WO1994006217A1

WO1994006217A1 - Spread spectrum communication system with adaptive power control

Info

Publication number: WO1994006217A1
Application number: PCT/GB1993/001882
Authority: WO
Inventors: Richard Frank Ormondroyd
Original assignee: Millicom Holdings (Uk) Ltd.
Priority date: 1992-09-07
Filing date: 1993-09-06
Publication date: 1994-03-17
Also published as: EP0611498A1; GB9218876D0

Abstract

A communication system comprises a number of subscribers connected in common to a base-station (1). Each subscriber may use an interface device termed a subscriber access device too (or SAD). The subscribers and the base-station communicate via duplex of dual simplex spread-spectrum channels, each channel comprising an up-link from a subscriber to the base-station and a down-link from the base-station to the subscriber. The base-station may transmit power control signals to respective subscriber stations on their down-links. The subscriber stations in response to the power control signal from the down-link modifies its transmission signal power following a feed back control strategy. The subscriber stations may also be arranged to transmit power control signals to the base-station to modify the powers of transmissions from the base-station. The communications systems may be implemented as a spread-spectrum cellular radio system. The cells of the system may be made dynamically resizable.

Description

SPREAD SPECTRUM COMMUNICATION SYSTEM WITH ADAPTIVE POWER CONTROL

BACKGROUND TO THE INVENTION

The present invention relates to a communication system in which a number of subscriber stations communicate via a node or base-station on a common transmission medium. Such a system might be used, for example, in a digital radio network of the type proposed for PCN (personal communications network) systems. It has been proposed to use direct-sequence spread- spectrum transmitters for digital radio networks used, for example, for transmitting analogue voice and video information and also digital data. In such a system a number of subscriber access devices (SADs) communicate with the central base-station using, for example, code division multiple-access (CDMA) ormultiple-divisionmultiple-access (MDMA) , which is a novel form of multiplexing related to CDMA and described in the present applicant's co-pending International Application PCT/GB 93/00199 entitled "Communication System" and filed 29th January 1993.

For the reasons discussed in further detail below, when using such direct-sequence spread-spectrum techniques it is desirable to be able to control the transmission power at least of the subscriber stations, and optionally also of the base-station. Such power control makes it possible firstly to prevent the overloading of the base- station receivers by many simultaneous radio transmissions which are co-channel, i.e. all operate on the same allocated frequency band (the so-called near-far effect) , and secondly to limit spectrum pollution. Considering the near-far effect, without power control, transmitters which are nearer to the base-station would be received with a much higher power than transmitters which are further away and so these nearer transmissions would swamp the weaker ones, producing the so-called "near-far" effect and possibly overloading the base-station receiver. The second main reason for using power control of each transmitter is to reduce the total power which is radiated beyond the perimeter of the network cell served by the base-station so that "spectrum pollution" is minimised. This is important for two reasons. Firstly, by reducing the power of level of spurious emissions, the interference range to other radio systems which are sharing the same frequency band is minimised and this improves the overall coverage area of the system, calculated on the basis of mutual non-interference operation. Secondly, it minimises co-channel interference to adjacent cells of the network and reduces the range required before the same spreading code may be reallocated to another network cell.

SUMMARY OF THE INVENTION

According to a first aspect of the present invention, there is provided a communication system comprising a base- station and a plurality of subscriber stations arranged to communicate with the base-station via duplex (or dual simplex) spread-spectrum channels each comprising an up¬ link from a subscriber to the base-station and a down-link from the base-station to the subscriber, in which the base- station transmits power control signals to respective subscriber stations on the down-links, and the subscriber station in response to the power control signal from its respective down-link modifies its transmission signal power according to a feedback control strategy.

In the arrangement adopted in the present invention, the duplex (or dual simplex) channel created in accordance with the media access protocol of the transmission system, is shared for use both for the transmission of power control signals from the base-station to a specific subscriber station and for the transmission of data. Thus, for example, in the situation where in accordance with the protocol a link has been established via the base-station between subscriber station A and subscriber station B the down-links to A and B are effectively shared between the transmission of data and the transmission of power control signals. This makes possible effective dynamic control of the power level of transmissions from each subscriber station active in the network. Preferably the base-station is arranged to monitor the power of transmissions received from the given subscriber station, to compare the received level with the desired level, and to transmit to the subscriber a control signal to modify the power transmission accordingly, thereby providing closed-loop control of the subscriber transmitter power.

Rudimentary power control might be obtained using open-loop control. With this type of control, the range of the transmitter from the base-station receiver is obtained and from this information the required transmitter power is calculated on the basis of some equation relating the transmitter power to range and prevailing atmospheric conditions. This type of control is very stable, but the error in control can be very large if the actual propagation loss differs significantly from the assumed loss used in the calculation, or if the loss increases due to resiting of either the base-station or the SAD for example. The preferred aspect of the present invention rather than using open-loop control, uses feedback control based on the actual received power at the base-station. An error signal representing the difference between the required received power and the actual received power is transmitted from the base-station back to the transmitter as the power control signal. The transmitter uses this error signal to modify its transmitter power so as to minimise the error. The error signal is repeatedly sent back to the transmitter, and in this way the received power remains close to the ideal value, even though the atmospheric conditions may have changed, or in the case of a mobile system the range may have changed.

