JP2003531507A - Pulse shaping device for mobile communication system - Google Patents

Abstract

SUMMARY A digital telecommunications system is disclosed in which pulse shaping for the uplink and downlink is independently designed according to desired cost parameters (eg, error function for amplitude, BER, bandwidth, energy, AFC). ing. Also disclosed is a transceiver designed to operate in such a system.

Description

Detailed Description of the Invention

Background Art In digital wireless telephones, a serial bit stream of data is transmitted over the air. The bit stream is used to modulate the carrier.
There are several types of modulation schemes used to transmit the data carried by the bit stream. For example, the modulation scheme used in GSM is Gaussian minimum shift keying (GMSK) and the modulation scheme used in CDMA systems is QPSK.

GMSK is a phase modulation that converts a serial bit stream into a carrier phase shift. The function of the modulation is to convert the input serial bit stream into an analog signal that modulates the carrier of the transmitter. In GMSK,
The output phase shift is filtered. The Gaussian function acts as a filter and removes the sharp edges of the digital pulse. Without this filtering, the bandwidth required to transmit the signal would be much higher. Even with a Gaussian filter, the GMSK system has been found to be spectrally inefficient. However, GMSK modulation provides a power efficient, constant amplitude signal.

In existing CDMA systems, different phase modulation schemes QPSK have been selected to provide higher bit rates. In QPSK, MSK
A quadrature signal is transmitted that doubles the data rate compared to modulation. In QPSK modulation, the output phase shift is Nyquist filtered to provide a square root cosine shaped pulse that increases spectral efficiency, reducing bit error rate by eliminating intersymbol interference. QPSK using square root cosine pulse shaping is spectrally efficient, allows high data rates and provides low BER.

DISCLOSURE OF THE INVENTION According to one aspect of the invention, there is provided a digital telecommunications system in which first and second communication devices communicate by transmitting pulses indicative of data in accordance with a respective predetermined modulation scheme. Provided. Each of the first and second devices comprises means for shaping the data pulse before transmission, the shaping being applied according to the standards of each system. The first and second devices may be different types of communication devices, eg fixed stations and mobile stations.

In prior art modulation schemes, the pulse functions used to shape the data stream had a pre-defined mathematical relationship. Example: Square root cosine

[Equation 1] For CDMA systems where QPSK modulation is used and for PDC and NADC systems where DQPSK modulation is used.

[0006]   Gaussian type

[Equation 2] For GSM where the MSK modulation scheme is used.

In conventional pulse waveforms with predefined mathematical relationships, only one parameter is variable for a given energy level. In the case of Gaussian pulses, the "sigma" can change the pulse broadening, changing the bandwidth at the expense of amplitude. For the square root cosine, the variable is the "alpha" that changes the frequency at which the cosine tail begins. This changes the bandwidth and results in power efficiency. The relationships between cost parameters are well-defined, with one improving and the other diminishing in a given way. There is no area to improve both cost parameters.

Since the trade-off that can be realized by changing one variable for a given mathematical function is severely limited, the most suitable pulse waveform for each modulation scheme is fairly clear. . The system designer determines the modulation method based on the advantages and disadvantages of the modulation method and selects the appropriate pulse waveform. One variable of the mathematical function is designed to provide an acceptable balance in the defined relationship between cost parameters.

In the absence of a predetermined mathematical relationship to the pulse shape, as in the present invention, the pulse shape is defined to meet desired cost parameters or system criteria. You have the freedom to choose new pulse shapes that can balance many system criteria with each other. The trade-off relationship between the two parameters is less restricted.

Only Gaussian and square root cosine pulses have hitherto been considered for use in modulators of telecommunication systems. If other pulse waveforms are used, there are different costs associated with different shaped pulses. By optimizing the pulse shape with respect to system criteria, the pulse shape that best fits the system requirements can be determined. The modulation scheme can be used as one of the variables in determining the desired cost balance. A typical list of system criteria or costs includes amplitude, bit error rate (BER), energy and bandwidth. These are defined in more detail below.

