MXPA98002470A - Switched radio communications networks distribute - Google Patents

Abstract

A plurality of transmitting and receiving stations are provided in randomly distributed locations of the switching circuitry within the same stations to route calls between stations in the network, using other stations in the network for retransmission of said calls, when necessary. To this end, each station (20) incorporates a call routing control unit that acts to select an additional station (21) into which a call from a source or origin S to a destination D will be transmitted for the purpose of retransmitting the call. The call routing control unit transmits an interrogation signal that will be received by other stations in the network, within the range of the transmitting station (20) and the call routing control unit of each of the other stations in the network. network, within range of the transmitting station (20) and the call routing control unit of each of the other stations (21) transmits a recognition signal when the station (21) is available to retransmit a call, in response to the interrogation signal received from the station (20), the acknowledgment signal will be transmitted after a delay indicating the conformity of the station (21) to retransmit the call to its intended destination. The call routing control unit of the station (20) then selects a station (21) to retransmit the call based on the reception of a station recognition signal, after the minimum delay.

Description

SWITCH OE NETWORKS SWITCHED DISTRIBUTED CIRCUIT FIELD OF THE INVENTION This invention relates to distributed circuit switched telecommunication networks and, more particularly, but not exclusively, relates to transmitting and receiving stations for said networks in which a plurality of these stations are provided at randomly distributed locations and wherein the switching circuitry is provided within the same stations to route calls between the stations of the network using other stations of the network for the retransmission of these calls when necessary.

BACKGROUND OF THE INVENTION In many countries, although there may be a telephone service to the towns and some major cities, the majority of the population does not have effective access to telephones. There is a need in these countries for a telephone network at a density such that substantially all the population lives within a few kilometers of a public telephone. However, this would require the installation of a network comprising P1156 / 98 X a large number of widely distributed telephones, which would be prohibitively expensive if a conventional wired telephony system were used. The Article "A Distributed Rural Radio System for Developing Countries", S.A.G. Chandler, S.J. Braitwait, H.R. Mgombelo et al, Fourth IEE Conference on Telecommunications, IEE Conference Publication No. 371, April 1993, describes a rural radiotelephony system, which, by virtue of its network structure without exchanges, is ideally suited to provide a service of basic telephony to sites with a wide separation.

SUMMARY OF THE INVENTION This radiotelephone system uses a network of cooperative radio nodes that do not require a telephone exchange or an interconnection infrastructure. Each node consists of a transmitting and receiving station comprising two digital transceivers of a single channel, at least one telephone interface and controllers containing software that implements a protocol to effect the required communication control. The links between nodes are fixed capacity links (as opposed to packet switched links or statistically multiplexed links) as required for the duplex conversation in telephone traffic. Each P1156 / 98MX transmitting and receiving station comprises a digital radio unit with solar energization with one or more telephones connected to it. Calls within a reasonable range (50 kilometers or similar in reasonably favorable terrain) are made by direct communication from station to station. However, beyond this scope, calls must be retransmitted by other stations within the network which are not being used at the time of making the calls. Calls outside the area served by the network or those that require an excessive number of stages or retransmission jumps can be routed through gateway nodes to the public service telephone network. The most similar systems currently in use are packet radio systems. However, these systems typically use a single statistically multiplexed radio channel, which usually results in a much lower data rate and always have an unacceptable delay variation for the duplex conversation. It is an object of the invention to provide an improved transmitter and receiver station for a distributed circuit switched telecommunications network. The invention is defined by the P115ó / 9tíMX accompanying claims.

BRIEF DESCRIPTION OF THE DRAWINGS In order that the invention is more fully understood, the embodiments of the invention will be described below, by way of example, with reference to the accompanying drawings, in which: Figure 1 is a Explanatory diagram of a distributed circuit switched telecommunications network; Figure 2 is a block diagram of a transmitting and receiving station in said network; Figures 3 and 4 are explanatory diagrams illustrating call routing methods in the network according to the invention; Figures 5, 6, 7 and 8 are explanatory diagrams illustrating the methods of call interruption in the network according to the invention; and Figures 9, 10, 10a, 11 and 12 are explanatory diagrams illustrating preferred features of the circuit used in a transmitting and receiving station according to the invention.