Preferably the desired level of received power (the reference power level) corresponds to the minimum acceptable level for reception, and the subscriber station responds to the power control signals by adjusting its transmission power so that the received power at the base is substantially equal to the said minimum power level. It is found that when spread-spectrum techniques are used, the optimum system performance is obtained by reducing the transmission power of the subscribers to the minimum level adequate for acceptable reception by the base. This seems counter-intuitive in view of the common practice for conventional radio systems, where each transmitter occupies its own allocated frequency channel. In such conventional systems, reducing the transmitter power of individual subscribers would tend to degrade rather than improve the level of service. However for co- channel multi-access networks such as CDMA and MDMA systems, each user represents a source of noise to all other users. Consequently, if transmitter A transmits at a very high power relative to transmitters B and C, the quality of A's received signal would be very good, but the quality of B's and C's signal would be poor because of the excessive noise and interference that A's transmission generates to the other users. Consequently, optimum noise performance for all uses is obtained when each signal is received at the same strength. The power control signal may be sent in the header of a frame carrying data communication traffic. In this case, it is preferred that the control signal comprises a command to increment or decrement the subscriber transmitter power by a predetermined amount, or by a number of predetermined steps.

Alternatively, the power control signal may be transmitted in a dedicated power control frame, such frames being transmitted intermittently, every few packets of data. For the rest of the time, data communication traffic is carried by the down-link. This arrangement is appropriate where the traffic is "bursty". Then the natural breaks in data transmission are used for the transmission of the power control signals. For data systems, the breaks may be the intervals between sending each character from a terminal keyboard. For voice transmission, the breaks may correspond to the natural pauses in speech (and similarly for video transmission) . With this method, preferably the power control signal includes data corresponding to an absolute power level.

Preferably the reference level used by the base- station in determining the error signal is variable depending upon the traffic carried by the base-station. Preferably the base-station is also arranged to vary its own transmission power levels in accordance with the amount of traffic carried.

This aspect of the invention takes advantage of the fact that when spread-spectrum techniques are used, the noise levels fall as the number of users falls and so the effective receiver sensitivity increases. In the preferred architecture adopted for the transmission system, the base- station is responsible for allocating channels and so has direct control of the amount of traffic. It is therefore straightforward for the base-station to adapt the reference level defining the required transmission powers, in accordance with the number of active channels.

According to a second aspect of the present invention, there is provided a communications system comprising a base-station and a plurality of subscriber stations arranged to communicate with the base station via duplex (or dual simplex) spread-spectrum channels each comprising an up-link from a subscriber to the base and a down-link from the base to the subscriber, in which one or more of the subscriber stations transmit power control signals to the base-station, and the base-station modifies the power level of transmissions to the respective subscriber station according to a closed-loop feedback strategy. Preferably the power control signal generated by each subscriber station is generated by comparing a received power level and comparing the received power level with a reference level, and outputting an error signal back to the subscriber station accordingly.

This aspect of the invention enables the SADs to monitor and control the power of transmissions received by them from the base-station. The sensitivity of a given SAD depends upon the number of interferers in the cell. For SADs near the base-station, their sensitivity will be relatively low, due to the high levels of interference from base-station transmissions to more distant SADs. In contrast, the more distant SADs receive less interference and so have a relatively higher sensitivity. The base- station can therefore use lower power for transmission to these more distant receivers. This has the added advantage of improving conditions for the nearer SADs. This aspect of the invention, which is preferably used in combination with the first aspect, makes possible the use of dynamically resized network cells, whereby as the number and location of subscribers in the system changes, the cell size is modified to optimise the base-station transmitter power in relation to the number of subscribers.

According to a third aspect of the present invention, there is provided a cellular spread-spectrum radio communication system, in which the cells making up the system are dynamically resizable. According to a fourth aspect of the present invention, there is provided a spread-spectrum cellular radio communication system in which the base-station is arranged to vary the power of transmissions in the system depending on the level of traffic. Preferably this is done by varying the reference power level used to generate a power control signal transmitted by the base-station to a SAD.

The present invention also encompasses radio telephones and base-stations adapted for use in systems in accordance with the different aspects of the invention. Examples of systems in accordance with the present invention will now be described in detail with reference to the accompanying drawings, in which: BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a schematic showing the architecture of one example of a network suitable for use in the present invention; Figure 2 is a block diagram of a digital radio telephone;

Figure 3 is a schematic of a duplex (or dual simplex) connection via a base-station;

Figure 4 is a diagram showing typical frame and slot structures of the MDMA protocol;

Figure 5 is a simplified schematic of the spread- spectrum spread-despread process;

Figures 6A and 6B are diagrams showing signal spectra before and after despreading; Figure 7 is a simplified schematic of a SAD transmitter including a power control system;

Figures 8A and 8B are schematic diagrams of the base- station structure and power control section of an MDMA base-station receiver respectively; and Figure 9 is a schematic of the power control system.

