GB2411800A

GB2411800A - Spread spectrum communication system using impulse modulation

Info

Publication number: GB2411800A
Application number: GB0404848A
Authority: GB
Inventors: Philip John Connor
Original assignee: Individual
Current assignee: Individual
Priority date: 2004-03-04
Filing date: 2004-03-04
Publication date: 2005-09-07
Anticipated expiration: 2024-03-04
Also published as: GB2411800B; GB0404848D0

Abstract

The invention is a communication system where information is conveyed by impulse modulation of a propagation medium linking a transmitting entity and a receiving entity. These impulses which are short in the time domain give rise to broadband waves in the frequency domain. These waves propagate from transmitter to receiver through the medium so carrying information. Information is encoded firstly, by grouping together impulses and impressing some pattern upon them to form a "Bit Group" and secondly by then grouping together these Bit Groups to form the larger scale pattern or "Word Group". By this means a single broadband communication channel is converted to a multiplicity of logical channels configurable as a multi-user network. The invention provides means for encryption, transmit power control, spectrum spreading, flow control, multiple antennae, power saving and noise reduction. The invention is not specific to a particular frequency or propagation medium and can be utilised for digital communication in a variety of electronic, acoustic and optical systems.

Description

241 1 800

SPREAD SPECTRUM COMMUNICATION SYSTEM

BACKGROUND TO THE INVENTION

Impulse modulation is not new. The first man-made electromagnetic transmissions made by Hertz in the 1 880s were impulse modulated radio waves generated by spark gaps. However, since that time communication systems have evolved to use the alternative of continuous wave methods with a sinusoidal carrier. Continuous wave methods have captured the majority of the application niches in communications.

Recently impulse techniques have been the subject of renewed interest since it has been recognised that they have many advantages for digital transmissions.

The present invention is a new technique of impulse modulation that has many advantages over alternatives. The invention greatly simplifies communications circuits reducing their cost and complexity and making many more applications economically feasible. The principles of the invention are not frequency or propagation medium dependent. The invention can be applied to communication systems using radio, optical, infrared, acoustic and ultrasonic data links. The invention can work at audio frequencies through to optical frequencies and above.

FEATURES OF THE INVENTION

The invention accepts as a prior art that impulses can be generated, travel over distance, and be detected and so form a communication system as exemplified by Hertz and more recently by Ross (US Patent No. 3,728, 632, 1973).

It is further acknowledged that a number of methods exist for encoding information with impulses. These are described extensively elsewhere in the literature and examples are:- OOK (On-Off Keying) where the presence of an impulse might represent a logic "1" and its absence a logic "0", PAM (Pulse Amplitude Modulation), PFM (Pulse Frequency Modulation), PPM (Pulse Position Modulation), PSM (Pulse Shape Modulation), et al. The present invention is described using the methods of OOK since this offers the most efficient spectrum utilisation but many of the modulation techniques listed above could be adapted for use with the invention.

The present invention is a means of encoding and decoding data with impulses such that multiple access networks are enabled. The essential feature of the invention is that a single binary data bit is provided with a code for the purpose of addressing either its destination or its functionality within a communication network. This coding of a single bit is achieved using a group of impulses impressed with some pattern and termed the "Bit Group". The pattern can be impressed by any of the established modulation methods as mentioned above. The Bit Groups can then be further grouped in time to convey a data byte or binary word which is termed the "Word Group". By arranging for the Bit Groups to be an orthogonal set it becomes possible to convert the single physical wideband channel into a multiplicity of logical channels which can be simultaneously used by a plurality of communicating modules. It is accepted that collisions of Bit Groups will occur on rare occasions and this is expected to be accommodated for with standard error recovery measures and re- transmission schemes.

The invention also straightforwardly and conveniently provides many of the functions required within communication networks. The invention enables the interconnection of a number of modules to form a network. There is means by which individual modules may be addressed within the protocol making a multi-point to multi-point topology practical. The invention has measures to broaden the spectrum and reduce peaks in the power density. The invention provides means to monitor quality of the coupling (reception) in the datalink along with provision to control power levels in the transmissions. The practical aspects of datalink management are attended to with methods to minimise circuit power consumption, control errors and overflows and overcome the effects of antenna nulls. The invention considers interference and noise and provides measures to reduce their deleterious affects. There are two special encryption methods that can be employed to make the link secure against interception by an eavesdropping third party. A means of "data blackening" is outlined to minimise spectrum utilization.