DETAILED DESCRIPTION OF THE INVENTION Figure 1 is a diagram showing the location of nodes in a hypothetical network of P115Ó / 98MX distributed circuit switched telecommunication comprising a series of fixed nodes 1 located randomly in which the transmitting and receiving stations are located. In addition to the nodes 1 of the network, several gateway nodes 2 are shown that provide access to the call to the public service telephone network in which telephone communication is carried out in a conventional manner by means of wired links under centralized control of a telephone exchange. As indicated by broken lines 4 of the Figure, calls can be made between nodes 1 of the network or between a network node 1 and a gateway node 2, either directly when the nodes are close enough to each other by means of other nodes 1 that are used to retransmit the calls. Figure 2 is a block diagram of a transmitting and receiving station 6 comprising two transmit / receive antennas 7,8; two digital transceivers of a single channel 9, 10; at least one telephone interface 11, 12 and associated telephones 13, 14, a transceiver interface 15 and a control unit 16 for effecting communication control between stations within the network. Each station 6 can be used to terminate up to two calls, that is, two calls that are made simultaneously using telephones 13 and 14, P115Ó / 98MX alternatively, retransmit a single call using simultaneously the two transceivers 9 and 10 to receive and retransmit the call information in both directions. Transceivers 9 and 10 normally use separate frequency channels, although, in an area small enough so that the bias in time is not problematic, an ultiplex by time division could be used. Normally, the two channels are selected from a set of 200 available channels although in certain situations a larger number of channels could be useful. The routing of calls within the network is based on the geographical location of the nodes in the network, rather than the provision of routing tables stored within the stations, as this provides a much better performance when there is no centralized office. The implementation of the routing and the initial establishment of the call are carried out using a specific call channel dedicated for this purpose in which the information is transported using asynchronous packets. Channels that transport separate traffic by which the conversation information is received and transmitted can be used either for fixed frame or asynchronous packet circuit switched transmission formats P1156 / 98MX depending on the system protocol and user requirements.

Call routing The routing of a call from a source station or origin S, in which the call is made, to a destination station D, to which the call is routed, is controlled by an appropriate routing algorithm implemented by cooperation between the control units of the source and destination stations and any other stations used to retransmit the call. The routing algorithm establishes a series of communication links that start from the source station F and end at the destination station D using an iterative process, if the destination station D is not within the reach of a single hop of the station Source S. The simplest method for such routing, shown in Figure 3, includes determining at each stage of the iteration the station within the radio range of the previous station that is closest to the destination station D. If the Nearest station is not closer than the previous station used in routing, then the path or route is blocked. Figure 3 shows this method applied to the retransmission of a call by means of two stations 20 and 21, the radio ranges associated with the P115ü / t'8MX source station S and stations 10 and 21 are shown by circles 22, 23 and 24. Next, the manner in which this routing algorithm is implemented by the control unit of each station will be described below. The control unit of the previous station will transmit an interrogation signal in the form of a CQL message that will be received by all the stations available for retransmission within range. If this interrogation signal is received by the destination station D, the control unit of the destination station provides an immediate recognition signal to indicate that the call can be retransmitted from the previous station directly to the destination station. The other stations receiving the interrogation signal provide an acknowledgment signal that is transmitted after a delay that increases with the distance of the station from the destination station D. The amount of this delay is calculated by each station in basis at the distance of the station from the destination station determined either from a list of the locations of the other stations that is stored within the station or from the use of the station numbers indicative of the grid references of the seasons. If the stations operate a CSMA protocol, so that P1156 / 98MX provide a recognition signal if the channel is already in use, most collisions can be avoided. As soon as the previous station receives a recognition signal from another station, it sends a confirmation signal to the selected station. The other stations within the range (but at a greater distance from the destination station D involve a greater delay before recognition) will also receive the confirmation signal and, in this way, the recognition of the interrogation signal will be inhibited, avoiding this way unnecessary congestion in the call channel. This routing algorithm has the disadvantage that it will not find all possible paths and routes and that the calls can therefore be unnecessarily blocked. For example, referring to Figure 4, a call may be blocked at station 21 due to the presence of an obstacle 26, such as a mountain or simply an area where there are no stations available to retransmit the call. One way to improve this situation is to route the algorithm slightly in a generalized way. Instead of selecting the station closest to the destination to retransmit the call at each stage, the next relay station can be selected at each stage so that P1.5Ó / 98MX will maximize the probability of completing the road or route to the destination. The probability of completing the road or route from any station varies inversely with the distance from the station in the absence of any knowledge that the road or route is blocked. However, if it has been determined that the path is blocked in station 21 of FIG. 4, so that there are no stations available at suitable locations to act as relays in accordance with the original algorithm, then a modified algorithm may be used to go back one step to station 20 and modify the probability distribution with the knowledge that station 21 is blocked. This results in the station 21 or the stations near the station 21 being less likely to be chosen, thereby enabling an additional station 25 to be selected for retransmission of the call in order to implement a path to the station. destination surrounding or deriving the obstacle 26. If the location of the station 21 is known by the additional stations, such as the 25, which can potentially be involved in the retransmission of the call, as will be the case if these additional stations reach to listen or detect a signal from station 21 that confirms the return or retraction, the P1156 / 98MX probability of completing the routing from any station or, at least an approximation to it, can be calculated by each of the stations on these bases, and hence, an appropriate delay determined to respond to the signal of interrogation of the previous station in order to implement the routing in the same way as previously.

