WO1993013611A1

WO1993013611A1 - Scanless tv transmission system

Info

Publication number: WO1993013611A1
Application number: PCT/US1992/010962
Authority: WO
Inventors: Elliot L. Gruenberg
Original assignee: Compfax Corporation
Priority date: 1991-12-23
Filing date: 1992-12-17
Publication date: 1993-07-08
Also published as: AU3327193A; HU9401888D0; JPH07502633A; BR9206980A; EP0619057A1; FI943010A0; HUT70716A; FI943010A; CA2117346A1

Abstract

A scanless TV transmission system is used for transmitting and storing graphic or other information requiring significantly less band width and memory than is currently required. A simultaneously available ensemble of information can be transmitted over a narrow band channel that is limited only by the rate of time change of the entire ensemble of transmitted bits, i.e. for television or motion video the limit is the persistence of vision. Reduction in band width reduces the power required for transmission. To attain these goals, a bit generator and operator, a switched invertor (using an amplifier, transmission loop and filter), a switch invertor, a lowpass filter, an amplifier, another transmission link and a filter are utilized.

Description

SCANLESS TV TRANSMISSION SYSTEM

Field of Invention

The present invention relates to apparatus and method for transmitting and storing graphic and other information in public and private facilities, which method requires significantly less use of bandwidth and memory of those facilities.

Description of the Prior Art

In Patent No. 5,029,210, issued in July of 1991, a cooperative communication system was described. This patent invention shows principles of operation of a communication system which have in part been used in the present invention.

Using those principles and additional ones, a new communications transmission system has been developed which greatly reduces the bandwidth requirements for transmitting television or other graphic material. Additionally, the method significantly reduces bandwidth needed for digital transmission used for any kind of information. Previous methods of TV transmission requires the material to be scanned by an electronic beam at high speed to cover a large number of pixels or illuminated spots in a raster-like field-of-view which might be called the scene. At the reception end, the received illumination values

_for the pixels are reproduced and located in corresponding pixel positions on the screen of a device such as a cathode ray tube.

In this method the pixels are transmitted sequentially, thus a scene containing 500 x 500 or 250,000 pixel locations must be

/ sent to the receiving location in a time interval of approximately 1/30th of a second which time interval is governed by the refresh time necessary for the persistence of vision. Such a method results in a transmission bandwidth requirement of 7 1/2 million hertz. However interlace methods are used presently to reduce the requirement to somewhat under 4 megahertz.

There is presently a demand for even higher definition than a 500 x 500 pixel raster, 1000 x 1000 or 1,000,000 pixel images are being contemplated in high definition TV (HDTV) .

The present transmission system eliminates the scanning and provides for a method of parallel transmission of information from the transmitted scene to the received screen bandwidth of the present system, that only requires in the order of 60 hertz.

It is therefore an objective of the invention to provide a transmission system requiring greatly reduced bandwidth yet allowing the transmission of high definition graphic material.

It is a further objective of this invention to provide a digital transmission of increased information transmission capacity operable on narrow and wide band transmission mediums.

It is an object of this invention to provide a digital transmission system for video and other data which requires drastically less power for transmission of the same information data rate.

It is an object of this invention to facilitate secure high data rate transmission.

-2- It is an object of this invention to reduce significantly the number and complexity of codes required to improve the rate of information transmission over narrow band facilities.

It is an object of this invention to improve the connectivity of networked computers and similar devices and terminals.

It is an object of this invention to facilitate two-way video transmission systems including picturephones.

It is an object of this invention to simplify transmission of three dimensional graphics.

It is also an object of this invention to store images and other blocks of data in substantially less memory space by efficiently encoding and retrieving such data speedily and completely.

g_-n._i._Ty <_f invention

In normal transmission of graphic material, the information is divided into elementary areas (pixels) and these areas are scanned by a beam at a rate determined by motion of objects in the scene, or by time of storage in the eye (flicker) and by the number of elements to be scanned. The receiver is synchronized to the scan rate so that the receiver places the pixel in the right location on the screen.

In such an arrangement the transmission rate is determined almost entirely by the requirement of placing the pixel in the right place and very little by the actual change of information in the scene. Since the plan of pixel location is known to the sender and the receiver, much savings can be made in transmission by sending a coded designation of location of pixel whenever a pixel is a "1" and send no such designation when the pixel is "0". (Grey Scales can be handled by a sequence of "bit" planes to be discussed separately but in the final analysis this amounts to an increase in the number of digital pixels) .