The invention will now be described in detail, and the claims clarified, by reference to a set of Figures and Embodiments.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1 shows an idealised delta-function like waveform which is hereinafter termed an "Impulse" which is very narrow in the time domain and so produces a broad spectrum in the frequency domain.

Figure 2 shows an overall block diagram of a one-way spread-spectrum communication system.

Figure 3 shows a number of time domain Impulses being grouped together to form a code that represents a single bit hereinafter termed a "Bit Group" which may contain any number or form of Impulses coded in any way.

Figure 4 shows how a data word is encoded using several instances of Bit Group 2 taken from the class of Bit Groups of the previous Figure 3. These several instances taken together comprise what is hereinafter termed the "Word Group" which may contain any number of Bit Groups.

Figure 5 shows two electronic modules both equipped with a transmitter and a receiver so as to form a two-way spread-spectrum communication system.

Figure 6 shows a three module example of a multi-module, spread-spectrum system. It illustrates how there are three ways that one module may become a "Master" issuing commands to the remaining Modules which become "Slaves".

Figure 7 shows how the position in time of the Impulses may be "dithered" to further broaden the spectrum without disrupting the Word Group. This "dithering" may also be applied to the Bit Grouping.

Figure 8 shows how a module may enter a power conserving "sleep mode" during the time interval between individual Bit Groupings.

Figure 9 shows how a known acceptance window can be used to reduce noise.

Figure 10 shows how a module may switch into a "sleep" mode as soon as there is a mismatch between the decoded Bit Group and the code that is being looked for.

Figure 11 shows how a plurality of Transmitters or Receivers can be arranged to produce signals within a single acceptance window and so overcome antenna null effects.

Figure 12 shows a circuit block diagram of a simple pulse height discriminator that is sensitive to either positive or negative Impulse spikes.

Figure 13 shows the signals at various points in the circuit of Figure 12.

Figure 14 shows "Bit Group Code Encryption" where a sequence of Word Groups is encrypted by changing the Bit Group code during the exchange of successive Word Groups.

Figure 15 shows ''Impulse Spacing Encryption" where a sequence of Word Groups may be encrypted by changing the time separation of the Bit Groups between successive Word Groups.

Figure 16 shows the Bit Grouping scheme that is utilised in an embodiment of the present invention in which Modules are addressed by the Bit Group Code to facilitate a multiple access network arrangement. The scheme produces quasi-orthogonal Bit Groups by using multiples of a fixed delay interval.

Figure 17 shows the Word Encoding scheme in an embodiment of the present the invention where a binary '1 ' (say) is indicated by the presence of say Bit Group 1 of Figure 16 in the required window whilst its absence in the window is indicative of a binary '0'. Non- contributing noise pulses are also shown.

Figure 18 shows an example of a word format that could be used according to the invention allowing commands, addresses and data to be exchanged as part of the word data.

Figure 19 shows a simplified block diagram of a Transmitter circuit to encode data with Bit Grouping and Word Grouping according to Figures 16 and 17.

Figure 20 shows a simplified block diagram of a Receiver circuit to decode Bit Groups and Word groups encoded according to Figures 16 and 17.

DETAILED DESCRIPTION AND PARTICULAR EXAMPLES

The invention is a spread-spectrum communication system utilising a new technique of data coding with time-domain Impulses. The invention provides a one way data-link such that an entity or module may send information to a plurality of other modules. This is a point to multi- point link. By employing two instances of the invention two-way communication is enabled so that the network comprises a multipoint-to- multipoint network.

In discussing such networks the adoption of the following terminology is helpful: The process of establishing a link between modules, transferring data and terminating the link is hereinafter termed a "Transaction". There may be a number of Transactions, between different modules, active at any one time. In a particular Transaction the modules form a group with only one "Master" and the remainder becoming "Slave" units. A Slave is defined as being a module acting under commands issued by a Master. In a transaction the data may flow from Master to Slave or from Slave to Master. The distinguishing feature is that only the Master sets- up the datalink, controls data transfer and then terminates the link. A slave cannot do this except by initiating a further transaction in which it is the Master. Master and Slave may be either a "Data Source" or a "Data Sink" for the transfer process, these terms referring to any transaction data payload rather than the controls signals. Data Sources and Data Sinks both transmit and receive information during the transaction. At the conclusion of the transaction the data payload has been transferred from the Source to the Sink.

The preferred embodiments of the invention will now be described with reference to the figures.