Examples of the present invention are described below in relation to MDMA and CDMA systems.

DESCRIPTION OF EXAMPLES Figure 1 shows a schematic of the overall system architecture of an MDMA system. A number of subscribers are connected in common to an access node or base-station 1. Each subscriber uses an interface device called a subscriber access device 2 (or SAD) which allows access to the physical network by the appropriate data terminal equipment 3 (DTE) which is connected on the subscribers side of the SAD. In the present example, all the subscribers are connected to the network via the medium of radio transmission and the common air interface is a radio transceiver. The access node 1 is represented by the base- station and it acts as a switch to route the information flow between the subscribers and also acts as the local network controller. Figure 2 is a block diagram of the radio telephone incorporating the MDMA SAD of the present example. The telephone might be used for voice telephony, or, for example, for providing a data link for a personal computer using a modem.

Subscribers may also send or receive information to another similar network via the access node, which routes the information through a conventional transmission network such as a dedicated broadband link (optical fibre or microwave link) to the access node of the other network. Furthermore, it is possible for subscribers to access other types of broadband service available on the ISDN network via the access node which can also route signals to an asynchronous transfer mode (ATM) switch. The mode of operation of the network is either "full duplex" or "dual simplex" which is accomplished by allocating two physical channels for the two different directions of transmission. In the case of the radio network, this is achieved by allocating two frequency channels, one for the up-link and one for the down-link.

The network is equivalent to a star network but uses shared transmission media rather than many separate wires. The main purpose of this radio network is to allow the network subscribers to have access to a range of digital information services, such as digitized voice, FAX, data, video and public utility telemetry etc. which are either currently available or have been proposed for the integrated services digital networks (ISDN) . The appropriate terminal equipment must be connected to the SAD via terminal ports, of course, but it is necessary for the

SAD to be transparent to these various communication services. The SAD not only acts as the interface device, but also as a switch. A feature of the access protocol is that it uses common channel signalling to control directly the switching function of the SAD to enable the information to be routed through the SAD to the appropriate terminal port on the SAD. The configuration of the SAD is also directly controlled by the MDMA access protocol via common channel signalling in the sense that it can adapt, dynamically, to different data formats in a way which optimizes the overall throughput of the system. However, the main requirement of the SAD is to gain, control and maintain access to the network by the subscriber according to the overall network access protocol.

The access node controls the operation of the shared channel on the basis of information supplied to it from each subscriber. This is also supplied via the access protocol. The access protocol to be described has four main layers. Even so, the access protocol may be assumed to operate at the lowest levels of the overall (end-to-end) protocol used by an application since it represents the transmission of the raw data at the bit level. For example, if the network allows a subscriber to transmit packetized data, the subscriber's own DTE may provide the necessary data link protocol to initiate packet retransmission etc, and the system is transparent to the format of the data. A feature of the proposed access protocol is that it is configurable to allow a great many different data link protocols to be handled by the MDMA network as well as a wide range of data formats including voice, data and video information as well as data transmission to the X.25 standard. For certain types of data transmission which currently use voice channel modems, e.g. the following data link protocols: V21, V22, V22bis, V23 etc the access protocol is completely transparent. The various layers of the protocol are described in further detail in the present applicant's above-cited co-pending application.

A similar architecture may be adopted for CDMA systems.

Code Division Multiple Access and Multiple Division Multiple Access Methods Code-division multiple-access (CDMA) networks use direct-sequence spread-spectrum code division multiplexing. In this method, all the subscribes to the network share the same frequency allocation but each transmission is encoded with a unique code. All the codes are from a set of codes that are virtually orthogonal. Because all transmissions are co-channel, each encoded transmission acts as noise to all the other users. In a receiver, the required message is extracted from the noise of all the other transmissions by correlating the received signal with a frequency and epoch synchronised locally generated replica code. The output is determined by the auto-correlation function of the code. For the case of a code called a maximal length sequence, for example, the decoded output is a maximum when the codes are perfectly synchronised, but the output decreases linearly as the phase error increases, and when the relative phase error between the two codes > | τ | , where T is the width of a code "chip", the correlation function is a minimum. Effectively, the correlation process synchronously demodulates the wanted data from the code and the integration process associated with the correlation process acts as a low pass filter which passes the wanted data but rejects the noise of the other CDMA signals according to the degree of cross-correlation between the local code and the other codes from the code set. Since the codes are not perfectly orthogonal, the result is "other user noise".