(I) Adjacent channel power (ACP) amplitude variation in the modulated signal. When the amplitude is a constant value 1, the error in the amplitude is given by the following equation. {Absolute value ^{2 -1 2} 2 (ii)} BER Error Function - To calculate this spectrum occupancy is required to determine the quantity of noise. This is given by: {Absolute value of interfering region} ² (iii) Energy error function Required energy-sum of squared sample points. (Iv) Bandwidth error function

In order to estimate the bandwidth of a pulse, we need the derivative of that pulse function, which is not yet known at this stage. This derivative can be approximated as being proportional to the difference between two adjacent pulse values. By freeing the traditional constraints on pulse shape, system designers are given more freedom. In particular, in the case of the present invention, it can be used to interrogate each element of the telecommunication system independently, thus making the system as a whole more efficient.

The main operating elements in a telecommunication system are fixed and mobile stations. The considerations for each of these are somewhat different. Both are bandwidth constrained by the system in which they operate. However, while base stations are relatively free from power issues, mobile stations are limited in the amount of power they can actually transmit.

The present invention allows pulse shaping for the uplink for fixed stations and the downlink for mobile stations to be independently optimized to take into account their particular requirements. Therefore, each system standard can be designed to improve the performance of each device. Alternatively, both criteria can be optimized to suit the weakest device. The present invention can be applied to any modulation method.

One method by which a pulse waveform can be designed is described below. Laurent is able to approximate Gaussian pulses by superposition of AM pulses (C ₀ , C ₁ ..., etc.), and that these pulses are a family of fixed pulses that are a function of cosine and sine showed that.

In the embodiments of the present invention, Laurent's theory that a pulse can be approximated by superposition of components is used as a criterion for identifying a pulse waveform satisfying the criterion required by a particular communication system. Has been done. This can be done as follows. First, the fixed function component in the Laurent's superposition expansion is replaced by one or more functions representing each unknown pulse component. Next, the cost function (eg, BER, bandwidth, amplitude, AFC) is examined. That is, the error from the value required by a particular system is taken into account. The weighting of the cost function can be varied and the result adjusted. The value of each function is then determined, for example, using a (commercially available) optimizer that minimizes these cost functions and thus gives a pulse waveform that meets the defined system conditions.

It is preferable to use two functions to optimize the pulse shape. This is because it provides a more optimal pulse shaping than using only one function. More specifically, the method can be implemented as follows.

[0018]   First, consider Laurent's formula. According to Laurent's formula:

[Equation 3]

[0019]   Here, S (t) is a signal at time t.

[Equation 4]

Instead of using Laurent's pulse, C _{k, n ′} , an alternative pulse PULSE
I want to use _{KK, n '} . It is not known yet, but I would like to determine an appropriate value according to the conditions of the error function required for it. By substituting this into the equation (1), the following equation is obtained.

[Equation 5]

As mentioned above, PULSE is not yet known, but in this embodiment,
It is real, non-zero, and has a maximum length of 8. In this embodiment, we choose to use two components (PULSE [0] and PULSE [1]) to make S. Therefore, M = 2. In other embodiments, the number of component pulses themselves can be used as a variable in determining the operating state for each link.

The equation (2) is expanded for M = 2, and the function _{AK is set} to the bit stream functions ∝ ₁ , ∝ ₂ . ． ． By replacing with, the following formula is obtained.

[Equation 6]

Since ∝ represents one bit, it must be a plus or minus one. Therefore, each term in equation (3) can be identified as to whether it is real or imaginary (assuming its pulse function is real). Example: taking the first term of this _{equation: J (α N-4 +} α N-3 + α N-2 + α N-1 + α N) α N-4, α N-2, α N = Odd → imaginary number ∝ _N−3 , ∝ _N−1 = even number → real number Therefore, the absolute value of this equation can be calculated as a function of bits (∝ _S ). It is necessary to decide which ∝ to transmit at time N (in the ideal system this is the signal received in baseband).

Examine equation (3) (eg for a simple receiver) and deduce that bit ∝ _N-4 is transmitted at time (N + 4) T (because it is independent). You can It is imaginary and its interfering (ie other imaginary) pulse must be considered. The real term in this equation can be ignored for both the interference term and the absolute value of these pulses.