The communication method involves the use of Impulse signals of very short duration in the time domain. These Impulses have the property of having a broad spectrum in the frequency domain. This is illustrated in the idealised waveforms of Figure 1. Impulse modulation can therefore form a natural spread spectrum communication system without the need for the circuitry that is commonly used to accomplish this spreading. Examples might be frequency- hopping circuits, devices to produce spreading codes or components for wavelength division multiplexing in optical systems.

The most basic element of the invention is illustrated in Figure 2 which shows a one-way spread-spectrum communication system. This comprises an Impulse Modulator with accompanying Broadband Transmitter and Antenna. This is linked via the transmission medium ( e.g. electromagnetic, acoustic or other waves) to the Antenna of a Broadband Receiver module and accompanying Impulse Demodulator. Modulator and Demodulator are connected to their individual host systems by some parallel or serial bus system according to any of the well-known and established methods. Taken together this arrangement comprises a one-way datalink. Information flows from Modulator through Transmitter and Receiver to Demodulator.

The Modulator takes in standard digital data from the Bus and converts it to a series of Impulses according to the principles of the present invention with protocols described below.

The Transmitter converts these Impulses to broadband waves in the transmission medium, which propagate to the Receiver. The Receiver detects the broadband signals and reconstructs the Impulses that generated them. These Impulses are then passed to the Demodulator to be converted back to the original standard digital data by the methods described in this invention.

This digital data is then presented as an output on the Bus for communication to the host system.

In one embodiment of the present invention the Modulator codes information in the Impulse waveforms by the method shown in Figure 3. A group of impulses is impressed with a particular pattern by OOK of impulses in particular time windows relative to other impulses.

The figure shows three different patterns. The presence of any one pattern is taken to indicate a logic 'I' (say) and its absence a logic 'O'. By this means a single binary bit is encoded by the presence of particular pattern in a group of Impulses. As shown in the diagram a number of different patterns can each represent a bit and each pattern can therefore be associated with a different recipient. This can be used to address communications to a particular module within a network of modules. This means a single data bit has been coded with an address. If the pattern is present then it is recognized as a logic '1' (or 'O'). Conversely if the pattern is absent it is recognised as a 'O' (or ' 1'). The patterns are chosen to be part of a set of patterns that are quasi- orthogonal in the sense that no one pattern can be confused with another. A suitable set of patterns will be discussed in more detail below (see Figure 15). In this fashion each Bit Grouping can indicate a unique address for a particular bit to be sent to a particular Module.

The Bit Group can also be used to offer a control function for the data bit and this is discussed later.

An important feature of the invention is that each and every Bit Group that is transmitted carries a code which means that its absolute position is time is not important to the coding scheme. The module address can be decoded whenever the bit group appears. In other words the Bit Group is free to appear at any point in time and its code can still be recognized. This is in contrast to alternative multiple-access schemes where a code is used to specify an acceptance window in which a single impulse must appear. In the latter case the single impulse codes the data and its position in time is tightly constrained for the coding scheme to operate successfully. This method does not have the power and flexibility of the Bit Group as will be seen later in discussions on UART type (Universal Asynchronous Receive and Transmit) operation, pulse "dithering" and encryption measures.

Another important feature of the invention is the wide tolerance on timing accuracy or time coherence. Timing is only important within the Bit Group which means any clocks only have to maintain agreement over the Bit Group duration which can be arbitrarily short.

Therefore the clocks in different modules do not need to have precisely the same frequency.

They also do not need to be in phase because it is possible to use an asynchronous, UART type decoding scheme. This again is in marked contrast to alternative schemes for single impulse multiple access which require high accuracy clocks which are in phase agreement over extended time periods.

Having modulated individual bits by the presence or absence of the Bit Grouping it is now necessary to encode data words of several bits. In one embodiment of the present invention the Word Grouping is achieved by producing a wave train with Bit Groups present or absent in successive time windows thus building up the required data word. This is illustrated in Figure 4.

This shows the Bit Group No. 2 of Figure 3 being used to build up the data word ' 1010'.

An important feature of Figure 4 is to recognise that the Bit Group 2 can appear anywhere within the windows indicated by the dotted lines and still be recognised as a logic '1'. In other words the timing requirements for the Word Group are even more relaxed than they were for the internal timing within the Bit Group. This has many advantages for spectrum spreading and encryption as will be seen later. It also means that the clock accuracy required in different modules is reduced compared with alternative single pulse multiple access schemes.