Prior methods of transmitting wide band data, such as TV, other than by the time division method just described, include assigning a separate frequency channel for each input (pixel) , or assigning a separate code per pixel. The receiver must employ a corresponding frequency channel for each pixel, the width of which must be wide enough to accommodate the possible time rate of change of the pixel information. Sufficient bandwidth must be provided for the transmission of any, or all, of the pixel information, or n(l/t) where t is the shortest time change of pixel information to be expected. Therefore, this method requires the same bandwidth as the sequentially transmitted scanned method presently in use.

If codes are used, the receiver codes ideally should be codes which provide correlation coefficient of one when correlated with the same transmit code and zero or a very low value when correlated with any other code after being convolved and integrated over the time period t. This method is described in "Handbook of Telemetry and Remote Control", pages 9-39 thru 9-52, (Gruenberg, McGraw Hill, 1967). Since n codes are required for n pixels, the bits must require a bandwidth of 1/nt. Thus, this method requires the same bandwidth as the sequential time division method.

This requirement can be reduced to 1/t by the present invention because these codes can be matched bit by bit by a -cooperative communication method in which each pixel transmit code is linked by feedback with all receive pixel codes, but ^"only the corresponding code will respond. The correct code will cause a constant transmission signal for the entire period t, because the bits at the receiver complement those of the transmitter for the full period t, whereas the wrong codes will not. This constant signal requires only a narrow transmission bandwidth, whereas the prior art coded modulation requires the full bandwidth of 1/nt to pass the individual bits of the code.

We have thus devised a cooperative communication system that will permit the transmission of any active "1" bits designated by the location code to the other party equipped with a decoder for each pixel. Transmission of codes from different pixels to geometrically corresponding pixels at the receiver is simultaneous and only the corresponding pixels respond. This transmission system makes use of elements of the cooperative communications system recently patented.

Thus for an n element picture, the transmission channel rate is r, rather than rn where r is the required rate of changing the scene sample. Thus, if r = 30 scenes per second, the conventional transmission rate is 30 x 1,000,000 = 30,000,000 pulses per sec. (one pulse per pixel position), whereas with the new method the transmission channel rate is 30 pulses per sec. (one pulse per scene) , an improvement of 1.000.000.

Each pixel, which uses a common transmission channel with the other pixels is connected to a communication system shown in Fig. 1, which includes new sending and receiving apparatus. This apparatus contains the orthogonal coding for each pixel and special modulating devices called Switched Inverters. The transmission of energy is dependent on the status of the inverters at each terminal. This apparatus allows each receiving pixel to respond only to a transmitting pixel with a corresponding code. If these n codes of length n were to be transmitted simultaneously, as by present means, a common channel bandwidth of n x (1/t) is required, where t is the time required by the change of information in the scene, resulting in no bandwidth savings. However, most significantly, the present invention provides a method which reduces the bandwidth requirement to 1/t. Instead of transmitting the n bit codes of length n to the receiver the codes are used to operate phase inverting switches in the transmitting and receiving apparatus. Each transmitter bit location (pixel) is linked to a corresponding bit location in the transmitter via a common pair of transmission paths allowing energy to pass from the transmitting point to the receiving location and from the receiving point back to the transmitting location. This energy is amplified and recirculated as taught in Patent No. 5,029,210, "Cooperative Communication System" so that energy will build up to detectable levels when the phase inverting switches adjust the phase inversion in the loop arrangements appropriately. When the inversions are complementary, i.e., the transmitter switch is inverting and the receiving switch is non-inverting, the conditions are set for energy build up. This build up is reinforced by the amplification in the loop. When this condition is sustained for a sufficient period of time, energy will build up to a detectable level even in a channel of restricted bandwidth, even though phase inversions at each location may change frequently (up to n times in a time t) provided they are complementary, i.e., inversion in one location non-inversion in the other.

Energy build-up will be suppressed whenever the codes are not complementary. Individual bit locations at the receiver individually and separately detect the presence of a complementary code at a corresponding bit location at the transmitter. (As mentioned before, the bit locations are connected to illuminated points on the scene in the TV camera and in the receiver to individual light generating points of the TV screen. However, in general these bit locations may be digit positions stored on shift registers or other accessible bit storage locations) . Energy will not build sufficiently to be detected if the receive and transmit codes do not match (correlate) , so that only complementary received bit positions will detect corresponding transmitted ones.

