A digital radio link system and a method of adjusting the transmission power in a digital radio link system.

0001The invention relates to a method of adjusting the transmission power in a digital radio link system by monitoring the received signal level and the error rate estimate at the receiving end.

0002On determining the transmission power of a digital radio link system, not only the so-called constant power dissipations caused by the link span but also random signal fades caused by multi-path propagation, for instance, have to be taken into account. Therefore the transmission power of the transmitter is often adjusted to a value considerably above the minimum transmission power which in normal conditions would suffice for interference-free re­ception at the other end of the link span. Excessive transmission power provides the link span with a power margin which ensures interference-free re­ception even in difficult conditions. On the other hand, however, high transmission power increases interference in adjacent channels or systems. This hampers the realization of dense radio link networks, efficient utilization of frequency bands and the re­use of the same frequency within the same area.

0003U.S. Patent 4,004,224 discloses a method in which transmission power within a certain link span is increased automatically if fading is observed within this specific span, so that a constant re­ception power is maintained at the receiver. The re­ceiver produces a signal representing the level of the received signal, and this signal is sent through a return channel to the transmitter for the adjust­ment of the transmission power.

0004In U.S. Patent 4,309,771, the bit error rate of the received digital signal is monitored in place of the received signal level, and the transmission power is adjusted through the return channel in such a way that the bit error rate of the received signal does not exceed a predetermined threshold value.

0005In FR Patent Specification 8518919 the range of adjustment is, in principle, divided into two por­tions. At low reception levels the transmitter is instructed through the return channel to increase power to compensate for fading. At high levels the error rate too is monitored. If the error rate is high irrespective of the sufficient level, the trans­mission power is increased temporarily so that it exceeds the normal backoff of 6 dB (such as a backoff of 3 dB).

0006The above-described methods aim at adjusting the transmission power so as to decrease the trans­mission power required in normal conditions in order to decrease the interference caused in other link spans or channels while enabling sufficient quality of data transmission even during the fading periods. These methods, however, are not sufficiently effective in all situations.

0007The object of the present invention is to provide an improved method of adjusting the trans­mission power of a digital radio link.

0008This is achieved by means of a method of the type disclosed in the beginning, which method accord­ing to the invention is characterized by monitoring the rate of change of the received signal level and the error rate estimate; and increasing the trans­mission power temporarily to a value close to the maximum transmission power if the rate of change of the received signal level or the error rate estimate exceeds a predetermined threshold value.

0009By monitoring the rate of change of the re­ceived signal level, rapid fades occurring in multi-­path propagation can be detected efficiently. With multi-path propagation, fading may be very rapid (as rapid as 100 dB/s and more), so that changes in the level of reception can be difficult to compensate for by real-time adjustment of the transmission power through the return channel due to, for instance, the delay of the return channel. In the method of the invention the transmission power is therefore set to its highest normal value for some time when rapid fading is detected. This makes it easier to control rapidly changing situations and stabilizes the adjustment. In addition, it is possible during multi-­path fading that the received signal level is suf­ficient but errors occur in the reception due to interaction between channels. In the preferred embodiment of the method of the invention the trans­mission power is therefore increased from a value close to the maximum transmission power to the maximum transmission power if the error rate estimate exceeds the predetermined threshold value in addition to the detected multi-path condition. This degrades the signal to some extent, whereby the long-term error rate can be of the order of 1E-7 ...1E-8. In a multi-path condition, it is, however, necessary to bring the error rates 1E-3...1E-6 under control, whereby the use of the maximum transmission power for a limited period of time is advisable.

0010If the monitored error rate estimate exceeds the predetermined threshold value in normal con­ditions (no multi-path condition), the transmission power is increased temporarily close to the maximum transmission power. In this way rapid response to errors is achieved. This kind of situation may occur due to temporary propagation conditions, whereby some remote system may disturb for some time.

0011In the preferred embodiment of the invention, slow variation in the level of reception, which may be due to variation in the k value, for instance, is compensated for by monitoring the level of reception. If the level of reception falls below a predetermined minimum, the transmission power is adjusted through the return channel by increasing it by a small in­crement. In this way the quality of the received signal remains sufficiently good while avoiding use of excessive transmission power and adjacent system interference.