Having communicated one data byte as described in Figure 4 further bytes can be transmitted at arbitrary intervals to whatever scheme is suitable for the particular application. For example the transmission may be "UARTlike" with the next byte occurring at any point in time. That is the next byte may occur immediately or alternatively after a long delay. This scheme allows for flow control, for other data traffic to clear or for the set-up and hold times of slower devices.

Alternatively bytes may occur at regular pre-determined intervals with the receiver in "sleep mode" between bytes. This option would be very power efficient but may have longer latency.

Yet another option might be a pre-arranged pseudo random byte spacing which might be used for covert communications. There are also other schemes where the receiver might sample the broadband transmission medium to look for its Bit Group and in the event of failing to find a match would enter a hold-offperiod before a re-try.

In other embodiments of the invention the quasi-orthogonal Bit Groups are used to act as separate communication channels to a single module. In this way a single module can have a multiplicity of independent data streams which are used for different purposes.

In yet other embodiments the Bit Groups convey information in addition to the Word Groups to act as command or control information which is not part of the data streams. For example this information can be used as a control signal to toggle the sense of the logic between "normal" and "complementary" for "data whitening" or "data blackening" purposes. This is discussed further below.

For many applications it is desirable to have two-way communications. For this an arrangement of the invention is employed with two transmitters and two receivers. This is illustrated in Figure 5. Information flows in either direction. This allows for the acknowledgement and error checking of the data exchanged. For example a hand-shaking protocol is possible where fresh data is only sent when it can be received thus avoiding data overflow. Similarly a copy of received data can be echoed back to the sender for confirmation that the message is error free.

The two-way link also means either module has the ability to become a Data-source or a Data- sink as necessary, these terms being as defined above.

In some applications a multiplicity of devices need to take part in communications. For this purpose the invention is configured in the ways shown in Figure 6 where, for simplicity, only three modules are shown although there could be more. It illustrates how the modules are in mutual communication and a number of topologies are possible. The "Master" module is defined as issuing commands to the remaining Modules, which become "Slaves". The Master originates, controls and terminates a data exchange transaction. Additionally the module which is the source of the data payload in a communication transaction is defined as the "Data Source" and the destination module as the "Data Sink". Any module may be a Data Source or a Data Sink irrespective of whether it is a Master or Slave in a transaction.

In multiple access networks the Bit Group can be used as a module address which is dynamically assigned by a Master module in the data exchange Transaction. At the conclusion of the Transaction the Bit Group is freed up for re-use in other Transactions. By making Bit Group assignment dynamic it is also possible to have the shortest and most simple bit groups at any one time, consistent with the number of modules active on the network and the amount of network transmission traffic. For example if a network only has 2 modules active then the Bit Group can be simpler than when a situation where there are 20 modules are active.

One arrangement that is advantageous is to have a special Bit Group used as a common address for several Modules. For example in multiple access networks a particular Bit Group can be designated as a "Paging Channel" to which all modules must respond when establishing the network and initiating a transaction.

Special Bit Groups can also be assigned for the purposes of clock synchronization between modules. For example if a bit group contains a section with all impulses present then this can be used to check the timing error between the clocks in the Transmitter and Receiver.

A common requirement in spread spectrum communication is to minimise peaks in the spectral power density either to meet regulatory requirement or as an aid to covert communications. The present invention provides means to achieve this by allowing the imposition of random fluctuations on the position in time of Impulses within either the Bit Groups or Word Groups or in both. This can be achieved without affecting the overall coding scheme provided these fluctuations are within pre-set bounds. This can be seen with reference to Figure 7 which shows the Word Group case. The normal waveform is shown in the top trace. Here the Bit Groups occur with a well-defined period. This periodicity will result in a peak in the frequency components present in the spectrum of the transmission. In the lower waveform in Figure 7 it can be seen that the Bit Groups are "dithered" (subject to time delays of random magnitude and of either sign) . The magnitude of these fluctuations is controlled such that the Word Decoder can still decode the Impulses. This "dithering" reduces the size of peaks in the transmission spectrum.

A further common requirement of communication systems is to minimise the power consumption of the receiving circuitry. A technique that saves power is to disable parts of the circuit during periods when no activity is expected. This is commonly termed a "sleep mode".

The present invention allows a sleep mode to be entered both within the pulse train comprising the Word Grouping and also in the period between successive Word Groupings. This is illustrated in Figure 8. Here we see the Receiver switching on for a short period to find the start of the sequence of Bit Groups and then for a shorter period at each expected location of a Bit Group. This is the 'within Word" sleep mode. It is relatively straightforward to see that this principle can be extended to allow sleep periods in the potentially longer interval between Words.