Interference should be minimized. For example, the term PULSE [0 in (N + 4) T
] Can be improved compared to the absolute values of all other terms to improve BER performance. Therefore, given the following ∝ sequence, {∝ _N , ∝ _N-1 , ... ∝ _N-7 } = {1,1,1,1,1,1,1,1,1}

The absolute value of the pulse at time ΔT can be calculated in terms of the unknown pulse. The absolute value of the interference term at time ΔT can also be calculated with the unknown pulse term. This is all possible combinations of 1, -1 for ∝ _N up to ∝ ₇ (ie
All 2 ⁸ = 256 possibilities). For each possibility both equations for the interference term and the absolute value are obtained. In this embodiment, the pulse must meet certain criteria for power, BER, AFC and bandwidth. Therefore, the error function for them is determined.

[0027]   Given an oversampling of 8, ΔT can have the following values:

[Equation 7] Obviously, the oversampling rate can be changed depending on the level of pulse sampling required.

The amplitude and BER cost are calculated for ΔT taking each of the above values. The total cost for each is the sum of all eight equations obtained over the possible sequences. The pulse can be specifically designed based on system requirements by weighting the above error function (eg, 0.3 for power, 0.3 for BER, and for bandwidth). 0.4, or 0 if the system only needs to consider bandwidth, for example, for power and BER, 1 for bandwidth). More weight can be added for everything that is causing problems. The only restriction is that the total weight must be equal to +1.

Here, the total error function is represented by unknown values, that is, terms of χ _{0, i} (i = _{0 to} 71) and χ _{i, j} (i = 0 to 55). In order to determine the appropriate value for the unknown, and thus infer the pulse shape, this equation is minimized using, for example, a conventional commercial optimizer. This method can be varied to make the pulse design and provides another system criterion, which is the number of pulse components used in shaping. The number of components used in each transmission direction can be different.

For a more detailed description, see British Patent Application No. 9805504 attached to this document.
． 9 and 9814300.1. This method uses a determined set of weighted cost parameters to, for example, determine optimal parameters for the uplink of a CDMA or GSM system, and then a different set of cost parameters for the downlink. Can be used to recalculate.

According to a second aspect of the invention, there is provided a transceiver for a digital telecommunication system with which first and second telecommunication devices communicate by transmitting information according to a predetermined modulation scheme, the transceiver of which is The first and second communication devices are molded prior to transmission,
A pulse indicative of data is transmitted, the transceiver having a processor for providing a data bit stream, a filter for shaping the transmitted pulse according to a first set of system standards, and a second set of predetermined system standards. A filter for receiving a data bit stream having a pulse shape determined according to a second set of system standards different from the first set of system standards.

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below by way of example with reference to the accompanying drawings. One of the first decisions to make in designing a transmitter is the duration of the phase pulse. This is really to determine the method for determining how pulse shaping is done. The duration of the phase pulse determines how much history is included in each signal. There are two transmitter elements that may be involved in pulse shaping: filters and lookup tables. The duration of the phase modulation pulse is relatively short, for example, 4T (however,
T is the duration of the symbol / bit), its "history", ie the effect of the previous data bit on the current data bit, lasts for 4T. Therefore, the tail of the previous 3 bits is superimposed on the current bit, and there are 2 ⁴ possibilities for each bit. If too much history is included, it consumes more processor time to map each of its possible histories, thus making the look-up table approach to pulse shaping impractical. However, with a fast enough adder, its output can be calculated in practice, at the cost of power consumption, among others.

In most systems, the handset or mobile unit is limited in terms of available power, one of its most important considerations being power consumption. One important factor in power consumption, and thus in busy and idle times, is the efficiency of the power amplifier. That handset is a consumer item cost and therefore complexity is also an issue. In embodiments of the invention, shaping of the pulse containing components of the transmit power for each of the uplink and downlink can be used to have a positive impact on the performance of the handset.

A description is given below with an example considering options for the design of a handset for GSM systems. Due to the large power constraints of the handset, the uplink (handset to base station) needs to be optimized for ACP (amplitude variation in the modulated signal). In a GSM system, the closer the pulse shape is to a pure Gaussian shape, the closer it is to a signal of constant amplitude.

If the handset uses the Laurent-type pulse component solution described above to shape the pulses, the more pulses of the Laurent sum used, the closer to the optimal shape. Laurent pulse shaping for handsets is usually
Achieved by calculating the shape of the Laurent pulse by optimizing for amplitude to provide a power advantage. For GSM, this means a closer approximation to the Gaussian form if amplitude is the dominant cost function.

