Control of power ratios for in-phase and quadrature channels in a communications system

【0001】

(Background of the Invention) The present invention is open to wireless communication systems using I / Q modulation, and particularly relates to controlling the power ratio of I channel to Q channel in such a system.

【0002】

Modulation methods that utilize in-phase (I: In-phase) signal components and orthogonal (Q: Quadrature) signal components are known. In some examples, such as uplink transmission in the IMT2000 Wideband Code Division Multiple Access (WCDMA) wireless communication system standard proposed in Europe and Japan, the I and Q components have different data channels (WCDMA). Hereinafter, in the present specification, "I channel" and "Q channel") are transmitted.<u style="single">like</u>I / Q modulation is used. In the proposed WCDMA system, the Common Control Channel (PCCH) is a Q channel with a spread coefficient of 256 and a data rate of 16 kbit / s.<u style="single">To</u>Is transmitted more. Meanwhile, traffic channels and<u style="single">Individual</u>A control channel (PDCH: Dedicated Control Channel) is an I channel with a data rate between 32 kbit / s (diffusion coefficient 128) and 1024 kbit / s (diffusion coefficient 4).<u style="single">To</u>Is transmitted more.

【0003】

The power requirements of each of these I and Q channels are, of course, different from each other. Therefore, the power levels of the I and Q channels are different before diffusion and scrambling are applied. First, it can be assumed that the power of a channel is proportional to the data rate of that channel. However, this is not a requirement, as quality of service (QoS) requirements vary from channel to channel. The PCCH channel has a pilot that requires QoS different from the QoS required for voice or data services multiplexed on the PDCH channel.

【0004】

I-channel and Q-channel power levels are controlled by a common power control algorithm. This algorithm increases or decreases the power to keep the signal power constant at the receiver. To achieve this, the algorithm will track Rayleigh fading, lognormal distribution fading, and path change losses due to changes in terminal-base station distance.

【0005】

Terminals in WCDMA systems face certain problems due to the demand for transmitters that feature good modulation accuracy. In order to obtain an accurate power difference of, for example, 3 dB between the I channel and the Q channel, the amplitude ratio in the terminal needs to have the following values.

【0006】

β = 1 / 2 = 0.707 [0007]

Q channel data sample to obtain this power ratio<u style="single">To</u>β<u style="single">Is multiplied</u>The resulting sample is fed to the diffusion-complex modulation circuit along with the sample from the I channel.

【0008】

Representing a value of 0.707 requires many bits for the multiplication performed for each sample transmitted, making it difficult to achieve such an accurate power ratio. As is well known, the computational load caused by multiplication is related to the operand lengths involved. Increased computing load<u style="single">By</u>In addition to increasing the calculation time, it can be said that the power requirement for executing the calculation increases.

【0009】

Furthermore, when designing a system such as WCDMA described above, assuming that the value of β can be changed continuously, in the system<u style="single">If the components are different</u>(For example<u style="single">Between terminals</u>Manufacturer<u style="single">If different</u>), When expressing β<u style="single">Occurs</u>Quantization error<u style="single">Is different</u>.. As a result of these inconsistencies, the performance of the system will be degraded.

【0010】

(Overview of the Invention) That is, an object of the present invention is to provide a power ratio control method for improving performance as compared with the conventional technique.

【0011】

According to one aspect of the invention, the aforementioned and other objectives are achieved by methods and devices used in transmitters in wireless communication systems such as WCDMA communication systems. Transmitter, a set of data I channel transmit Le, send another data set in the Q channel. The transmitter generates a gain signal β and digital data related to the Q channel.<u style="single">To</u>Gain signal β<u style="single">Multiply</u>。<u style="single">here,</u>Gain signal β,<u style="single">Just with a given number of bits</u>Representable finite value<u style="single">(For example, the bit length after the decimal point is 4</u><u style="single">bit</u><u style="single">)</u>To<u style="single">Limits</u>This reduces the complexity of the multiplication operation.

【0012】

According to another aspect of the present invention, the modulation in accuracy due to the quantization of β can be eliminated by using the same quantization size for β in all the components in the wireless communication system.

【0013】

(Detailed Description) Hereinafter, various features of the present invention will be described with reference to the drawings. In the drawings, the same parts shall be given the same reference number.

【0014】

FIG. 1 is a block diagram of a transmitter of a wireless communication system that operates according to the present invention. As described in the background section of the invention, the transmitter uses I-channel and Q-channel. The data sample related to the Q channel 103 is supplied to one end of the input of the multiplier 105, and the other end of the input of the multiplier 105 receives the value β from the power ratio control circuit 107. As described above, the multiplier 105 multiplies the Q channel data by β to control the power ratio. The multiplied Q channel data is supplied to the diffusion-complex modulation circuit 109 together with the data relating to the I channel 101. The resulting I-channel and Q-channel signals are fed to the first and second D / A converters 111 and 113, respectively. The analog signals supplied by the D / A converters 111 and 113 are supplied to the first and second mixers 115 and 117, respectively. The mixer 115 mixes using a cosine signal, while the mixer 117 mixes using a sine signal. The outputs from the first and second mixers 115, 117 are connected to coupling means 119 (eg, adder), and the output is supplied to the power amplifier 121 for pre-transmission amplification.