Communication systems are usually subject to interference and noise tending to disrupt the datalink. In an Impulse modulation system the noise will be manifest as rogue Impulses at random positions in time. In the present invention the effect of these on the signal being transmitted is reduced be employing an acceptance window around the expected position of a wanted signal. This is illustrated in Figure 9. This shows the two noise pulses do not contribute to the "Bit Detected" signal in the lower trace because they fall outside the "Acceptance Window" which has been predetermined as the range of acceptable positions for the wanted signal.

When conserving power using a "sleep mode" it is advantageous to enter sleep mode at the earliest opportunity. In the present invention it is only necessary to decode a few Impulses within a Bit Group to determine if a message is addressed to a particular recipient or not. If it is not, then sleep mode can be immediately entered. This is illustrated in Figure 10, which shows how a module may switch into a "sleep" mode as soon as there is a mismatch between the decoded Bit Group and the required code for the receiving module. The codes match for the first three positions but then a mismatch causes "sleep mode" to be entered.

A common problem in communication systems is that transmitting and receiving antennae have nulls in their directivity lobes. It is common practice to employ a multiplicity of antennae with different orientations in order to overcome this problem. The present invention is so conceived as to allow the simple use of multiple antennae. This is illustrated in Figure 11, which for clarity shows a single Impulse Bit Grouping. The top trace shows the analogue impulse signal that is exchanged between Transmitter (Tx) and Receiver (Rx). This needs to be above the threshold to generate a logic ' 1'. With three Tx/Rx pairs there will be 3 signals of varying amplitude in the analogue signal. When the Impulses are transmitted there are time delays introduced such that the signals from all three antennae arrive in the one acceptance window.

This allows a common discrimination threshold to be applied. In this example only two pairs have a signal above threshold. These can be logically 'OR'ed together to derive a combined bit detected signal for the window. This provides an element of redundancy in the signals such that any one signal can provide the information for decoding without the need to determine which signal is contributing.

The present invention envisages that the orientation of the transmitting antenna with respect to the receiving antenna may be unknown. The polarity of the received Impulse may be positive or negative. This is accommodated in the circuit of Figure 12 which shows how a pulse height discriminator can be made sensitive to either positive or negative Impulse spikes. The waveforms in this circuit are illustrated in Figure 13.

In a communication network it is a common requirement to monitor the transmission power levels so that the power levels for those modules close to the receiver can be adjusted downward. The present invention provides for this functionality. The received signal power or quality of the data link in terms of the coupling efficiency can be determined by measuring the absolute amplitude at the Receiver of the Impulse signals. This is easy to see with reference back to Figure 11. The height of the peaks in the analogue signal measured in absolute units give the coupling efficiency for each antenna. This information may be used simply to confirm they are adequate and the link is thus noise resistant. Alternatively this received power information may be used to derive a control signal to be communicated back to the Transmitter thus providing means for active power control. At the Transmitter the power radiated in the broadband signal is controlled by modification of the amplitude of the impulse waveform.

These two features taken together allow easy monitoring and control of the power utilised in the link. This is important for reducing power consumption, minimising spectrum utilization, controlling unwanted emissions and reducing inter module interference.

Communication systems are often required to be secure or covert such that it is not easy for a third party to intercept a transmission between two other parties. The present invention includes two encryption methods. Figure 14 shows one technique which hereinafter is termed "Bit Group Code Encryption". This is where a transmission is encrypted by changing the Bit Group code over time in a way only known to the parties participating in the exchange. In Figure 14 it can be seen that the Word Group number N uses a particular pattern of impulses in the Bit Group. The Bit Group pattern (or code) then changes for the next (N+1) word. In this example the change only occurs as the word number increments but the change could occur in any pre- arranged fashion. In the extreme this method can be applied to successive bits such that every Bit Group uses a new code. The encoder and decoder have been made aware of the sequence of these changes by any of a number of means such as the exchange of an encryption key or the seed for a pseudo random sequence. If, as is the intention, an intercepting third party does not have this information then it cannot decode the message.