0012By means of the method of the present inven­tion, the transmission power can be adjusted in such a way that it is maintained at an optimal value in normal conditions while providing rapid response to interferences, thus ensuring interference-free reception.

0013The invention is also concerned with a digital radio link system according to claim 8 for applying the method of the invention.

0014The invention will now be described by means of exemplifying embodiments with reference to the attached drawings, wherein <ul id="ul0001" list-style="none"><li>Figure 1 shows a block diagram of a digital radio link system according to the invention;</li><li>Figure 2 shows a flow chart of the transmission power adjustment unit 4 of Figure 1;</li><li>Figure 3 shows the circuit diagram of the transmission power adjustment unit 4 of Figure 1; and</li><li>Figure 4 illustrates graphically the adjustment of transmission power according to the invention in an imaginary reception situation.</li></ul>

0015Figure 1 shows a radio link system according to the invention, comprising a transmitter unit 1 and a receiver unit 2. In the transmitter unit 1, a trans­mitter or modulator part 6 applies a modulated high-­frequency signal to a high-frequency amplifier stage 7, such as an output stage, which amplifies the signal and applies it to a transmitter antenna 3. At the other end of a link span 15, a receiver antenna 12 receives the signal sent by the transmitter unit 1 and applies it to a receiver 8, in which the signal is detected.

0016A block 9 in the receiver unit 2 monitors the rate of change of the received signal level, par­ticularly rapid level drops, to detect a multi-path condition, and produces a 1-bit signal S_mp the state of which is 1, if a multi-path condition occurs, and 0 if no multi-path condition is present. In the case of Figure l the block 9 compares the present value of the AGC voltage P_rec of the receiver 8 with a short­term average of the voltage or merely with a pre­ceding value. If the present value of the AGC voltage P_rec deviates at least to a predetermined extent from the short-term average or the preceding value, the change is interpreted to have been caused by a rapid fade, and the state of the output signal S_mp of the block 1 is given the value 1.

0017A block 10 in the receiver unit monitors the received signal level, and produces a 1-bit signal S_min, if the received signal level is below a pre­determined threshold value L_min. In the preferred embodiment, the block 10 also monitors the value of the AGC voltage P_rec of the receiver 8. The threshold value L_min is set so that the residual error rate of the system will be sufficiently good. The limit value L_min is preferably of the order of about 10 to 12 dB above the threshold level of the receiver 8.

0018A block 11 in the receiver unit produces a 1-­bit signal S_err the state of which is 1 if an error condition occurs at the reception and 0 if no error condition occurs. The error condition may be, for instance, N errors per time unit (e.g. per one minute). In the preferred embodiment of the invention the signal S_err obtains the state 1 immediately after N errors have occurred even though the time would not have run out. The errors can be counted, e.g., from the parity of the frame. The error signal can also be produced from the pseudo-error signal occurring in connection with the receiver 8. When the estimated error rate is greater than a predetermined threshold value, the state of the signal S_err is given the value 1.

0019The binary signals generated by the blocks 9, 10 and 11 are multiplexed by a multiplexer 13 and sent through a low-capacity return control channel 14 to the transmitter unit 1. In the transmitter unit 1 a demultiplexer 5 demultiplexes the received signal and applies the signals S_mp, S_min and S_err to the transmission power adjustment block 6. The trans­mission power adjustment block generates, on the basis of these signals, an adjusting voltage c(t) which controls the amplifier stage 7 of the trans­mitter unit 1. By means of the adjusting voltage c(t) a high-frequency power P_out applied by the amplifier stage 7 to the antenna 3 can be adjusted in response to the quality of the signal received by the receiver unit 2.

0020The functions of the transmission power adjust­ment unit 4 are illustrated by the operational diagram of Figure 2.

0021To begin with, the presence of a multi-path condition is monitored by detecting the state of the signal S_mp representing an excessive rate of change of the signal level. If a multi-path condition is de­tected (S_mp = 1), the state of the signal S_err re­presenting an error condition is monitored at point 22 to find out whether an error condition is present. If S_err = 1, a multi-path condition and error condition are simultaneously present, as a result of which a maximum transmission power P_max is set at point 23, and the transmission is continued at this value for a period of time T1 from the last detection of simultaneous multi-path and error condition. With a 16QAM signal, the maximum transmission power may be, e.g., 2 to 4 dB and with a 4QAM signal about 0 to 2 dB below the level of a l dB compression point. The time period T1 may be of the order of 5 to 10 minutes, for instance. This use of maximum trans­mission power aims at minimizing the time during which the error rate at the reception is inferior to 10⁻³...10⁻⁶.

