 (A3); and JP2000106535 (A) offer some additional protection.