In order for the correlation process to extract correctly all the data bits, it is usual for one sequence period to correspond to each data bit (although this is not essential) . Consequently, the "chip" rate of the sequence is much higher than the data rate. As a consequence, the bandwidth of the transmitted signal is much wider than the bandwidth of the baseband data. Such a method of multiplexing is generally referred to as direct-sequence spread-spectrum multiplexing. The spread-despread process is illustrated in Figure 5. To the wanted signal, the spread-despread process is entirely transparent. However, the other users appear as broadband noise, and, assuming them to be uncorrelated to the wanted signal, are not despread. Nevertheless, the wanted signal is corrupted by the noise of the other users which pass through the post correlation filter, as shown in Figure 6.

The ability of a spread-spectrum receiver to reject

"other-user noise" and interference is termed the process gain. To illustrate this, let the received power in the wanted signal be P_r and the total noise power over the same transmission bandwidth, B_t be P_n.

The despread-process is assumed to be lossless so the power in the data bandwidth is still P_r. However, the noise power density is:

B,

and the noise power over the data bandwidth, B_d, is:

Thus, the output signal to nose ratio, SNR₀, is related to the input signal to noise ratio, SNR_1n, by:

B„

SNR₀ = SNR_in - = SNR_in . G_p

where G is the process gain. Consider now the case where the noise is due only to the noise produced by the other spread-spectrum users. In this case, assume that the network supports X users, of which X-l act as noise to each other user.

If all transmissions are received with the same power, P_r, the effective output SNR is given by:

B- _ i

SNR₀ = _{{ χ} _ _ι} — = _{{ χ} _ _ι) G_p

Consequently, since the spread-spectrum modulation technique provides finite processing gain, it is possible for the spread-spectrum system to act as a multiple access system supporting X users only if the quality of each service is limited to having a finite SNR₀. This is a totally different way of operating a network than with frequency division multiplexing (FDM) or time division multiplexing (TDM) , where the quality of service is limited only by extraneous interference such as adjacent channel interference (for the FDM case) or inter-symbol interference (for the TDM case) . When the system also has to contend with noise and interference, as well as "other user" interference the result is similar. In this case the output SNR is simply:

where N_th is the extraneous noise power and J is the interference power within the transmission bandwidth.

The CDMA system may be considered as the physical layer of the network architecture which carries the data across the transmission medium. Multiple division multiple-access (MDMA) is a modified form of CDMA which incorporates aspects of time division multiplexing with CDMA to increase the overall spectrum utilization of the network beyond what would be possible with CDMA alone. The MDMA system differs from the CDMA system in that the underlying data is transmitted synchronously and because of this it is possible to use time division multiplexing as a form of sub-multiplexing. Consequently, CDMA multiplexing creates a number of virtual channels (say X) . The despread data on each of these virtual channels is formatted as a time frame of specific length, T_f, which is sub-divided into Y shorter time slots, each of duration T_f/Y. Each time slot is allocated to a specific SAD using conventional TDM techniques. Consequently, the total number of SADs which may use the network simultaneously is N_tot = X.Y. Each time slot comprises a short guard band of about 6 μs, leading and trailing synchronisation bits, a header code and the data bits, as shown in Figure 4. In this type of transmission, each SAD transmits in burst mode. The SAD transmission only takes place for the duration of the time slot (or time slots) which have been allocated to it. The header code is split into a number of sub-codes which perform various systems tasks, including an address to the appropriate DTE connected to the SAD. The mode of operation of the CDMA system and the MDMA system can be identical, depending on the media access protocol used. An essential difference between CDMA and MDMA is that for CDMA, the virtual channel is dedicated a single SAD, whereas for the MDMA system, communications can be established between Y SADs for the same code allocation. The MDMA protocol is described in further detail in our co-pending application cited above.

The need for power control in direct-sequence spread- spectrum systems When all spread-spectrum users are received at the base-station with equal power, P_r, let X users be supported. Under these conditions, the output SNR is then:

SNR_Ω = £—-

° {X - 1

If one of these spread-spectrum signals is now received at a power L.P_r (where for simplicity L is an integer) , by virtue of being closer to the base-station than the others, the total noise power can only be the same as the previous case if the total number of users is now reduced to X - L. As more transmissions are received with higher powers, the total number of users that can be supported is reduced. This is the "near-far" effect in which the available process gain becomes utilised rejecting the strong signals from transmissions near to the base- station and is then insufficient to process the wanted but weaker transmission from a more distant transmitter.

Consequently, in order eliminate the near-far effect, power control of each transmitter must be carried out to ensure that all transmissions are received at the base- station with equal power. The method by which this is carried out in the preferred method is described in the next section.

Power Control Module in the SAD - A First Example

The transmitter power of each SAD is not fixed but is controlled by a variable gain amplifier (1') as shown in the schematic diagram of Figure 7. The gain of this amplifier (and hence the transmitter output power) is controlled by a digital word (2'). The digital word is output from the SAD controller (3'). The SAD controller examines the header code within the CDMA or MDMA data format, shown in Figure 2. The header code performs several functions, but one of them is to pass on the destination address of terminal type (e.g. voice or video codecs, data terminal, group 4 facsimile etc.) to the SAD controller and this is used to pass the data which follows the header to the correct device. One of the permitted destinations is to the SAD transmitter power control module (4') via (3') and (2'). The data in this case contains an error signal sent by the base-station which represents the instantaneous difference between the actual received signal power of the SAD transmission at the central base-station and the required signal power. A schematic of the base-station part of the power controller is shown in Figure 8.