The amplitude of energy in the common channel transmission link flowing from transmitter to receiver and receiver to transmitter may increase with the number of simultaneously energized pixels (or bit locations registering "1" bits) , but the bandwidth will remain fixed and controlled by the narrow band filter. A limiter may be used to limit the rise in amplitude in the links.

A receive bit location will not register a bit "1" if the corresponding transmit bit location is not actively transmitting a bit. Other transmit bits will cause energy oltage to be present on the common transmit link. Thus, voltage is limited at each bit location. The receiver switched inverter of such a bit location, however, will chop up this voltage, rendering it unsuitable to pass through the low pass or narrow pass filter of that bit location which passes energy to the receive to transmit link and which feeds signal to the bit detector. Thus interference among the bit locations is eliminated.

It is therefore seen that the present invention can transmit a simultaneously available ensemble of information over a narrow band channel, the bandwidth of which is determined only by the rate of time change of the entire ensemble of transmitted bits. In the case of TV or motion video this rate of change is determined primarily by the persistence of vision. It may also be determined by motion within the scene. Sending changes in individual bit locations will reduce the number of simultaneously transmitted bits in the ensemble. This effect may be used to further reduce the transmit rate and bandwidth.

The reduction in bandwidth requirement also reduces the transmit power requirement in proportion to bandwidth where the signals compete with noise. This power reduction is significant. In the case of TV the power can be reduced by a factor of as much as 1,000,000 or transmit power of 100 milliwatt would provide the same useful signal as a present transmitter of 100 KW, for example.

The usefulness of this system is not limited to TV but may be applied to any digital transmission system such as digital telephone. Digital phones using PCM require 64,000 bits per second to transmit a voice conversation. Applying this invention to this situation would, for example, allow such a phone conversation to be transmitted in a lKhz bandwidth over voice frequency lines. Such a system could use 64 codes repetitiously used each millisecond. Refinements discussed below could reduce the number of codes to 16.

The instant invention therefore provides the following useful features.

Reduced TV Transmission Bandwidth

The transmission of TV is limited only by the changes in the televised scene thus reducing Bandwidth requirements by factors of as much as 1,000,000. Reduced Digital Data. Voice and Music Transmission

Transmission of digital voice may be reduced to bandwidth requirements that can be accommodated by existing analog transmission facilities or narrower band. Much higher transmission rates for digital data can then be provided by narrow band telephone and radio facilities.

Flat Plate TV Reception

TV reception apparatus may easily be instrumented as a flat plate, using this technology in conjunction with existing flat plate display technology, which employ methods to drive each pixel individually.

Low Transmission Power

This results from the low transmission bandwidth requirement. Radio Frequency transmission power requirement may be accordingly reduced up to 1,000,000 fold or more.

Interference

Interference is reduced by the cooperative transmission format.

Security

Transmission security compatible with total security methods described in Patent Nos. 4,805,216 and 5,029,210 can be provided. Three Dimensional Representation May Be Accommodated

A third set of codes may be used to specify pixel locations in three dimensions. These codes can provide isometric representation.

May be used to Address Computer Memories (DMA)

Random Access memories may be remotely addressed over narrow band links. Access Time may be greatly reduced by simultaneous transmission.

Facsimile Transmission of High Resolution Material

Over narrow band facilities, resolution and speed of transmission are both enhanced by the number of pixels simultaneously transmitted.

Color and Gray Scale of Expanded Range

These are easily accommodated, such attributes are digitally encoded and may be sent by successive bit planes.

Changing Pixels

Changing pixels may be handled separately from non-changing ones further reducing bandwidth requirements.

Random Accessing of Networked computers and Space Vehicles

A large number of computers may work in parallel at high speeds although connected by narrowband links. Interspersed commands are possible. Space vehicles may send much larger bit rates longer distances by sending large amounts of data in parallel. Power to communicate will be vastly lower reducing communication equipment requirements (such as antennas) .

Brief Description of the Drawings

FIG. 1 is an overview of a Scanless TV System.

FIG. 2 is the Transmission System.

FIG. 3 shows examples of codes used in the system.

FIG. 4 is a diagram to facilitate understanding of the operation of the Transmission System.

FIGS. 5 A and B, are more detailed views of the transmission encoder and decoder units, and C2 is one embodiment of the common storage unit for codes.

FIG. 6 is a diagram of the system used for reduced bandwidth digital transmission.

FIG. 7 shows method of encoding gray scale and other attributes.

FIG. 8 shows the multidimensional binary encoding system.

FIG. 9 shows details of Types of Switched Inverters.

FIG. 10 shows a single channel transmission system.