Call interruption When a station is retransmitting a call between other stations in the network, both transceivers of the station will be in use and, accordingly, the station will not be available to send or receive a new call (which terminates at that station). station) during this retransmission, unless a special call interruption facility is available to allow the re-routing of the existing call to allow a new call to be made. Figure 5 shows in diagrammatic form the information transported by the two traffic channels 27, 28 and the call channel 29 of a station that is retransmitting a call between two other stations. During a normal telephone conversation, although a complete transmission path will be provided P1156 / 98MX duplex between the two telephones, normally only one part will speak at a time and there will be silence in both directions for a significant fraction of the transmission time, for example, during spaces or separations between words and syllables. There is no need to transmit information during these periods of silence, in addition to information on the duration of silence. In the described system, the information of the conversation is divided into 10 ms frames that are actually transmitted in packages of approximately 4 ms in duration that allows the information in the two directions to be transmitted in alternate form. When a station is retransmitting a call (or simultaneously terminating two calls), the two transceivers are used to communicate with the other two stations using different frequencies. As shown in Figure 5, the transmitters of both transceivers are synchronized to transmit simultaneously, so that, as the conversation information is being transmitted in one direction to a station in the first traffic channel, the conversation information is transmitting in the opposite direction to the other station in the second traffic channel. The two transceivers are available to receive the information in intermediate periods 30 PU.56 / 98 X between successive succession periods 31, guard intervals 32 that will be provided between the transmission and reception periods in order to avoid any overlap of these periods. If both transceivers were busy receiving signals in one or the other direction during the reception periods, it would not be possible for the relay station to receive a call made to it. However, during each period 30, each transceiver is actually receiving conversation information only during a proportion of the time and, there are periods of silence. In these short silence periods, null conversation packets can be received indicating that the receiver is available to receive the interruption signals from a station that wishes to call, so as to allow the station to interrupt the retransmission of the existing call to accept the new call . The re-routing of the previously retransmitted call can be carried out automatically, so that there will be no separation or significant space in the transmission of the call. A possible method by means of which the retransmitted call can be interrupted, will be described with reference to Figure 6, which shows a station that is in use to retransmit a call that will be effected by another station that transmits interrupt packets 34 in P115ó / 9ß X the call channel. During the transmission periods 31 and the periods 33 during which the conversation information is being received, none of the transceivers is available to receive the interrupt packets 34. However, the reception of a null conversation packet 35 in one of the traffic channels indicating a period of silence, cause the transceiver to communicate quickly to receive information. in the call channel and, thus allow the existence of an interrupt packet 34 to be detected in the call channel and to be recognized by the transmission of an acknowledgment signal 36 in the call channel. Since a call to any station has priority over an existing call will be transmitted by that station, the acknowledgment signal 36 will indicate the call of the telephone at the called station and, will result in the transmission of the conversation starting with the calling party using one of the traffic channels if the call will be accepted or taken by the phone that will be off-hook. This causes the existing relay link to be broken and the alternate routing of the existing call to be implemented. The retransmitted call can be interrupted in a similar way if the telephone will be off-hook to make a call to another station. Although the brief interruption in the transmission P115ó / 9aMX of the existing call in this method should not be serious, it is possible that the existing call could be interrupted. In this method, the routing of the new call to the station that is currently retransmitting a call proceeds in accordance with the routing algorithm described above until the penultimate station is reached. As with other stations, this penultimate station incorporates a table of neighbors indicating those stations that are within the guaranteed radio range as determined by the geographical distance of those stations and the pick up of signals from said stations and, the list of neighborhood of the penultimate station will include the destination station. The penultimate station will try to reach the destination station by repeatedly calling on the call channel and, if the call is accepted by the destination station, an acknowledgment signal indicating the start of the call or dialing will be returned. the destination station. Otherwise, a signal is returned or returned from the destination station indicating that it is busy. This method has the difficulty of requiring very fast channel switching which results in an increase in the complexity of the circuit to provide the required switching frequency and may encounter difficulties due to the stabilization time P1156 / 98MX of the demodulator levels. Figure 7 illustrates an alternative interruption method that overcomes the need for fast channel switching by replacing the neighborhood tables referred to above contained by the stations with tables of distribution signal registers or channel distribution sent between stations when routes are established for existing calls. In this way, the routing of a call to a destination station that is currently retransmitting a call on its traffic channels, the penultimate station will refer to the registers of other listened stations and the frequencies of the channels to which they have been sent. Distributed the transceivers of the stations listened to. These registers are entered when their frequency installation commands reach them to listen to the station and, they are deleted or suppressed when subsequent attempts to call the corresponding station on that channel fail or, the corresponding station that calls another station on that channel is heard. the call channel. The penultimate station then attempts to reach or connect to the destination station by repeatedly calling on the traffic channel or on each traffic channel of the destination station, in accordance with the previously recorded channel distribution information. During the use of the channel P115Ó / 98MX called to transmit or receive the conversation information, reception of the call signal in accordance with the CSMA protocol will be inhibited. However, upon receipt of a null conversation packet 35 indicating a period of silence in the called channel, the channel will be available to receive an interrupt packet 37 from the calling station and, this will then allow the transmission of a signal of acknowledgment 38 on the called channel indicating the beginning of the call or dialing at the called station and the placement or installation of the call in the manner already described. The frequency stabilization time requirements in the last channel of this method are not demanding. However, there is still the possibility that the existing retransmitted call could be interrupted. Next, a modification of the interruption method described above will be described with reference to Figure 8 which substantially eliminates the possibility that the existing retransmitted call will be cut off during the interruption. Figure 8 shows in diagrammatic form a call from a source station S to a destination station D that will be transmitted by several other stations including a station R to which a new call will be directed from a station 41 by means of the relay stations including to the penultimate P115Ó / 98MX station 1. As previously indicated, periods of silence or spaces in the data transmission are recognized by the terminal stations and, special null conversation packets are transmitted, shorter than the normal conversation packets used to carry signals ordinary. The null conversation packets contain information that allows the listening stations to identify which station has transmitted the null conversation packet and to synchronize with the null conversation packet, as well as to prepare the normally receiving station to listen for interruptions during the empty rest of the time slot. Since the penultimate station I can only be sure to hear the signals from station R and not necessarily from station A or station B adjacent to station R in the opposite direction, the time of the interrupt packet of station I should be able to be determined only by the signals from station R. A good method to achieve this would be to use the null conversation packets from station R to synchronize to station I but, that station I does not transmit an interrupt packet until station R stop transmitting null conversation packets This means that the subscriber at one end has started talking making it more likely that the other subscriber has P115U / 9HMX stopped talking Station I will continue trying until a connection is made. If a null burst is heard from station A, this problem does not arise and the interrupt packet is transmitted immediately. When a call request is received from station R from the penultimate station I or, when a telephone from station R is off-hook to initiate a call, a message is passed to an adjacent station A on the call route between the source station S and the destination station D that instructs the re-routing of the call. This will cause one of the transceivers of station A to temporarily interrupt the transmission to station R in order to transmit a CQL message in order to set an alternative route for the call. Station A then waits for a response from a potential replacement relay station using the normal routing procedure already described, with the exception that the transceiver of station A returns to transmit to station R in the traffic channel after sending the CQL message, so that additional communication with the potential replacement of the relay station should be made in the traffic channel instead of in the call channel as would normally be the case. The routing procedure is then continued in the normal way until a P1150 / 98MX connection to the destination station D or to a penultimate station 45 if both transceivers are used at the destination station D. Only when confirmation is received from the R station that this connection has been made is interrupted the retransmission of the call by the station R and the initiation of the dialing at the R station is made (or the routing of the call begins if a call is made from the R station). While this method avoids the loss of existing retransmitted calls, the method has a slightly adverse effect on the availability of being called by the same stations, which are currently retransmitting calls since these stations will not be available to make or receive calls. if they are the only stations capable of acting as relays of existing calls. With this method, the existing retransmitted call may suffer a transmission cut in an address equal to twice the frequency stabilization time plus the time to send a CQL message and receive a response to it. However, this will still be less than the break or cut of the transmission provided by the interruption of the call in the other described methods.