0022If point 22 gives as a result S_err 0 (no error condition), the transmission power is set to a value P₁ at point 25, which is slightly below the maximum transmission power, and with a 16QAM signal, for instance, about 6 to 8 dB below the compression level of 1 dB. With the transmission power P₁, distortions are small and it is thus possible to achieve a good residual error rate (BBER). The trans­mission is continued at the power P₁ for a time period T2 from the last detection of multi-path condition. The time period T2 is of the order of 15 to 20 min, for instance.

0023If point 21 gives as a result S_mp = 0, that is, there is no multi-path condition present, the possible presence of an error condition is checked at point 24 by means of the signal S_err. If S_err = 1 at point 24, one goes to point 25, at which the trans­mission power is set to the above-mentioned value P₁ for the time period T2. Errors occurring without multi-path condition may be due to interference caused by exceptional propagation conditions. In such a case the situation may continue for a long period of time and it is not advisable to send the maximum transmission power P_max.

0024If point 24 gives as a result S_err 0, the received signal level is monitored at point 26 by means of the signal S_min. If S_min = 1, that is, the received signal level is below a minimum reception level L, the preceding value P_n-1 of the transmission power is increased by an increment k·d , where k is a proportional coefficient (≧ 1, typically between 2 and 4) and d is a basic increase. After the increase, the maximum value of the transmission power should not exceed the above-mentioned power P₁. By selecting k > 1, the increase in the transmission power is more rapid than its decrease.

0025If point 26 gives as a result S_min 0, that is, the level of reception is above the minimum reception level L, the transmission power is de­creased from the preceding value P_n-1 by the in­crement d. These changes are typically below 1 dB, because the updating is carried out, for instance, once a minute or more frequently. The rate of change of the transmission power should be less than 1 dB/s in order to prevent the interpretation of the change as a multi-path condition at the receiving end. The transmission power will not fall below a pre­determined value P₀ which is the transmission power in normal conditions. The link span is so dimensioned that the transmission power P₀ in normal conditions results in a reception level which is above the minimum reception level L. Accordingly, the trans­mission power is nearly constantly P₀, and the system operates about 12 to 14 dB above the threshold level of the receiver. Higher power levels are applied only exceptionally, that is, in case of multi-path and error condition.

0026Figure 3 shows diagrammatically the principles of one possible realization of the unit 4 for adjust­ing the transmission power.

0027The signals S_mp and S_err are applied to an AND gate 41 the output of which is connected to a flip­flop 42. If both signals S_mp and S_err have the state l, the gate 41 and the flip-flop 42 reset and start a counter 43 which counts the time period T1. The out­put of the flip-flow 42 causes an output LOAD1 in another flip-flop 44 to have the state 1 during the counting. When the counting of the counter 43 is com­pleted after the time period T1, the counter 43 sets the output LOAD1 of the flip-flop 44 to the state 0. The signal LOAD1 controls a memory circuit 45. During each cycle of a clock signal CL, a count correspond­ing to the maximum power P_max is loaded from the memory circuit 45 to a counter 55 as long as the signal LOADl has the state 1.