The spread-spectrum signal is first de-spread using the synchronised local replica code (5'). For the purpose of the power control module, the mean signal level in the de-spread signal is then obtained using a conventional piece of circuitry such as, but not necessarily restricted to, a mean signal detector and averager implemented using a digital signal processing circuit (6'). This value of mean signal level is compared with a value representing the required value of signal level within (6') . The difference in the two values (7') is the required error signal.

A feature of the preferred method, which is entirely new, is that the reference power level is variable depending upon the traffic carried by the central base- station. Thus if the traffic is low (i.e. the number of SADs accessing the network is low) the reference power can be reduced in order to take advantage of the improved receiver sensitivity when the number of CDMA channels is low. This facility to adjust the reference power level according to the number of users is permissible because the central base-station effectively controls the number of simultaneous users.

At periodic intervals, the error signal (7') is transmitted by the central base-station to the appropriate SAD by coding the header code, as appropriate, to select the power control module in the SAD in preference to other DTE equipment. For the MDMA system, this involves selecting the correct CDMA code and the correct user time slot. For CDMA systems, there is no underlying time division multiplexing and the header code is associated with one code for all time and hence one SAD, as described above. The error signal data is then transmitted during the time slot using the data encoder/multiplexer (8'). The period which is used for updating the error signal can be set by the system designer. Since each SAD decodes every header sent to it by the central base-station and sends the data to the appropriate terminal (including the power control module) the repetition rate can be as often as the frame repetition rate or much slower, determined by, say, the packet length of "real" data or the natural breaks in PCM encoded speech, so that power control does not interrupt the data flow. Alternatively, the process is sufficiently flexible that the central base-station can update the error control signal at a predetermined time during data flow. The overall control system is shown schematically in Figure 9.

In practice, the control loop drawn in Figure 9 is not time continuous but uses sampled analogue values which are transmitted to the SAD digitally. When the propagation loss between the SAD and the base station varies very slowly, as anticipated in the application of the technique to a fixed site MDMA system, the update period of the samples around the loop can be long and in this case the precise method of implementing the control loop is relatively unimportant. Let a signal sample, S_ref, represent the required power level in the base-station receiver for a given CDMA virtual channel. Let S_act be a signal sample representing the actual received power at the base-station. The error signal sample generated in the base-station is S_err = S_ref - ^s _aCf This signal sample is effectively amplified in the base-station by K^ and the signal sample K^S^,. is transmitted back to the SAD via the down-link channel. The error signal sample is detected and further amplified by a system of gain K_sad in the voltage controlled power amplifier which controls the SAD transmitter output power. Consequently:

^Serr ^{= S}ref ~ ^Sact

^Sact ~~ ^K s ^{, K}sad - ^Serr

Thus , Sa c ( 1 + κ_sad. κ_bε)

If the two gains, K_sad and K^, are high, it will be seen that in the limit, K_sad.K_bs→<», then S_act tends to be the constant reference power S_ref, as required. The precise value of S_ref is set by the required bit error rate in each base station CDMA virtual channel. This may be found by calculation, or, as is used in current microwave links, may use available circuitry for estimating the bit error rate directly from the received signals so that S_ref may be set dynamically under prevailing conditions.

If the fluctuations are very fast (as in the case of fast fading for mobile systems) special digital signal processing techniques are necessary in order to ensure the feedback controller remains stable. Such techniques have only recently been possible with the advent of digital signal processing techniques. This circuit element is vital for mobile SADs but not for fixed-site SADs.

Power Control Module in the SAD - A Second Example The constituent parts of the module are essentially the same as the first example. In this case however, the precise error signal (7) ' is not sent back to the SAD. In this case, the comparator in (6') produces a unique two bit code if i) the received and reference power levels are the same, ii) if the received power is greater than the reference power and iii) if the received power is smaller than the reference power, respectively. This two bit word is coded into the header code of the packet sent to the SAD being controlled. In this case, when the SAD dissembles the packet, the two bits of power control information are recovered and the SAD transmitter power either, i) remains constant, ii) is reduced by a preset amount or iii) is increased by a preset amount. The magnitude of the power increment can be also sent in the header code from the base-station. A three bit code, for example, would allow eight increment sizes to be uniquely coded (for example code 000 = 0.01 dB, code 001 = 0.05 dB, etc to code 111 = 1.0 dB) . It will be clear that the only major difference in circuitry in the SAD lies in the A/D conversion circuitry in the base-station and the D/A conversion circuitry, shown in Figure 9. This is replaced with a form of "delta-mod" coding, but this is entirely practicable with current digital signal processing circuits.