P1156 / 98MX Mobile Stations When mobile transmit and receive stations are used in the network, each mobile station is associated with a base node in which a normal fixed transmitter and receiver station is located. When the mobile station is on, it transmits log packets to its base at regular intervals, say once every ten minutes, these log packets will be retransmitted through the normal fixed stations within the network when necessary. The transmitted registration packets, when retransmitted by other stations, will contain information identifying the first station used to effect this retransmission and, thus, indicate to the base station the approximate location of the mobile station, i.e. the location within of the reach of the first relay station. This record of the location of the mobile station will be updated regularly with the reception of the registration signals by the base station. All call requests to the mobile station are initially routed to the base station which then retransmits these call requests to the mobile station in the normal packet radio mode, using, when necessary, the first relay station (which is nearby). to the mobile station according to P1156 / 98MX is indicated by the registration information) and possibly other stations for retransmitting the call request to the mobile station. The reception of the call request retransmitted in this way by the mobile station results in dialing or calling at the mobile station, when the telephone set at the mobile station is off-hook to accept the call, the routing of the return call is implemented. to the calling station in accordance with the routing algorithm described. If the calling station is closer to the mobile station than the base station, this routing would often result in the call not being retransmitted by the base station. It should be noted that this routing in response to picking up the telephone to accept a call is a preferred feature used in the network, if the call routing had been initiated to reserve capacity in the stations that will be used in the retransmission of the call at the moment in When the call is initiated, unnecessary network congestion would be caused by the fact that these stations would be reserved for use while the telephone at the called station is ringing and remains unanswered.