Detailed Description of the Invention

In FIG. 1, item 1 is the new sending apparatus for transmission of scene material item 3. Item 2 is the new receiving apparatus for converting the transmitted information to usable information for the receiving screen. Scene 3 may be made up of individual pixels lined up in rows and columns. These pixels are light sensitive and may be sensitive to specific colors. Each pixel such as 5 is connected to the sending apparatus via lines 10 through IN and 20 through 2N. Optionally, each pixel can be connected to the send apparatus 1 via two connections; thus pixel 5 may be connected by lines 10

-_. and 2N as will be shown in detail later. The use of two connections per pixel can result in a great saving in the number of individual codes that must be used in the transmission system. The connections 10 through 2N are individually connected to bit generators in the send apparatus, such as 100 and 107 in FIG. 2. The send apparatus is connected to the receive apparatus via transmission link 112. A return link 113 is also used. As will be seen, these links require very small bandwidths in the order of 60 hertz. The receive apparatus is connected via its bit receivers such as 119 and 123 in FIG. 2 to the screen. The screen consists of individual points of illumination. The value of illumination of individual pixels is determined by the bit receivers. Again each pixel may be individually connected via lines 30 through 4N but it is a: advantageous to use two connections per pixel, one row connection and one column connection. The row connections being 30 through 3N and column connections are 40 through 4N. Thus the illumination of pixel 7 is determined by said lines 30 and 40 corresponding to the transmission lines 10 and 20. The illumination of 7 is governed by the use of an AND circuit which logically adds the outputs of 30 and 40. The understanding of the transmission system for accomplishing the transmission of images from scene 3 to received scene 4 requires the use of FIG. 2 which details the transmission system. Each pixel of the sending unit is connected to a bit generator such as number 100 shown in FIG. 2, the system is specifically useful for transmitting digital information or information in binary form. When the pixels are not in digital form, it would be necessary to convert those pixel values into binary form. One approach to this would be to send the gray scale information in the form of bit planes.^' Bit planes are scenes containing bits of a common value or level of significance. Thus, gray scale could require 4 bits which can be transmitted separately and successively, one bit plane at a time. This will cause a bandwidth increase by a factor of 4 or an increase from 30 to 120 hertz, for example.

Bit Generator 100 transmits bits when they are present at the pixel location to XOR operator 102; a locally generated code is also provided to 102 by coder 101. The codes generated by 101 and 108 are orthogonal codes of length n where n is governed by the number of pixels. As mentioned before, one method would be to provide a bit generator or each pixel in the raster which would, for a 1000 x 1000 raster, require a code length of 1,000,000. FIG. 3 shows simplified codes for the case of n = 3. Orthogonal codes have the property of a correlation coefficient between codes of 0 with an autocorrelation of unity. Such codes are beneficial in the present system. The codes are repeated for the time interval of 1/30th of a second so that he smallest bit interval would be l/(30th x 1,000) of a second for n = 1,000. One method of generating these codes is by a continuously revolving multiple head magnetic drum or disc of the right speed which would contain all of the necessary codes such as 102 and 109. These codes will remain fixed for the full operation of the system.

The output of 102 is used to control a switched inverter 103. The operation of 103 is to provide an inversion of phase (a shift of 180 degrees) when it is switched from one position to the other. Inverter 103 has only these two operating positions. This inversion is provided by switching in or out one stage of amplification as described in U.S. Patent No. 5,029,210.

Switched inverter 103 operates in a loop consisting of 103, amplifier 105 the transmission link 112, the filter 114, which comprises one transmission link, the switch inverter 117, lowpass filter 118, amplifier 115, other transmission link 113 and filter 106.

Switched inverter 103 and switched inverter 117 of the receiver act cooperatively in this loop. The inverters 103 and 117 operating with the amplification provided by 104, 105 and 115 cause an oscillation to occur when one or the other switch inverter is providing an inversion. If both 103 and 117 are providing the inversion or 180 degree phase shift at all frequencies there will not be an oscillation in this loop. If both of the inverters are not inverting, providing zero degree phase shift, there will also not be an oscillation. The oscillatory state of the loop is detected by the bit receiver 119 which rectifies the oscillation and converts it into a "1" bit. This "1" bit activates a pixel in the received screen. However a bit will not be received unless the oscillation lasts 1/30th of a second. Filters 114 and 118 prevent the reception of signals unless they are sustained for this period of time. Of course this period can be made to be other values depending on circumstances and can be longer than 1/3Oth of a second for input material that does not change that rapidly. Decoder 116 operates the switched inverter in similar fashion to 101. Only when the decoding code is complementary to the transmitted code will a low frequency sustained oscillation occur for the period of 1/3Oth of a second. Should other codes be generated, such as by 108, for example, which do not provide a complementary code sustained over this period, the filters 114 and 118 would prevent any such transmission.