Power Control Protocol

The foregoing has highlighted that there is no essential difference between network data and power control data - only the header code is different. Consequently, there is no change in data format required for the power control. In this section, the way in which the power control system utilizes the media access protocol will be described. The technique is essentially the same for both CDMA and MDMA as long as they use this type of access protocol.

In this type of access protocol, the first process by which a SAD gains access to the network is when it is programmed by the SMARTCARD. This logs the SAD onto the network. This is accomplished by the SAD "hailing" the central base-station using a specific "hailing" or pilot code on the up-link which is constantly monitored by the base-station. When the SAD is programmed for the first time, the SAD hails the base-station at a very low power to prevent the transient overload of the total transmitter power within the network. The power level of the SAD is then automatically increased, progressively, until an acknowledgement is received from the base-station confirming that the request to log onto the system has been received. At this point, the SAD transmitter power is then fixed and this represents the coarse power level appropriate to the range of the SAD from the base-station. Because the log-on process does not imply an intention to transmit data, there is no requirement at this stage to obtain precise power control, only that there is sufficient power to allow the SAD to communicate with the network without swamping the system. This value of power level is then stored in the power control module of the SAD and an allowance is made when storing this level to include an extra margin of power, called the "fade margin", which allows for the possible increase in path loss due to changes in atmospheric conditions. The second part of the power control protocol takes place when a SAD requests access to the network for the purpose of transmitting information. In this case the SAD hails the base-station. (In this case the header used by the SAD reflects the fact that the SAD is not logging-on but requesting access.) The base-station establishes the link between SADs and then refines the transmission powers of both SAD transmitters. The establishment of the link results in the base-station allocating a specific code and, for the case of the MDMA system only, a specific time slot (or multiple time slots) to both SADs. SAD A communicates to SAD B on the up-link from A to the base-station. At the base-station, A's message is retransmitted to SAD B on B's down-link. Messages from B are transmitted to A on B's up¬ link and received on A's down-link. The power control information is sent to both SAD A and SAD B on their respective down-links, interspersed with data as described above. However, accurate power control is first set up before any data is allowed to be transmitted, and this forms part of the access protocol and the power control protocol. When accurate power control has been established, the link is established and only then does the central base-station allow data transmission between the SADs. Accurate power control is deemed to have been established when successive error signal samples (7') fall to an acceptably low value.

To take into account variations in the propagation loss due to atmospheric fluctuations, or other factors, the instantaneous power control error signal is periodically transmitted to maintain optimum receiver power.

Although revisions to the SAD transmitter power are constantly updated, the revisions are not held in memory when the SAD terminates its call. Only the power setting established during the initial log-on process is stored and it is this value which is used every time the SAD attempts to gain access to the network. (This is because this value allows for the fade margin, whereas the accurate power control settings do not.)

Power Control of Base-Station Transmitter Powers

In the following, the requirements for dynamic power control for the base-station power for CDMA and MDMA systems are essentially the same. For the sake of clarity, the case for the CDMA system will be considered first, followed by the MDMA system, where different.

Power Control of the CDMA Base-Station It is usual for a CDMA base-station to transmit all CDMA channels at the same power level. In this case, the power level of each CDMA channel reduces with respect to distance from the base-station at the same rate. Consequently, the condition that each SAD receives all the CDMA channels at the same power is guaranteed at all SAD sites in the network.

This method of operation works well if the density of users accessing the network simultaneously is uniform across the cell. Where the density of simultaneous users is not uniform and the number of SADs is fewer than the base-station capacity, there are performance benefits to be gained by carefully controlling the base-station transmission powers to the various SADs. Furthermore, the technique allows dynamic resizing of the cells and this enables the radio-link performance and the cost effectiveness of the network to remain optimised as the number of subscribers in a given geographic region changes. These are two innovative features of this part of the present invention. These are described in broader detail below.

Power Control for Minimum Base-Station Power Transmission

Consider the case of a fixed cell size where the density of simultaneous SAD users is not distributed uniformly across the cell and there is a higher density of users close to the base-station. (It is assumed that there are fewer simultaneous users than the base-station capacity) . In this case, it is much better to transmit to the many near SADs at a relatively low power. Transmissions to the few distant SADs takes place at a higher power to take account of the increased path loss. "Other-user" interference to the distant SADs from transmissions to the near SADs is now much lower than for the case of equal transmission power for each CDMA channel and the sensitivity of distant SADs is increased as a result. This allows the transmission powers to these distant SADs to be lower than it would have been when all the powers were equal. This has the further advantage that it reduces the interference power to the near SADs from transmissions to the distant SADs.

Clearly, in this case, power control of the base station reduces the overall base-station power and this reduces the interference to other, non-CDMA, bandsharing users.