P115Ó / 98 X Transmission / reception switching In the system described, each transceiver transmits and receives on the same frequency and it is therefore essential to suppress the transmission signal by a large amount in order to ensure that this will not interfere with the reception of the signal that will be received by the transceiver. The output of the transmitter must therefore be attenuated in much more than the loss of the radio transmission path, that is, of the order of 140 dB. When the switching from the transmission mode to the reception mode must be fast, it is significantly less than one millisecond, it is difficult, without or impossible, to turn the oscillators off or on or to shift their frequency as is normally done in the transceivers to press to speak. Therefore, this places severe requirements on the switching circuits that will be used for switching between the transmission and reception modes in these transceivers. In accordance with the foregoing, as shown in the block diagram of Figure 9, one of the frequency signals that is mixed to produce the carrier frequency signal of the transmission or, the same carrier frequency signal in the case of An FM transmitter is derived from a frequency divider 50 comprising a digital bistable circuit (a jog circuit). These P1156 / 98MX digital bistable circuits can be turned off instantly using a logic gate. In the transmission mode, the frequency signal of a frequency source 51 is divided in two by the frequency divider 50 and is supplied to a modulator 52 in which the signal of the baseband for the transmission is used to modulate the carrier signal emitted by the frequency divider 50 to produce a modulated signal for amplification by an amplifier 53 and transmission by means of the transmit / receive switch 54 and the antenna 55. When the transceiver will be used in the reception mode , the transmit / receive switch 54 is switched to the reception position, as shown in Figure 9, by the application of an appropriate control signal and, at the same time, the control signal is applied to an inhibition input. connected to a logic gate within the frequency divider 50 so that the transmit carrier frequency signal is inhibited. The fact that the carrier power is not present in any of the input signals to the circuit and is only generated by the non-linear flip-flop operation of the flip-flop means that the carrier signal is completely turned off by this operation. This does not eliminate the need to turn off the amplifier P115Ó / 98MX 53, since otherwise this would still amplify any noise signal but, the requirements to turn off the amplifier 53 are much less severe than would otherwise be the case. The signal received from the same frequency is supplied by means of the switch 54 to the reception amplifier 56 and from there, to the reception section of the transceiver.