【0015】

According to one aspect of the invention, the transmitter further includes a power ratio controller 107 that generates the value β by the method described below. The principle of operation of the power ratio controller 107 is derived from the fact that, at least in part, in a wireless communication system such as CDMA, all signals are transmitted simultaneously at the same carrier frequency. All signals other than the desired signal are considered interference at the receiver. When received by the base station, all received signals need to have approximately equal energy per transmission bit in order to minimize interference from each user and thereby optimize cell capacity.

【0016】

In the uplink channel of the I / Q modulation method such as the WCDMA system described above, the spread coefficient of the Q channel is set to 256. Here, the diffusion coefficient of the I channel can be any value of 128, 64, 32, 16, 8, or 4. The purpose of the gain coefficient β is to ensure the optimization of system performance. In the conventional system, there is no limit to the possible values of β (that is, in the conventional system, β is continuous.<u style="single">Typical</u>Written as a parameter<u style="single">Was</u>。)。

【0017】

The choice of β value can have a significant impact on system performance. For example, the speed of information transmitted in a CDMA system<u style="single">Scale</u>There is a "chip rate" as. In a typical system, the base chip rate is fc = 4.096Mchip / s, and other chip rates such as 1.024, 2.048, 8.192, or 16.384Mchip / s can be defined depending on the application. See Figure 1. The digital information signal supplied to the Q channel 103 has an oversampling rate of 4 (OS), and each sample can be represented by the number of bits Nb. Then, the multiplication by β of each sample obtained as a result requires fs = fc · OS = 16.385Mops in 4.096Mchip / s mode.

【0018】

All additional bits in the multiplier (ie, the number of bits in Nb * β) make each multiplication operation more complicated and also increase the additional bits per sample that need to be transmitted and processed. Therefore, the number of bits involved in multiplication must be minimized in order to achieve a power efficient design. This can be achieved by limiting β to a value that can be accurately represented with a relatively small number of bits. However, as mentioned above, the purpose of the gain coefficient β is to ensure the optimization of system performance, and this cannot be achieved with any β value. Rather, it is necessary to perform an analysis to determine an appropriate β value that can reduce the complexity of the operations required for multiplication operations without excessively degrading system performance.

【0019】

An example of the analysis is as follows. In the realization of a typical transmitter, β is assumed to be in the form of a signal with a finite number of bits. Therefore, β_IDEALRepresents an "ideal" value that guarantees optimized system performance, and the β signal is a small amount of quantum generated by the approximation of the ideal value to the ideal value with a finite number of bits. It shows the sum with the noise conversion noise. That is, β = β_IDEAL+ (Quantization noise).

【0020】

In the examples<u style="single">Is possible</u>As long as (for example, β_IDEAL=0.5<u style="single">so,</u>β<u style="single">Is the bit length after the decimal point</u>At least 1 bit<u style="single">Binary number</u>Expressed in<u style="single">If</u>)<u style="single">、</u>β<u style="single">Is</u>β_IDEALAccurately represent<u style="single">To do</u>thing<u style="single">But</u>Possible<u style="single">Is</u>By<u style="single">That</u>Beta is selected, and in all other cases, by rounding up β to the next largest representable number.<u style="single">That</u>β is selected. Truncation is the reason why it is always better to always round up rather than rounding to the nearest expressible number (which in some cases is truncation).<u style="single">By</u>I to Q power ratio decreased<u style="single">Resulting in</u>Because. Therefore, the power of the I channel should be increased so that its performance is relatively improved compared to the power of the Q channel. In this way for beta value selection<u style="single">Is</u>, The additional power transmitted when this β increases slightly, with a bit size that can represent β<u style="single">plural</u>To be a candidate<u style="single">about,</u>Can be calculated.

【0021】

FIG. 2 shows, as a function of the desired gain value β.<u style="single">Multiple</u>It is a graph which shows the additional transmission power (in dB) required for each candidate of the quantization size (that is, the number of bits used to represent β) of. The graph shows the representation of 3bit, 4bit, 5bit, and 6bit β. It can be seen from the graph that when the 3bit representation is used, the additional transmit power required in the worst case is 0.5 dB. By comparison, when the 4-bit representation is used, the additional transmit power required in the worst case is only 0.25 dB, and in most cases much less. As the number of bits used to represent β increases, the additional transmit power required decreases. When deciding how much β value is most convenient to use<u style="single">To</u>Another factor considered is the step size for adjusting the transmit power. In many communication systems, transmission power adjustments are made only in discrete quantities.<u style="single">。</u>In a typical WCDMA system, the minimum transmission power adjustment range is 0.25 dB. So if you use 5bit or 6bit to represent β, the additional transmission power requires at most about 0.12dB (see Figure 2), but small adjustments are not possible, so the wireless communication system In practice the power must be increased to 0.25 dB. Therefore, even if 4 bits or more are used, it only increases the complexity of multiplication and does not save additional transmission power. Therefore, it is best to use 4 bits for β.