The second encryption technique is illustrated in Figure 15 and is hereinafter termed "Impulse Spacing Encryption". This is where the encryption is effected by changing, in a predefined way, the separation in time of Impulses either within a Bit Code or within a Word Code or both together. Again this might be a repetitive sequence or a pseudo- random sequence with a seed number known to both parties but not to the intercepting or eavesdropping third party. As an example in Figure 15 only the spacing of Bit Groups in successive words is modulated in this way but in the extreme every Impulse could have a varying positions in time to provide encryption.

To enhance both these methods of encryption the present invention envisages the transmission of "dummy" Impulses which form no part of themessage but merely serve to confuse a potential unauthoAsed interceptor of a communication. These false Impulses can be positioned or patterned to look like the correct Impulse patterns or Bit Groups but also be chosen so as not to disrupt the correct message.

As mentioned previously, there are a number of ways to achieve quasiorthogonal patterns for the Bit Grouping. As an example consider an embodiment using the method shown in Figure 16 which shows a simple Bit Group scheme using multiples of a delay interval. This coding provides four possible options for an individual bit pattern. Each pattern is not easily confused with another unless noise pulses are present or two bit groups are closely spaced. For this embodiment each Bit Group option is assigned to a module address. That is the address is coded by the separation in time of the two Impulses composing the bit group. This time separation is caused to be some multiple of a clock period which is known to both transmitter and receiver. In this way the addresses can be coded and decoded in hardware at the Transmitter and Receiver are described in Figures 19 and 20.

In this same embodiment now consider a module with Address O of Figure 16. The actual data may be encoded in the Word Grouping an example of which is shown in Figure 17. This waveform illustrates a real situation with noise and other pulses present but the module will only respond to the pattern of impulses shown as Bit Group 1 in Figure 16.

The Figure 17 waveform shows a typical real-life situation and exhibits several features indicated by letters as follows: Feature A is an Impulse pair with the correct separation and correctly positioned within an acceptance window. This will cause Bit O to be set to '1'.

Feature B is a single noise Impulse which will be ignored since it is not paired with another Impulse.

Feature C is an Impulse pair correctly positioned within an acceptance window but the separation of the pulses is too great and so Bit 1 remains reset at '0'.

Feature D is a correct pair in the correct window. Bit 2 is set to ' 1'.

Feature E shows an Impulse pair having the correct separation but not positioned within the acceptance window and therefore Bit 3 remains reset at '0'.

The present invention does not place restrictions on the length of the data word that can be used. As an example consider a word format as shown in Figure 18. This utilises a 16 bit word with 1 bit as a start bit (always present), 3 bits as an address code, 3 bits as a command code, 8 bits as a data byte and finally a 1 bit as a stop bit (always present). All this information is exchanged as part of the Word Group and is separate from the Bit Group coding.

Similarly the present invention does not place restrictions on the size of the Bit Group in terms of the number of impulses it may contain. This may contain as many or as few Impulses as are required for the application in question. Also, as previously discussed, the Bit Group may be dynamically assigned and change with time as the system accommodates for changes in the flow of data traffic. These changes may be in the number of active modules or in the requirement of existing modules for more or fewer data channels.

The coding and decoding of Bit Groups and Word Groups is straightforward using established and standard digital techniques. For example Figure 19 shows a block diagram of a logic circuit that can encode Bit Groups for a transmitter operating according to the embodiment described in Figures 16 and 17. In this figure the data to be sent is presented to the "Set" input of an RS flip-flop. If a data bit is present then this starts the parallel to serial shift register running and the counter is enabled to count. The shift register sends out a serial train according to the code pre- set on its data inputs and at a rate determined by the Tx Impulse Clock input. Each rising edge in the serial train is converted to an impulses at the Impulse Transmitter and then transmitted.

The burst counter resets the RS flip-flop when a complete bit group has been transmitted. The system then waits for the next data bit. The Bit Group is only transmitted for a logic ' 1 ' or a logic 'O' and not for broth.

Similarly Figure 20 shows a block diagram of a logic circuit to decode the Bit Group in the Receiver. The impulses are detected and pulse height discriminated and arrive as logic signals at the RS flip-flop. Any pulse above threshold will set the "Q" output to I for the remainder of the Rx Impulse Clock period and then it is clocked into the shift register at the next clock rising edge and the RS flip-flop is reset. The shin register continuously shifts through the data presenting the parallel outputs to the comparator. Each time the data stream matches the pre-set Bit Group code a logic signal will appear at the comparator output. The Impulse clocks at the Receiver and Transmitter do not need to be in phase and the periods only need to match sufficiently to maintain coherence over the duration of the Bit Group for the decoder to work.