Transmitter modulation ramp Since the information is transmitted in this system in short bursts, measures must be taken to prevent the transient envelope variation of each burst from causing the spectral dispersion that could result in interference for users of the adjacent channels. To this end, the system uses linear modulation of the carrier frequency in order to provide optimum bandwidth efficiency and provides an up and down ramp with very short rise and fall times at the beginning of the transmission bursts. in order to minimize the spectral dispersion. In opposition to the more established constant envelope schemes, linear modulation systems have allowed orthogonal modulation (in which each symbol waveform can be received P1156 / 98MX independently of all other symbol waveforms) to approximate the theoretical limits of bandwidth efficiency. However, the use of frequency modulation-type envelope conformation for these systems has been claimed to cause substantial spectral dispersion of the linearly modulated burst mode signals. While all the modulation in amplitude of the linearly modulated signal t associated with the ascending and descending ramp would seem to necessarily cause some spectral dispersion, it is considered that these known systems use an incorrect approximation to the problem, leaning very strongly to the approach that must take with constant envelope schemes. The modulation system according to the invention uses an approximation which results in very short ascending and descending ramps and, in principle, without any spectral dispersion. In contrast to the known system in which the ascending and descending ramping is performed by the full amplitude modulation of the linearly modulated signal, this system includes the generation for each symbol of the data that will be transmitted from a symbol waveform that It has portions of ascending ramp and descending ramp of limited duration. A series of waveforms from P11SÓ / 9HMX symbol overlapping in time such as for example the warms 61, 62, 63 and 64 shown in Figure 10a, representative of the digital symbol frequency 1101, combine with a transverse filter 70 as shown in Figure 10 in order to produce a combined warm 60 having ascending ramp and descending ramp portions 65 and 66 of limited duration. The generation of the combined warm by convolving a pulse stream representing the data sequence with warms of individual symbols can be implemented using stored and precalculated or calcified warms in real time using a digital signal processor followed by an anti-pseudonym filter. The output of the cross filter 70 is then supplied to a modulator 71 in order to modulate a frequency carrier signal fc to produce the linearly modulated signal required for transmission. The symbol warms are theoretically of infinite duration such that an approximation is made by truncating each warm to provide portions of ascending ramp and descending ramp of limited duration. However, the power associated with the ascending ramp and descending ramp portions can be made extremely small with a length of If only a few symbol periods are truncated, the truncation will be imposed by considerations of causality since the warms are usually symmetric. The power spectrum of a signal resulting from an uncorrelated sequence of data would be the same as the energy spectrum of the individual symbol pulses. The correlation of the data sequence would cause some attenuation at certain frequencies but would not cause the spectral dispersion of the signal. If each transmitted burst is generated in accordance with the system of the invention by convoking the sequence of finite-length data with the symbol warm, the spectrum of the burst would have the same spectrum as the continuous signal. Although the length of the symbol warm would cause the power to be transmitted for a few symbols before the first present present symbol (the center of the impulse) this is all the ramping that is necessary. In practice, this only requires precursors of two or three symbols that is much less than what is required in conventional schemes.