When using a Laurent pulse to create the desired pulse shape, the larger the BT product (bandwidth × symbol period), the greater the power in the first pulse. For GSM, the BT product is 0.3, and a phase pulse of duration 4T can actually be used, the smaller the BT product, the longer the pulse duration.

If desired, other cost functions can be included in the pulse shape optimization for the handset as described above. This is the case when the efficiency of the power amplifier is only one of the parameters that is considered important. When the BT product is large, a single pulse provides enough power for signal transmission. This may be the case for Gaussian or near Gaussian shaping, which is preferred for transmission in existing GSM systems.

Another consideration other than the listed cost function for handsets is:
This is the calculation overhead for demodulation. This is the downlink (base station to handset)
It depends on the method of pulse transmission to the handset in. Quantify this,
It can be added to the optimization routine.

For the base station, the efficiency of the power amplifier is less important than for the handset.
This is because there are few restrictions on the power supply. So the designer has one choice. Transmissions at a base station can be optimized for power and other costs with weightings determined for efficient operation of that base station. However, it is advisable to consider the handset when designing the output of the base station, as the power efficiency of the handset is an important factor in the system. I mean,
This is because the handset needs to demodulate the signal, and the type and composition of the signal affects the computational overhead of the handset and consequently the power consumption.

The available power of the base station is much higher than that of the handset. This means that a Gaussian pulse is desirable, and if the pulse is produced by superposition of pulse components, a single pulse component can carry sufficient power to carry the data. The superposition of two pulses to provide an output pulse at the base station has some advantages in the power efficiency of the base station, but it complicates the transmission process and offers little advantage to the handset.

Taking the overall approach to the system, it is more advantageous to reduce the computational overhead of the handset by limiting the number of pulses transmitted by the base station. Therefore, ideally in a GSM system the number of pulses used to shape the modulated signal transmitted from the base station should be kept at one.

Conventional GSM includes a frame structure as shown in FIG. FIG. 1 illustrates a GSM transmitter suitable for use in a handset according to one embodiment of the present invention. The bit sequence 101 to be transmitted is the frame of the transmitter
Input to builder 102, which puts the bits in the appropriate portion of the burst within the time slot of the TDMA frame. Next, the bit stream modulator 1
It is transferred to 04. This conventional modulator is a GMSK modulator. However, in the preferred embodiment, the signal is not Gaussian filtered. Instead, the look-up table 106 defines a pulse function whose shape is formed from the calculation of the first two pulses of a Laurentian continuous pulse. Instead of a look-up table, it is also possible to use a multi-tap FIR digital filter whose characteristics have been calculated to provide pulse shaping corresponding to the first two Laurentian pulses of the optimization. Clock 105 provides the carrier signal as is conventional.

The modulated signal is input to the DA converter 107. This analog signal is then reconstructed by reconstruction filter 108. This filter may typically be equipped with analog filters, such as switched capacitor filters and RC filters to perform some spectral shaping, primarily to handle residual shaping. Finally, the signal is amplified by power amplifier 109 and transmitted via antenna 110.

The transmitter at the base station looks similar, but the look-up table or multi-tap FIR digital filter provides a pulse function whose shape is formed only from the first pulse of the Laurentian continuous pulse. Hold. Since the amplitude variation of a single shaped Laurent pulse is greater than that of two composite pulses,
It is unavoidable that the efficiency of the base station transmitter is no greater than that of the handset. However, this trade-off because power constraints are tolerated and the cost of providing a more linear PA (power amplifier) at the system level is less important for consumer product handsets. Is easily accepted.

FIG. 3 shows a GSM receiver according to one embodiment of the invention. Similar receivers are available to both handsets and base stations. In this embodiment, the handset is formed by the first pulse of the Laurentian continuous pulse and receives the pulse transmitted from the base station. The base station receives the pulses transmitted by the handset. This pulse consists of two Laurent type pulses to optimize power efficiency in the handset. For the Gaussian shape, most of the power is in the first pulse of the Laurentian continuous pulse, so there is usually enough power in the first Laurentian pulse that if only this pulse is decoded by the base station. It will be good. In situations where there is not enough power during the first pulse, the base station can be designed to decode both the first and second pulses of the continuous pulse.