For a variety of reasons it is common practice in networks to balance the numbers of '1 's and O's in a digital transmission (sometimes called "data whitening"). In an Impulse system it is desirable to minimise the number of impulses used to convey information since this minimises the spectral power utilised. This could be termed "data blackening" since it reduces the overall spectral power density of the transmission. As an example consider in the present invention, if a logic channel is predominantly at logic ' 1', say, then it would be best if this were represented under OOK by the case of the Bit Group not being present. To be more general, when there is an imbalance then whichever logic level predominates should become the "impulse off" signal.

In the present invention this is arranged by assigning a control signal, either within the Bit Group or within the Word Group, that indicates whether the logic in the signal being transmitted is "normal" or "complementary". That is this control signal says whether the presence of Bit Group is chosen to indicates a logic '1' or logic '0'. In this way the sense of the logic can constantly be toggling between normal and complementary throughout the communication to maintain any required bias. A delay is introduced such that the circuitry can "look ahead" for a preponderance of any logic state. The sense of the logic is then switched between the "normal" and "complementary" setting to make the dominant state the "impulse off" case. The Receiver will be able to determine the setting of"normal" or complementary" from the control signal. By this means the invention can reduce the number of impulses that are required to transmit information.

The embodiments of invention described above are only a few illustrations of the many possibilities. A large number of embodiments and applications directly follow, in a straightforward manner, from the generic descriptions of the invention that are given above. All these are intended to be covered within the claims herein.

Claims

What is claimed is: Claim 1. A spread-spectrum communication system for the exchange of data between separate entities or modules with said system comprising: (a) A Modulator providing means to encode data according to two concurrently operating coding schemes namely: 1. The Bit Group where a plurality of discrete impulse signals represent a single bit by being grouped together to form a pattern or code that conveys information about that single binary data bit;
2. The Word Group where the method of conveying information or intelligence is the repetitive re-occurrence of instances of the Bit Group so forming a pattern or code, (b) A Transmitter providing means to convert the electrical impulses into impulse waves with a broad spectrum propagating in a medium such as free space, (c) A Receiver which extracts energy from the propagating broadband impulse waves and thus provides a signal output consisting of a train of electrical impulses corresponding to the received broadband impulse waves, (d) A Demodulator providing means to distinguish between different Bit Groups in order to extract intelligence from the received signal and so reconstruct the Word Groups so extracting information in the inverse fashion to the Modulator, Claim 2. A spread-spectrum communication system as in Claim l where each entity or module is provided with both a sending channel and a receiving channel to give two-way communications.
Claim 3. A spread-spectrum communication system as in Claim 1 and 2 where a plurality of modules are communicating to become an ad-hoc network exchanging data.
Claim 4. A spread-spectrum communication system as in claim 1,2 and 3 where the information exchanged in the Word Group includes components that specify address, command or data information.
Claim 5. A spread-spectrum communication system as in Claim 1,2 and 3 where the information exchanged provides means for a sender to stop transmitting data when the receiver is not able to store the received data.
Claim 6. A spread-spectrum communication system as in claim 1,2 and 3 where the word length is only of a single bit meaning the Word Group only comprises one Bit Group. for $
Claim 7. A spread-spectrum communication system as in Claim 1,2 and 3 where the information exchanged includes copies of messages reflected back to the sender to enable verification as an error free transmission.
Claim 8. A spread-spectrum communication system as in Claim 1 and where, for the Word Group, the transmitted impulses are caused to be subject to varying pseudo- random time delays with the purpose of further spreading the impulse wave spectrum with consequential reductions of the peak spectral power density.
Claim 9. A spread-spectrum communication system as in Claim I where the receiving channel is capable of being switched into a low-power "sleep" mode for intervals when it is pre-determined from a setting or from previous received datasets that there will be no transmissions by the sending channel.
Claim 10. A spread-spectrum communication system as in Claim I but where the time windows for the Word Group modulation method are width restricted in order to reduce the noise by lowering the frequency with which noise or interference pulses are detected in preference to a wanted signal namely a particular Bit Group.
Claim 11. A spread-spectrum communication system as in Claim I where provision is made that if the data decoded from Bit Grouping reveals that the transmission is not intended for a particular recipient then the said recipient switches to a low-power "sleep" mode without further decoding.
Claim 12. A spread-spectrum communication system as in Claim I employing a multiplicity of Transmitter components to drive a multiplicity of antennae with trains of impulses that are delayed with respect to each other but arranged such that all impulses occur within one time window of the Word Group modulation method for the purposes of overcoming nulls in the directivity lobes of said antennae.
Claim 13. A spread-spectrum communication system as in Claim I and utilising a multiplicity of Receivers and Antennae in the manner similar to that described in the previous Claim 12 in order to eliminate nulls in the said Antenna sensitivity lobes by electronically delaying the received signals so that they are presented in the same de- modulation window of the Word Encoding.
Claim 14. A spread-spectrum communication system as in Claim I which has at the receiving unit a means of measuring the received impulse amplitude with this information then being used to signal to the transmitter to cause an adjustment of the transmitter pulse amplitude to bring the received signal within an optimal band and thereby optimising pulse power and reducing unwanted emissions.
Claim 15. A spread-spectrum communication system as in Claim I and I I where a number of transmitters and antennae are used and the receiver selects the antenna giving best signal quality for the link with other transmitters being then instructed to tad off Claim
16. A spread-spectrum communication system as in Claim I where impulses of both negative and positive polarity are detected by the demodulator and the system can respond' to either type.
Claim 17. A spread-spectrum communication system as in Claim 2 where communications are acknowledged by the receiving entity sending a copy of the received data back to the originating entity as confirmation of the correct receipt of the information.
Claim 18. A spread-spectrum communication system as in Claim 1 and 14 where data flow control is achieved by means of the acknowledge signal of Claim 13 in that fresh data is only sent by the transmitting entity when an acknowledge is received from the receiving entity so that overflows of data at the data receiving entity are avoided.