Synchronization of the receiver The performance of all digital receivers depends critically on the synchronization of a signal P1156 / 9BMX clock used to determine the sampling time of the received and filtered signal. The conventional technique for controlling the clock signal uses the zero crossings of the received warm to synchronize a local timing oscillator using a phase locked circuit. In this case, the bandwidth is a compromise between the smoothing of the random time variations of the zero crossings and the speed at which the phase locked circuit can achieve the assurance. Random time variations are caused not only by the effects of noise but also by the fact that zero crossings occur at times or times that depend on the sequence of data that will be transmitted. In this way, performance in this system is limited even in the absence of noise. The orthogonal modulation of the transmitted signal means that the symbol warms obtained from the signal received after the filtering are such that each symbol warm determines the value of the signal at one, and only at one, point. sampling in a regular sequence of sampling points. Although the symbol warms overlap, all but one of the symbol warms will be zero at each sampling point. Examples of the modulation formats for which this is true are QAM, QPSK, and pi / 4 DQPSK. In P115Ó / 98MX could also include the filtered signals from the base station sent over an FM or PM channel. If the limiter discriminator integrator techniques are used to receive any form of orthogonal PSK signal or correct amplitude scaling signal (AGC) used with QAM modulation, the x signal in the absence of noise will have a value (xQ, x? ), at each sampling point selected from an infinite set of values (1,1), (1,0), (0,1) and (0,0), the particular value (xQ, Xi), determines the received symbol. As shown in Figure 11, which is a graph of the signal x versus time t showing a change in the value of x between two sampling points, the signal x will be found in an error in a quantity dv if the time real sampling t4? ? ? ? ? , is in a relative error with the correct sampling time Tc in an amount dt. If the error v and its derivative of the time dv / ct (which is the same as the time derivative dx / dt of the signal) is measured, the timing error can be estimated as dv / (dv / dt) and, this can be used amount to adjust the phase of the timing clock signal. However, in the presence of noise, disproportionate errors will occur if dv / dt is small. Therefore, it would be preferable to scale the adjustment amount by a factor of | dv / dt |, the magnitude of P115Ó / 98MX the slope of the signal, which results in an adjustment amount of k.dv / sgn (dv / dt) where k is a weighting factor that will be larger the smaller the slope and sgn (dv / dt) represents the sign of dv / dt, so that the adjustment amount can also be written as k.dv / sgn (dv / dt). The time of the clock signals based on this phase adjustment is superior to the conventional zero crossing technique, since, although the adjustment will be degraded by additive noise, it will not be affected by the data sequence in the r & amp; Thus, the main cause of the fluctuation in time is eliminated. Figure 12 is a block diagram of a pi / 4 DQPSK receiver that uses this phase adjustment to control the sampling time. In this receiver, a received signal x is provided by a reception section consisting of an antenna 80, a tuning circuit 81 provided with a local oscillator, a tied filter 82, a limiter 83, a discriminator 84 and an integrator of symbol 85. The signal x is supplied both to an analog-to-digital converter 86 and to a differentiator 87 followed by a comparator 88. The output of the converter 86 is a digital word, of which, the most significant bits (x0, X? ), are representative of the value of the symbol received provided that the escalation is P 15Ó / 9HMX is correct and is emitted from the circuit as the received data (x0, i), while the least significant bits from the output of the converter 86 constitute the complement of two of the error quantity dv. The output of the differentiator 87 is the derivative of the time dv / dt of the signal error, which is equal to the time derivative dx / dt of the signal and the output of the comparator 88 is a digital signal representing sgn (dv / dt) . The process of multiplying the error of the signal dv by sgn (dv / dt), corresponds to deducting the complement of two of dv if dv / dt is negative. Thus, the multiplication of dv by sgn (dv / dt) can be approximated by supplying the least significant bits of the output of the converter 86 and the output of the comparator 88 to a set of exclusive OR gates 89 to provide an output corresponding to the complement of dv that approaches the complement of two (only with an error in the least significant bit). The output of the gates 89 is supplied to a counter 90 provided with a crystal frequency reference oscillator 91 which in turn supplies the clock signal to the analog to digital converter 86 counting down from a high frequency reference provided by the oscillator 91. The counter is reset by applying a reset signal supplied by an inverter 92 that receives an input P115 / 98MX from the gate 89 associated with the most significant bit of the least significant bits emitted from the converter 86 to control the clock signal. The gain and the errors of displacement in the signal x can also be a main cause of problems when using the DQPSK pi / 4 limiter discriminator reception as well as for multilevel QAM schemes and additional parts of the receiver circuit are provided to compensate for said errors . The displacement error can be approximated by the least significant bits of the output of the analog-to-digital converter 86, ie, those bits other than the data bits, so that offset error compensation can be obtained by using these bits to adjust recursively or recurrently the level of displacement. This can be done by supplying the least significant bits to an accumulator or an ascending / descending counter 93 that counts up when the estimated displacement is low and counts down when the estimated displacement is high and, supplying the resulting signal to a digital converter to analog 94 and to a potentiometer 95 that controls the displacement of the signal x. A more sophisticated system would be to use all the error bits and adjust the displacement in an amount proportional to the error.

P115Ó / 98MX In order to provide gain compensation, use is made of the fact that the gain error can be approximated by the error provided by the least significant bits of the output of the analog-to-digital converter 86 divided by the voltage corresponding to the symbol in question and, to the fact that the performance will be very little different if instead only the sign of the voltage of the symbol is used. This approach is easy to implement since division by the sign of the symbol can be achieved by passing the most significant bit of the output of the converter 86 through a set of exclusive OR gates 96 with the least significant bits as shown in the Figure . The output v.sgnx = v.x0 of the gates 96 is supplied to an accumulator or to an up / down counter 97 which is connected to a pointer meter 98 to effect the required compensation of the gain error. However, the fact that the system could adapt to a false null point should be considered if the permissible range of the parameters were very large. The same principle, that is, the minimization of the error in the sampling time, can be used in demodulators based on the digital signal processing with the difference that, since the output of a differential modulator signal P115Ó / 98MX would be in Cartesian coordinates rather than in polar coordinates, some form of approximation would be needed to avoid unnecessarily long calculations. However, it was found that even the simplest approximation using a constant magnitude of adjustment performs adequately in practice.