【0022】

From the above point of view, one embodiment of the present invention is the minimum power adjustment amount ΔP._ADJ<u style="single">To</u>Predetermined<u style="single">Obtained by multiplying by a multiple K</u>Not below the value<u style="single">At the limit</u>Maximum additional transmit power requirement P_EXTRA<u style="single">Becomes the minimum</u>Bit number N_βso<u style="single">、</u>It involves quantizing β (ie, expressing β). In the above example, with K = 1, only the minimum possible power adjustment amount is required as a result of the quantization of β. However, in some other examples, N outweighs the benefits of further reducing additional transmit power._βIf you want to prioritize the benefits of keeping the value small, it is desirable to select different K values.<u style="single">N</u>_{<u style="single">β</u>}<u style="single">of</u>One method of selecting a value is shown in the flowchart of FIG. In step 301, quantization size N_βIs initialized to the minimum possible value (ie 1). Then P_EXTRABut given N_βDetermined as a function of (step 303). And P_EXTRAThe value of is K P_ADJCompared to the amount of (judgment block 305), P_EXTRAThe value of is K P_ADJWith the above (through the yes path from decision block 305), N_βThe value of is incremented by 1 (step 307). The meaning behind this step is the desired K · P_ADJIt is to try different quantization sizes to see if the additional transmit power required can be reduced without dropping below the minimum of. N_βAfter adjusting, the process returns to step 303 and the process is repeated.

【0023】

P_EXTRAThe value of is K P_ADJIf it turns out to be smaller (through the no path from decision block 305), it means that too many bits are used for the β representation. Correspondingly, N_βThe value is adjusted to return to the previously accepted level (step 309), and the β quantization level selection process ends.

【0024】

Table 1 below shows a typical set of βs selected to be close to the data rate ratio. For each ratio, the ideal value β, along with the value proposed based on using 4 bits to represent β as determined above._IDEALIt is shown.

【0025】<img he="104" id="000002" wi="150" file="2_0003575751.tif" img-format="tif" img-content="drawing" /> 【0026】

Another factor to consider in choosing the quantization size for the representation of β is related to the effect of what modulation error this brings. In many cellular telephone systems, for example, standards (these systems must operate according to this standard) place a limit on the amount of modulation error that can be accepted by any given terminal. Wireless communication systems operate under standards that place no restrictions on β values<u style="single">Suppose</u>(That is, the system is β_IDEALSuppose that is used. ) And a terminal that uses the quantized value of β<u style="single">If you used</u>Modulation error when compared to the signal expected by the system<u style="single">Will occur</u>.. Figure 4 shows<u style="single">Multiple</u>Ideal gain parameter β for each of the quantization size candidates of_IDEALIt is a graph which shows the result of the modulation error plotted as a function of. When using a 4-bit β value, the modulation error is about 6%. This is a generally accepted amount of modulation error, and it is necessary to impose strict limits on other components in the wireless terminal so as not to exceed the allowable modulation error limit.

【0027】

According to another aspect of the present invention, the problem of modulation error due to the quantization of β is rather an unrestricted ideal β._IDEALExplained by the design of the entire wireless communication system based on the β value quantized in. For each transmitter, the same quantization size for β (eg N)_βBy using = 4), the system is not subject to any modulation error due to this quantization. This has the dual effect of facilitating terminal design without degrading system performance and eliminating system inconsistencies that occur when different manufacturers quantize at different sizes.

【0028】

The technique for obtaining the gain β has been described above. In use, these βs are generated by the power ratio controller 107, as shown in FIG. The power ratio control device 107 is realized, for example, by storing one or more β values required by the digital storage device in the digital storage device. A preferred measure for the selection of β quantization values has been described as β = 4 as a particular value suitable for use in the typical WCDMA systems described above. However, it will be readily appreciated by those skilled in the art that other measures can be used to select the quantization value of β. For example, P_EXTRA K P_ADJMinimum P that satisfies the relationship_EXTRAN that produces the value of_βYou don't have to work hard to find the value all the time. Alternatively, there can be different trade-offs between the modulation accuracy preferred in other cases and the additional transmit power required. The present invention describes these<u style="single">side</u>Includes.

【0029】

The present invention has been described with reference to specific examples. However, it will be clear that the present invention can be embodied in a specific embodiment different from the preferred embodiment described above. This can be done without departing from the gist of the present invention. Suitable embodiments are for illustration purposes only and may be of any kind.<u style="single">Limited</u>Should not be considered either. The scope of the present invention is given by the claims, not by the above description, and all variations and equivalents are by the claims.<u style="single">Definition</u>It is intended to be included within the scope of the present invention.

[Simple explanation of drawings]

FIG. 1 is a block diagram of a transmitter of a wireless communication system that operates according to one aspect of the present invention.

FIG. 2 is a graph showing the additional transmit power required as a function of the desired gain value β for each of the large number of quantization size candidates.

FIG. 3 is a flowchart showing an example of a method for selecting β according to one aspect of the present invention.

FIG. 4 is the ideal gain parameter β for each of the many quantization size candidates._IDEALIt is a graph which shows the result of the modulation accuracy plotted as a function of.