Claim
l 9. A spread-spectrum communication system as in Claim I and which uses a plurality of transmitters and receivers to improve the communication link.
Claim 20. A spread-spectrum communication system as in Claim l where additional spreading or an encryption scheme is effected by means of the Bit Group pattern for successive Word Groups changing in some manner known only to Sender and Receiver to make interception by third parties more difficult such as for example a pseudo-random sequence with a seed number known only to Sender and Receiver.
Claim 21. A spread-spectrum communication system as in Claim 1 where additional spreading or an encryption scheme is effected by means of the recurrence frequency of Bit Groups in successive Word Groups changing in some manner known only to Sender and Receiver such that interception by third parties is made more difficult.
Claim 22. A spread-spectrum communication system as in claim l where impulses are detected by a simple pulse height disciminator where the shape of the impulse waveform is unimportant provided it is of sufficient amplitude to be detected above the system noise floor.
Claim 23. A spread-spectrum communication system as in claims 1,2 and 3 where the transmission medium is not free space but another medium supporting the propagation of waves be they guided or unguided and in any frequency band of the electromagnetic, acoustic or ultrasonic spectra.
Claim 24. A spread-spectrum communication system as in claims 1,2 and 3 where a component of the Bit Group or the Word Group conveys information to indicate if logic signals are to be regarded as normal or inverted for the purposes of introducing a bias in the number of logic bits of a particular type.
Claim 25. A spread-spectrum communication system as in claims 1,2 and 3 where spurious or "dummy" impulses are introduced which form no part of the intelligence to be communicated but are intended to confuse a third party intercepting the transmission.
Claim 26. A spread-spectrum communication system as in claims 1,2 and 3 where some Bit Groups are utilised in broadcast fashion to act as a "paging channel" to address many modules simultaneously.
Claim 27. A spread-spectrum communication system as in claims 1,2 and 3 where the information conveyed is constructed such that the receiving party can measure the frequency of the impulse signals and so deduce the sending party's clock frequency and so synchronise or apply a correction to its own clock frequency.
Claim 28. A spread-spectrum communication system as in claims 1,2 and 3 where the propagation medium is an optical fibre or a plurality of optical fibres and the waves are electromagnetic waves at optical wavelengths.
Claim 29. A spread-spectrum communication system as in claims 1,2 and 3 where the propagation medium is an electrical signal cable and the waves are galvanic currents.
Claim 30. A spread-spectrum communication system as in claims 1,2 and 3 where the propagation medium is free space and the waves are radio waves.
Claim 31. A spread-spectrum communication system as in claims 1,2 and 3 where the waves are ultrasonic or sonic waves in a solid, liquid or gaseous supporting medium.
Claim 32. A spread-spectrum communication system as in claims 1,2 and 3 where the information is conveyed by microwaves in waveguides or in free space.
Claim 33. A spread-spectrum communication system substantially as described in the

description and diagrams of preferred embodiments.