Amplifiers 104, 105 and 115 and possibly the amplifiers in 103 and 117 provide sufficient amplification to overcome the losses around the loop. Thus the operation of the transmission of one pixel from the transmitting scene 3 to the received screen 4 is as follows: if a pixel is illuminated such that a bit must be sent, bit generator 100 will provide for 1/3Oth of second a 1 bit level. At the same time the coder 101 is constantly sending an orthogonal type code consisting of a total of "1" or "0" levels, the minimum width of which is 1/nth of 1/3Oth of a second. Since decoder 116 has the same code as 101, and the XOR unit 102 converts the 101 code to the complement of 116, a sustained oscillation will occur around this loop. This sustained oscillation will be detected by 119, the bit receiver, which will illuminate the corresponding pixel in the receiver screen.

Another pixel will energize bit generator 107 when it is suitably illuminated or activated. The operation of this channel is entirely similar to the operation of the channels connected to bit generator 100, however, coder 108 has a code different than that of 101 and corresponds to the code contained in decoder 120. Once again the activated bit generator mixes with the decoder in XOR unit 109 which controls switched inverter 110 which is part of the loop containing amplifier 111 and both these units are part of a loop containing 105, 112, 114, 121, 122, 115, 113 and 106. Note that 105 amplifier, 112 transmission link, 114 filter, 115 amplifier, 113 transmission link and 106 filter are common to both of these loops. This means that operation of any of the pixels two of which are shown in FIG. 2, uses a common transmission path and requires exactly the same narrow bandwidth primarily determined by the change of scene (or bit ensemble) requirement. It should be noted that if it is useful to send only changes in the pixels which may change less frequently than the persistence of vision, it would be possible to further narrow the bandwidth of 112 and 113 transmission link. This requires modification of the bit generator to provide the change transmission capability. Modification of the bit receivers is required to modify pixel value when changes are received.

Operation of the Transmission system

FIG. 4 can be used to follow the operation of the transmission system shown in FIG. 2. In the FIG., column 1 indicates the value of the bit being transmitted either "1" or "0", column 2 indicates the arbitrarily chosen number of the pixel position, column 3 shows the transmit code, column 4 shows the transmitted bit waveform, column 5 shows the resulting state of the transmission switched inverter as to whether it is in an inverting ("1") or a non-inverting (or "0") state. This state is the XOR product of column 2 and 3. When in the inverting state a block is shown; when it is in a non-inverting state no block is shown. Column 6 shows the state of the receiver switched inverter and column 7 shows the reception of a bit wave form (a "1") or a non-reception ("0"). Non-reception is shown by a line indicating zero energy; finally, in column 8 is shown the state of the transmission lines which are items 112 and 113 in FIG. 2.

The first line represents the transmission from pixel 1. In this case pixel 1 has a transmission code of all zeroes as shown in column 3. The XOR operation on the transmit code is shown in column 3 and results in the transmission switched inverter (item 103) in FIG. 2 to be in the invert state. The corresponding receive pixel position in FIG. 2 has a corresponding code in decoder 116 so that the switched inverter 117 is non-inverting as shown by the zero amplitude in column 6 and the bit receiver 119 will receive an amplitude designating receipt of a "1" Bit, as shown in column 7. There will be an amplitude on the transmission channel as shown at column 8.

Pixel 2 has a different code orthogonal to that used in Pixel 1. As shown in column 2 a "1" bit is being transmitted which is XOR processed with the code of column 3 to produce the transmitted pattern of inversions presented in column 5. This pattern is complementary to the inversion pattern of the receiver pattern of column 6 resulting in a "1" bit being present at pixel 2 as shown in column in column 7. There will be an amplitude on the transmission link as shown in column 8. If both pixel 1 and 2 are active the amplitude on the transmission line could rise to the sum of the two. However, the system can limit this amplitude without changing the overall operation.

Rows 2 and 3 shows the operation when "0" bits are sent from pixels 1 and 2. The transmit codes shown in column 3 are the same as before and the inversion patterns are complementary. The transmit bits result in "0" bits being received as shown in column 7 and there is no amplitude on the transmission line (column 8) .