Base-station power control fails, however, when there are more SADs transmitting from the perimeter of the cell than near to the base-station. In this case, most CDMA channels are transmitted at full power and reducing the transmission power to SADs which are close to the base- station simply worsens the input signal to noise ratio to those SADs and their performance suffers as a result. The mechanism for controlling the base-station power is very similar to controlling the SAD power, and is initiated by the SAD. Each CDMA channel of the base-station has a separate power control circuit which is identical in basic form to that used in the SAD transmitter. A data signal from the SAD, which is addressed to the power control module of that CDMA channel, is used to control the gain of that output amplifier of that CDMA channel. The data signal sent from the SAD represents the error between the actual power received by the SAD and the desired power of the SAD. The general system structure of Figure 8b is still applicable. It is assumed, at the outset, that the base-station hailing channel is either at a sufficiently high power that SADs at all ranges can ultimately synchronise their codes to the base-station. The power of the hailing channel is not variable. When SADs log-on to the network via the hailing channel, they do so by progressively increasing their transmitter power until it is correctly received by the base-station. This value of SAD transmitter power (plus a fade margin) is then stored in the SAD. This value of SAD transmitter power is also sent back to the base-station as data to the base-station power control processor of that CDMA channel, and the power of the base-station transmission to that SAD on the assigned CDMA channel is adjusted to match the SAD transmitter power using the gain controlled amplifier on that channel. Once the link between a SAD and a base-station channel has been established, the SAD monitors the base-station power it receives and it periodically sends back error information to modify the base-station power level for that CDMA channel. This value of base-station power is stored in the power control processor. At the same time, the base- station monitors the SAD power and periodically sends back error signals causing the SAD to adjust its power, as described earlier. The error signal transmitted back to the base-station may represent an absolute error, or may simply request the power to be incremented or decremented by a fixed amount, as described earlier.

Whether it is appropriate to control the power of individual CDMA channels depends on the activity and distribution of users within the cell. Because the base- station has access to the database of SADs and their location, cell activity and user distribution can be determined by the base-station and this information can be used to decide whether base-station power control is appropriate or not using a microprocessor controller.

However, even when it is not appropriate to control the power of individual CDMA channels (for example; because there is a high density of users on the cell perimeter) , it is still possible to use base-station power control to reduce the level of radio emissions from the cell.

In this case, the power in the base-station transmissions of all the CDMA channels is equalised, but the precise power level is adjusted to provide the minimum required signal level at the current furthest SAD within the cell. If the furthest SAD logs-off, the power level of all CDMA channels is reduced to the level which will allow communications with the new furthest SAD, etc. Conversely, if an additional SAD logs-on to the network and has the furthest range from the base-station, the power in the CDMA channel between the new user and the base-station is set and the power in the all other CDMA channels increases to the same value.

This is easy to achieve because the base-station power control processor keeps a record of all the various SAD transmitter powers which are currently logged-on to the network and the corresponding base-station powers used on the down-link channel.

Power Control for Dynamic Cell Resizing In this case, the maximum base-station power is adjusted according to the cell diameter. For large cell diameters the peak power is large, whilst for smaller cells the peak base-station transmitter power is reduced. Cell resizing may be carried out manually from the base-station or, within certain cell-size constraints, automatically under the control of the SADs. The maximum cell size is determined by the power of the hailing channel from the base-station because all SADs within the cell must be capable of receiving the signal on the hailing channel. If it is required to modify the cell- size from its nominal size then this can be achieved by:- 1) changing the power of the hailing channel. In this case, the network is only available to those SADs which are able to receive the hailing channel.

2) modifying the list of SADs stored in the database which is accessed by the base-station. In this case the power of the hailing channel signal is not reduced, although SAD and base-station transmission powers are reduced. Consequently, if the cell is to be reduced in size, a particular base-station may refuse to acknowledge the log-on request of a SAD beyond a certain range. This is possible because the SAD stores the value of transmitter power it needs to establish contact with the base-station and sends this to the base-station as part of the log-on procedure. This value of SAD transmitter power gives an estimate of the range of the SAD from the base-station. Moreover, the base-station may also have access to the physical location of the SAD when it gives its physical identification number during the log-on procedure because this will be stored in the base-station's database.

Once it has been decided to resize a cell on the basis of which SADs can or cannot gain access to the base- station, because the logged-on SADs determine the power of each CDMA channel, as described earlier, it is a relatively simple process to control the base-station power to levels which are appropriate to the maximum SAD range within the new cell.

Power Control of the MDMA Base-Station The MDMA system has an extra degree of freedom due to the MDMA data format, in which one spread-spectrum code (defining the virtual channel) is used to gain access to several SADs using time division multiplexing beneath the spread-spectrum multiplexing. Consequently, it is not possible to uniquely target a single SAD in the same way as a conventional CDMA system, where there is no underlying time division multiplexing. The method of dynamic power control of the base-station transmitter power in this case can be more complex than described above. (Although the CDMA system has only X CDMA channels, the MDMA has X virtual channels and Y time slots, giving X.Y channels in total and it is impractical to individually control the base-station power of all X.Y channels) . For the case where SAD users are uniformly distributed within the cell, or where there is a high density of users at the perimeter of the cell, power control of individual channels would not he appropriated, as described above, and in this case the power of all transmissions would be identical and at a level which allows reliable communications with the furthest SAD in the cell. In this case the method of power control is identical to that described for the CDMA case: each SAD reports to the base station the level of power it needs to gain access to the network and the largest of these values sets the base- station transmitter power.