P115Ó / 9TMX

Claims (8)

1. NOVELTY OF THE INVENTION Having described the present invention, it is considered as a novelty and, therefore, the content of the following CLAIMS is claimed as property. A transmitting and receiving station for a telecommunications network in which a plurality of said stations will be provided in randomly distributed locations and where the switching circuitry is provided within the same stations for call routing between stations of the network using other stations of the network to retransmit these calls when necessary, characterized in that the station incorporates a means of distribution of calls that acts to interrupt an existing call that will be retransmitted by the station for the purpose of receiving a new call at the station , coming from an additional station in the network and comprising: (a) a call monitoring means for receiving, during time intervals between the retransmission of the voice or data signals within the existing call, an interruption signal which indicates to the station a call request from the station additional; and (b) a means of accepting calls so that P115Ó / 9TMX the station interrupts the retransmission of the existing call, in response to the interrupt signal, to allow the new call to be received from the additional station. A station according to claim 1, wherein the call interruption means incorporates a re-routing means which, in response to receiving the interrupt signal, selects an internal route for retransmitting the existing call and, in the case of If an alternative route is available, re-route the existing call to allow the new call to be received from the additional station. A station according to claim 1 or 2, wherein the call interruption means includes a null transmission means for transmitting null packets indicative of gaps or gaps in the voice or data signals that will be transmitted within the existing call , the call monitoring means is adapted to receive the interruption signal during the transmission of the null packets. 4. A station according to claim 3, wherein the call monitoring means is adapted to receive the interrupt signal only after the transmission of a null packet sequence has ceased. P115Ó / 98MX 5. A mobile transmitter and receiver station for use in a telecommunications network in which a plurality of transmitter and receiver stations will be provided in randomly distributed locations and where the switching circuitry is provided within the same stations for routing calls between stations of the network, using other stations in the network to retransmit these calls when necessary, the mobile station comprises: (a) a recording means for transmitting the registration signals to a fixed base station for transmission and reception associated with the station mobile that uses other fixed stations in the network to retransmit the registration signals when necessary, such that the base station receives an indication of the approximate position of the mobile station from the location of the first fixed station used in the retransmission of the registration signs; and (b) means for receiving calls to receive a call from an additional station in the network retransmitted by means of the base station and routed to the mobile station by the base station, based on the approximate position of the mobile station indicated by the registration signals and / or initiates the routing of the call back to the additional station following the P1156 / 98MX acceptance of the call. 6. A transmitting and receiving station for use in a telecommunications network, comprising a signaling means for transmitting and receiving signals on the same frequency, a transmission means for transmitting signals by means of the signaling means, a receiving means for receiving signals by means of the signaling means and, a switching means for switching to the signaling means between the transmission means and the receiving means in response to a control signal, wherein the transmission means incorporates a frequency source for supplying a frequency signal, a bistable circuit means for producing, by frequency division of the frequency signal, a carrier frequency for the signals that will be transmitted and a means of inhibiting to turn off the bistable circuit means to inhibit the carrier frequency in response to the control signal. 7. A digital transmitter for transmitting digitally modulated signals in short bursts, comprising a symbol waveform generating means for generating, for each symbol of the digital data to be transmitted, a symbol waveform of limited duration, a waveform convolution medium combine a series of these wave series of P115o / 98MX symbol, such that each waveform is delayed with respect to a previous waveform and overlaps with it, so as to produce a combined waveform representative of a sequence of digital data symbols that will be transmitted and a signal modulation means for linearly modulating a carrier waveform with the combined waveform so as to provide the modulated signals for transmission in short bursts without producing a substantial spectral dispersion but, without using separate waveforms of ascending ramp and descending ramp. 8. A digital receiver for receiving digitally modulated signals, comprising a receiving means for receiving the digitally modulated signals and for producing from the same symbol signals indicative of the symbols within the transmitted digital data, a sampling means for sampling the symbol signals at the sampling points determined by a clock signal, in order to provide a received data output signal corresponding to the transmitted digital data and an error output signal and, a sampling synchronization means to synchronize the sampling of the symbol signals by the sampling means, adjusting the timing of the clock signal to minimize the error output signal. P115Ó / 98MX SUMMARY OF THE INVENTION A plurality of transmitter and receiver stations are provided in randomly distributed locations within a telecommunications network and switching circuitry is provided within the same stations to route calls between stations in the network, using other stations of the network for the retransmission of said calls, when necessary. To this end, each station (20) incorporates a call routing control unit that acts to select an additional station (21) to which a call from a source or origin S to a destination D will be transmitted for the purpose of retransmit the call The call routing control unit transmits an interrogation signal that will be received by other stations in the network, within the range of the transmitting station (20) and the call routing control unit of each of the other stations ( 21) transmits a recognition signal when the station (21) is available to retransmit a call, in response to the interrogation signal received from the station (20), the recognition signal will be transmitted after a delay indicating the conformity of the station (21) to retransmit the call to its intended destination. The control unit P1156 / 98MX call routing station (20) then selects a station (21) to retransmit the call based on the reception of a station recognition signal, after the minimum delay. P1156 / 98MX