Rows 5 and 6 shows the case when the pixels send different bits. The inversion patterns (columns 5 & 6) remain complementary for each pixel position. Each receive pixel receives its corresponding pixel value, but the amplitude of the transmission line can only rise to the value caused by one active pixel. Note that the received bit level of pixel is zero corresponding to the transmitted level despite of the presence of amplitude on the transmission line. This is a result of the action of the receive switched inverter. This receive bit level will be zero at the end of the bit interval.

Rows 7 and 8 shows the operation for a Pixel 3 which has a different code that pixels 1 and 2 and is orthogonal to both of them. This code causes different inversion patterns than the others but provides complementary transmit and receive inversions (columns 5 and 6) and correctly conveys the bit level to the receiver. The coders 101 and 108 and other coders used for the rest of the transmit pixels require fixed codes which are repeated for each bit interval. These codes are stored in a common storage unit 200 as shown in Fig. 5a. This unit is a multi track memory with multiple readouts 201 a - - n. The storage medium may be an endless magnetic tape, a magnetic drum, a bubble magnetic memory, or a preprogrammed read only memory (PROM) . The speed and phasing of the codes stored in this memory must be coordinated with the decoder (such as 116, 120) which are stored in similar units 300 as the coders. The synchronization is effected by the use of one of the tracks, 202, dedicated to synchronization. The Transmit Coder Unit sends the synchronization code in the same manner as a pixel code. All these codes are constantly repeated. The decoder unit contains a controller unit 303 which reads the output of the sync-channel detector and adjusts the memory unit 300 in time reference and time rate (if required) until a constant amplitude input to the control unit is achieved. Then the receiver coder apparatus is time phase locked to the transmitter. The transmitter is governed in time and phase by Timing Unit 203, which in an electronic embodiment is a clock supplying regular clock read pulses. In the case of the endless tape or belt embodiment shown in figure 50, the timing is controlled by the speed of the tape or belt drive motor.

The same transmission system used for implementing the scanless TV system can be used for reducing the bandwidth requirement of a digital transmission system for a given information rate. In this case as shown in FIG. 6 the TV system is replaced by the transmission buffers 51 and 52. 51 receives serial digital transmission information from transmission line 50 and 51 delivers information to serial transmission line 53. The buffer 51 feeds parallel information into Transmitting apparatus 1. The parallel information bits are the same as pixels in our previous description which are fed into the bit generators which provide input to the XOR units which encodes the Switched Inverters. The n parallel bits are transmitted via the common channels 112 and 13 to the receiving apparatus 2, which conveys the reception to parallel bit positions of the buffer 52 - every t seconds.

The buffer is serially read out by a local clock. The buffers contains an intermediate storage so that new data can be received while the data received at the end of the previous T second interval is being clocked out.

This procedure allows the data rate of the transmission line's 112, 113 to be 1/n of the data rate of lines 50 and 53. Where n is the number of parallel bit positions buffered. Thus if the data rate of the original lines is 100,000 bits and n is 100, the effective bit rate of the transmission lines 112 and 113 is 1,000 bits per second. Thus digital voice may easily be transmitted on voice frequency channels. T-l channel operating at 1,544 MBPS may also operate over voice frequency line if n = 1500.

Gray Tones and Other Pixel Attributes

The code word can be extended to include specification of gray scale, color and other attributes of a given pixel location. This is done by assigning additional identification codes to a pixel input, so that a pixel may have access to three, four or more such codes. Gray tone or scale may be represented three or four or more bits to represent 8, 16 or more levels. Tint and hue may also be denoted by additional bits. (The procedure is similar for those attributes.) Each of the gray scale bits controls a separate ID code as shown in Figure 7. For pixel 1 gray scale bits 430, 431, 432 control the sending of I codes I , I_lb, and I_lc, which are respectively 433a,b, and c, respectively. Similarly for pixel 2 bits 434, 435, and 436 control 1^, 1^, and 1.^, 437a, b, and c, respectively.

At the decoder the code sequence 455 is decoded as usual. However, the additional bits per pixel are decoded separately; i.e., a separate output at pixel 1 is obtained for I , I_lb, and I,_c at 438, 439, and 440, and for pixel 2 separate outputs are obtained for 1^, 1^,, and I_2c at 441, 442, an 443. The three outputs for pixel 1 are presented to digital to analog converter (D/A converter) 444 which derives the gray scale output for pixel 1. D/A converter 445 uses the separate outputs 441, 442, and 443 to provide the gray scale output for pixel 2.