Dynamic power control only becomes a problem when there is high density of users near to the base-station. In this case, in order to minimise the total power transmitted from the base-station, it is necessary to allocate the same CDMA virtual channel to groups of SAD users which have a similar range from the base-station, identifying individual users by means of their time slot allocation. In this way, all time slots in a particular CDMA virtual channel are transmitted at the same power level. The precise value of this power is set by the furthest of the SADs from within that group of SADs. Again, this is easy to accomplish because each SAD sends the value of its transmitter power level back to the base- station.

In this way, it is not necessary for the all the SADs in a group to have precisely the same range, but the system will still work. In fact, it would be possible for the SADs to be scattered over the entire cell. However, in this extreme case there would be little benefit in using power control at all. As described above, the proposed system makes novel use of the media access protocol applicable, for example, to either CDMA or MDMA multiple access cellular radio systems to effect: i) remote dynamic power control of the individual SAD transmitters on the basis of equalising the received signals at the base-station ii) variable SAD and base-station power levels on the basis of the number and distribution of users sharing the network iϋ) dynamic control of the base-station power to facilitate dynamic cell resizing.

Claims

1. A communication system comprising a base-station (1) and a plurality of subscriber stations (2,3) arranged to communicate with the base-station via duplex or dual simplex spread-spectrum channels, each channel comprising an uplink from a subscriber to the base-station and a down-link from the base-station to the subscriber, characterised in that the base-station (1) is arranged to transmit power control signals to respective subscriber stations on the down-links, and the subscriber stations are responsive to power control signals from respective down¬ links to modify the subscriber station transmission signal power according to a feedback control strategy.

2. A system according to claim 1, in which the base- station is arranged to monitor the power of transmissions received from a subscriber station, to compare the received level with the desired level, and to transmit to the subscriber a control signal to modify the power transmission accordingly, thereby providing closed-loop control of the subscriber transmitter power.

3. A system according to claim 2, in which the reference power level corresponds to a minimum acceptable level for reception, and the subscriber station is arranged to respond to the power control signal output by the base- station by adjusting its transmission power so that the received power at the base is substantially equal to the said minimum power level.

4. A system according to any one of the proceeding claims, in which the base-station is arranged to transmit a power control signal in the header of a frame carrying data communications traffic.

5. A system according to any one of claims 1 to 3, in which the base-station is arranged to transmit the power control signal in a dedicated power control frame transmitted intermittently between data carrying frames.

6. A system according to claim 5, in which the power control signal includes data corresponding to an absolute power level.

7. A system according to claim 4, in which the power control signal comprises a command to increment or decrement the subscriber transmitter power

8. A system according to any preceding claim, in which the base-station is arranged to vary a reference level used in determining the power control signal in dependence upon the traffic carried by the base-station.

9. A system according to claim 8, in which the base- station also varies its own transmission power levels in accordance with the amount of traffic carried.

10. A method of operating a communication system comprising a base-station and a plurality of subscriber stations, including the steps of communicating between the subscriber stations and base-station via duplex or dual simplex spread-spectrum channels, each channel comprising an up-link from a subscriber to the base-station and a down-link from the base-station to the subscriber, characterised by the step of transmitting power control signals from the base-station to respective subscriber stations on the down-links, and modifying at each respective subscriber in response to the power control signal on its respective down-link the subscriber station transmission signal power.

11. A method according to claim 10, further comprising transmitting power control signals from a subscriber station to a base-station and modifying the power of transmissions from the base-station to the respective subscriber station accordingly.

12. A communication system comprising a base-station (1) and a plurality of subscriber stations (2,3) arranged to communicate with the base-station via duplex or dual simplex spread-spectrum channels each comprising an up-link from a subscriber to the base and a down-link from the base to the subscriber, characterised in that the subscriber station is arranged to transmit power control signals to the base-station and the base-station is responsive to the received power control signal to modify the power level of transmissions to the respective subscriber station according to a closed-loop feedback strategy.

13. A system according to claim 12, in which the subscriber stations are arranged to generate the power control signal by comparing a received power level with a reference level and outputting an error signal back to the base-station accordingly.

14. A cellular spread-spectrum radio communication system, in which the cells making up the system are dynamically resizable.

15. A cellular spread-spectrum radio communication system, in which a base-station is arranged to vary the power of transmissions in the system depending upon the level of traffic carried by the system.

16. A system according to claim 15, in which the base- station varies the power of transmissions by varying a reference power level used to generate a power control signal transmitted by the base-station to a subscriber station.

17. A radio subscriber station adapted for use in a system according to any one of the preceding claims.

18. A radio base-station adapted for use in a system according to any one of the preceding claims.