0028The signals S_mp and S_err are also applied to an exclusive-OR gate 46 which controls, through a flip-­flop 47, a counter 48 counting the time period T2. The counter 48 is reset and started whenever any of the signals S_mp and S_err has the state 1, that is, when either a multi-path or error condition is present. An output signal from a flip-flop 49 is set to the state 1 during the counting of the counter 48 and it is applied to one input in an AND gate 50. The signal LOAD1 is applied to another input, that is, an inverting input, in the AND gate, whereby the output signal LOAD2 of the gate 50 has the state 1 only if the signal LOAD1 has the state 0 and the output signal of the flip-flop 49 has the state 1. A count corresponding to the transmission power P₁ is applied from a memory circuit 51 to the counter 55 during each cycle of the clock signal CL as long as the signal LOAD2 has the state 1. The gate 50 ensures that the maximum power P_max is loaded in the first place in case of competition between P_max and P₁. The signal S_min is applied directly to AND gates 53, 54 and the signals LOAD1 and LOAD2 through a NOR gate 52. In this way, the adjustment of the transmission power takes place by means of the signal S_min only if neither multi-path condition nor error condition is present, whereby the signals LOAD1 and LOAD2 both have the state 0 and the output of the gate 52 has the state 1. The signal S_min is thereby able to affect the counter 55 through the gates 53 and 54. If the signal S_min = i.e., the received level is below the minimum reception level L, an in­crease k · d is applied to the input of the counter 55. If the signal S_min 0, i.e., the minimum re­ception level is exceeded, the increment d is extracted from the contents of the counter 55 through an input D. The counter 55 controls internally that the transmission power adjusted therethrough does not exceed the power P₁ or is not below the power P₀.

0029The above-described operation can, of course, also be realized as a microprocessor programme.

0030The digital output of the counter 55 is applied to a D/A converter 56 which produces an analog vol­tage. If required, this voltage can be subjected to a nonlinear conversion in the block 57 if it is desired that the power increment steps on the dB scale should be substantially of the same size. As a result, an adjusting voltage c(t) is achieved, by means of which one amplifier in the amplifier sequence of the trans­mitter is adjusted, such as the high-frequency terminal amplifier.

0031The counting block 6 described above can be realized in a simpler form. One alternative is to apply the same time period T1 for sending both the maximum power P_max and the power P₁. Thereby one cir­cuit suffices for the counting of time. In the micro­processor implementation, it might be possible to save a few program steps in this way.

0032Another alternative, probably applicable with constant-amplitude modulation methods, is to use a single higher power value P₁ in place of the powers P1 and P_max and correspondingly a single time period T2. This simplifies the realization considerably.

0033The adjusting principle could also be modified by monitoring possible exceeding of a predetermined maximum reception level U. If the received level is between L and U, there is no need to change the power level. If the level U is exceeded, the transmission power is decreased. An advantage of this alternative would be the less frequent changes in transmission power and more reliable detection of multi-path con­dition. A drawback is the need of an additional bit to transfer information to the transmitter.

0034Figure 4 illustrates the operation of the adjusting method of the invention in an imaginary reception situation. At point A the received signal level is below the predetermined minimum reception level L, whereby the transmitter is adjusted to some extent (an increment k · d is added to the counter 55) to increase the power. At point B it is to be seen that the reception level exceeds the minimum reception level L and the increment d is extracted from the counter 55. The same takes place during the following clock cycle at point C.

0035At point D, a rapid change in the received signal level as compared with the short-term average is detected, such as a difference of 3 to 5 dB or a rate of change which is greater than 3 to 5 dB/s. The power P₁ close to the maximum power is thereby adjusted in the transmitter. At point E both multi-­path and error condition are detected, and the maximum transmission power P_max is connected to the transmitter for the time period T1. No further errors are detected, but the multi-path condition continues. At point F, the transmission power is decreased from the maximum power P_max to the value P₁ after the time period T1 has elapsed from the last detection E of errors. At point G the time period T2 has elapsed from the last detection of multi-path condition, whereby the transmission at the power P₁ is interrupted and the adjustment is begun to be carried out with the signal S_min. It is thereby observed at point G that the minimum reception level L is exceeded and the output power P_out is decreased by the increment d. The same takes place at point H within the following clock cycle and the transmission power falls gradually until the normal transmission power P₀ is achieved at point I.

0036At point K, an error condition occurs without the multi-path fading. The transmission power is thereby increased to the value P₁. The power is transmitted for the time period T2 after the last detection of an error condition. At point L the adjustment again takes place with the signal S_min which decreases the transmission power gradually or monotonously toward the value P₀.

0037The attached figures and the description related to them are only intended to illustrate the present invention. In their details the method and the system of the invention may vary within the scope of the attached claims.

0038The features disclosed in the foregoing des­cription, in the following claims and/or in the accompanying drawings may, both separately and in any combination thereof, be material for realising the invention in diverse forms thereof.