This method of gray tone or attribute coding increases the total number "virtual" pixels to be encoded by a factor g, so that: n = pg where p is the number of pixel locations and g is the number of gray tone (or attribute) bits log₂L where L = no. of tones and n is the total number of bits to be encoded. This causes the bit length b = log₂n to be longer than that required for the actual number of pixels. Though the code C is longer, the compression is greater because the compression ratio improvement is

L/g Another method is to send the gray tones one frame at a time. This way the bandwidth must increase and the compression ratio does not improve.

Multidimensional and Binary Encoding

In multidimensional operation transmission codes may be used by several pixels according to the demands of the binary representation of each pixel. As shown in Figure 8, the pixels first activate the identification coders 450, 451, ... 453, for each active input. These I codes are b = log₂n parallel (simultaneous) bits which feed the connection matrix 454. Individual I bits activate several output bits in the connection matrix. The output of this matrix designates code 455 which is comprised of C_xl&I,C_yj&I,C_x2&I C_y&I...C_xb&I,C_yb&I. This composite code represents one or many or all of the input pixels. A given transmit code bit may be activated by several pixels. The total number of transmit codes of 2b² provides sufficient codes for transmitting all the pixels whether or not all or a few pixels are transmitted simultaneously, because each receiver pixel location is uniquely masked to be responsive to only those multiple groups of codes (bits) which apply to the specific pixel location. Combined Voice and Video

Storage Buffers as given in FIG. 6 for digital transmission may be used to send combined sound and video in the same apparatus. If the video requires C = 2,000 codes, for b=2, additional codes may be added for the sound. If digital sound at the rate 100,000 bits per sec accompanies the video, arid the picture delivery rate is 100 frames per sec, then 1,000 bits of sound may be transmitted simultaneously per frame by the use of C= 2(1,000)^Iβ = 66 additional codes to the 2,000 codes used for video.

Transmission Medium 112. 113

Very few restrictions are placed on the transmission link 112 and 113. The common channel transmission may be translated to any suitable frequency by mixing with an oscillation on transmission and down converting the signal at the receiving end. Both 112 and 113 may be handled in the same fashion. The mixed oscillations may be used for selective addressing. Both wired or wireless facilities may be used.

The system may also operate directly with a retrodirective oscillation loop so as to provide automatic directivity as taught in U.S. Patent No. 5,757,335, as well as increased transmission capacity.

Switched Inverter Implementation

The switched inverter may be implemented using a digitally controlled analog switch 500 shown in FIG. 9. The chip type 4016 or 4066 may be used. They switch radio frequency signals at input 498. The control signal 497 is digital and switches channel outputs pin 2 and pin 4 inversely; that is, pin 2 passes signal when pin 4 does not and vice versa. In FIG. 9a, the output of pin 2 passes through a linear amplifier 501 of odd number of stages and therefore the signal experiences a 180 degree phase shift or phase inversion with respect to signals passing through pin 4. Either signal is sent to output 503.

In FIG. 9b, the output of pin 2 passes through a transformer 505 irt which the polarity of the signals in the secondary winding is reversed with respect to ground. The output of pin 4 is fed to a transformer 504, the polarity of the secondary of which is not inverted with respect to ground. The result of this is that the output of the transformers 503 will be shifted 180 degrees out of phase in accordance with the state of the input 497.

Single Channel Transmission Method

A single channel may be used for transmission instead of a separate forward and return channel by making use of an additional 90 degree phase shifter in each terminal as shown in Figure 10. The phase shifter is installed in the return line to cause the phase around the local loop to be always 90 degrees regardless of the setting of the inverting switch. The phase around the loop formed by the path around both terminals contains both 90 degree shifts as well as the phase inversions of both terminals. This merely inverts the sense of the inverting switches which can be compensated for by changing the sense of the codes applied to the inverting switches. As Figure 10 shows only one transmission link 112 is used. Filter 106 is connected to 112 instead of transmission link 113 which is not used. The new phase shifter 401 is connected to 106 and to amplifier 104 and to other branches (via z) . The local terminal path is completed via switch inverter 103 which connects to the transmission link 112 via amplifier 115. (Other branches connect to 112 through 105 via w) . This local path will not oscillate because the loop phase is always a multiple of 90 degrees.

The other terminal path includes filter 114, switch inverter 117, the new phase shifter 402 and amplifier 115. Other branches connect through x and y, Again the local loop phase is 90 degrees or a multiple thereof. The mutual loop containing the two terminals, however, can oscillate because the switch inverters can cause the loop phase to be multiples of 180 degrees.

Claims

I Claim;

1. A transmission and reception system which comprises a first apparatus at a sending location which converts multiple simultaneously available patterns of digital information bits into simultaneously transmittable information codes using a common information transmission channel and a second apparatus at a receiving location capable of converting signals received from the common channel into simultaneously available receivable bit indications which correspond to the original pattern of simultaneously available information bits.

2. The transmission and reception system of claim 1, wherein the first apparatus comprises switched invertors for controlling oscillatory energy in the information transmission channel.

3. The transmission and reception system of claim 1, wherein the first apparatus comprises amplification and phase inversion means which are cooperatively operative at transmission and reception locations.

4. The transmission and reception system of claim 1, wherein coding means are provided for each individual ensemble member of the first apparatus which combine with bits of ensemble to produce transmitted output only if corresponding complementary coding means exist at the second apparatus. 5. The transmission and reception system of claim 1, wherein the first and second apparatuses comprise means to reduce the length of codes and number of codes by providing two or more codes per transmitter ensemble member and providing the correct combination of the output coded for designating the received member of the transmitted ensemble transmit.

6. The transmission and reception system of claim 1, wherein the first apparatus converts spatial patterns of bits to signals transmittable in common channels of narrow bandwidth and variable amplitude which operates cooperatively within the second apparatus to recover the spatial pattern.

7. The transmission and reception system of claim 6, wherein there is formed a retrodirective oscillating loop controlled by the first and the second apparatus cooperatively.

8^~. The transmission and reception system of claim 5, wherein ^{• '}the means to reduce the length of codes and number of codes utilize Logical AND units.

9. A transmission system comprising a source of information, a buffer to hold a block of this information in a specific and identified geometric arrangement of bits, the system of claim 1, a transmission link for transmission to and from each station and a means for assembling the received information into a buffer from which the corresponding geometrical set of information bits is extracted. 10. A cooperative communications system comprising first and second communicating stations each containing apparatus consisting of several branches, a communications link connecting first and second stations in a communications loop, each branch containing predetermined number of amplifying stages, each branch containing phase inverting means responsive to sequential coding means, each branch encoded substantially differently, means to modify said codes by information bits such as to invert the code at particular branches of the sending station apparatus, means to detect status of each branch of the receiving apparatus and thus to detect presence of bit information at a branch of the receiving with corresponding non inverted code to that of the particular branch of the transmitting station, and a synchronizing means which uses a single branch modulated with a standard information pattern.

11. A cooperative communications system of claim 10, wherein each of the stations include a 90 degree phase shift unit in the loop path and coupling unit so that the communications link comprises one wire or transmission channel for both the forward path and the return path of the communications loop.

12. An apparatus for use in conjunction with the system of claim 1 and the communications system of claim 10, which comprises means to code the information organized in the form of an array of independent sources into a set of bits substantially fewer in number, means for these bits to modify the phase of particular branches of the communications system transmitting station, and means to decode the receive bits so as to completely receive the original organization of bits without loss of information.

13. An apparatus of claim 12, wherein the encoding means consists of a means to organize the data inputs into a geometric matrix, means for marking each data input position with a binary code designating the input uniquely, means to provide multiples of the coded inputs in accordance with the position of the inputs in the geometrical arrangement of the inputs, and means to combine the multiple outputs on a logical OR basis on a bit by bit basis.

14. An apparatus of claim 12, wherein the decoding means for recovering each input comprises: means to compare specific parts of the received code, which had been determined by geometric encoding, with the code identifying the specific input and comparing the results of the comparison with each other and accepting the result to be a true output when all these comparisons provide a "1" output. 16. An apparatus of claim 12, wherein the encoding means and decoding means are capable of operating with multiple bits per pixel and means to derive correct pixel gray scale, color and tint from the multiple bits per pixel.

17. The transmission and reception system of claim 4, wherein the coding means comprise continuous multitrack magnetic medium driven at substantially constant rate at the transmission terminal and a similar multitrack unit at the receive station driven at substantially the same rate under control of synchronizing signals derived from a synchronizing track; the other tracks of the multitrack units containing unique orthogonal or shift register code sequences.

18. The transmission and reception of system of claim 4, wherein multiple code sequences are derived from a preprogrammed read only memory driven by a clock oscillator.

19. The apparatus of claim 12, wherein the coding and decoding are performed by a fixed program digital computer.