For a receiver that decodes a single component pulse, the receiver downconverts the received signal to the bandwidth of receiver 120. Then this signal is a complex sequencer 1
21 and then to demodulator 122 to provide bit sequence 123 in a conventional manner. The principles for optimizing uplink and downlink pulse shaping in spread spectrum systems are the same as in GSM, but the cost parameters involved are somewhat complex. This is because bandwidth was the first consideration in determining the modulation scheme for spread spectrum systems, leaving much work to be done in terms of power efficiency by shaping the pulses. There is. Next, other cost parameters need to be closely monitored to provide a generally acceptable solution.

In spread spectrum systems, a unique digital code is used to distinguish mobile stations rather than separate RF frequencies or channels. The code is shared by both the mobile station and the base station and is called a pseudo-random binary code, one type of which is the Gold code.

In the introduction, the problems for various systems were explained. In a CDMA system, the typical square root cosine shaping chosen to improve the bandwidth efficiency of the output pulse has poor ACP efficiency for use in non-linear amplifiers (constant amplitude variation uses GMSK shaping). The power of the handset is very important (as it is much larger than in GSM). Several criteria can be taken into consideration when designing and considering a method of creating a pulse waveform for CDMA. The important criteria are the output energy of the signal and the efficiency of the power amplifier. In this embodiment of the invention, as in the GSM embodiment, an important design factor for the handset is the efficiency of the power amplifier in providing sufficient power for the base station to receive. The handset is portable and its power source can be carried with it, thus limiting power. Due to the large amplitude variations, the non-linearity of the power amplifier causes the spectrum to spread wider than ideal for a typical square root cosine pulse waveform.

The efficiency of the power amplifier can be improved by using the Laurent type pulse as described above and creating a pulse waveform optimized for power efficiency. This pulse shape can be modified in a complex manner to improve efficiency significantly. A problem arises as to how many Laurent type pulses should be included in pulse shaping.

In order to apply the Laurent pulse method to optimize the pulse waveform, it is necessary to determine the duration of the phase pulse. This affects BT. For a CDMA with a phase pulse duration of 6T, BT is approximately 0.15. The higher the BT product, the more power is present in the first pulse of the Laurentian series. For GSM, most of the power is in the first pulse. The BT of 0.15 is small, which means more than one pulse is needed to provide the required power for transmission. It has been found that at least two Laurent type pulses are required to provide the desired power or reduction in the effective bandwidth for a BT of 0.15.

FIG. 4 illustrates a code division multiple access (CDMA) transmitter according to one embodiment of the invention. CDMA has traditionally been a dedicated physical data channel (DPDCH).
) And a dedicated physical control channel (DPCCH). The bit sequence 301 to be transmitted is input to the transmitter's frame builder 302, which puts the bits in the appropriate portion of the frame (ie, in the DPDCH).

The bit stream is then spread over its spectrum by a Gold code encoder. Gold code encoder 303 operates as follows. The bit streams are {c ₀ , c ₁ . ． ． c _N−1 }, and the frame sequence (ie, M symbol bits) {f ₀ f ₁ .. ． ． f _M−1 }, the output of the Gold code encoder 303 is a sequence of N × M terms with the following elements: {F ₀ c ₀ , f ₀ c ₁ ,. ． ． _{f 0 c N-1, f} 1 c 0. ． ． f ₁ c _N-1 .
． ． ． ． ． f _M-1 c ₀ . ． ． f _M-1 c _N-1 } Therefore, there are MN chips to modulate.

The modulator 304 uses these MNs output by the Gold code encoder 303.
The chips are modulated onto the carrier wave output by the clock 305. C such as IS95
The modulator 304 commonly used in DMA systems is a linear QPSK modulator. The bandwidth of the signal output by the modulator 304 is determined by the lookup table 3
It is directly related to the spectrum of the pulse used to create 06. Conventionally, C
For a DMA system, this look-up table will contain the data defining the square root cosine. However, in this embodiment of the invention, the look-up table defines different pulses, the shape of which is optimized with respect to the desired cost function and represents the first two AM pulses according to Laurent's superposition theory. Created using a look-up table that holds the data, these pulses are a fixed family of pulses that are a function of cos and sin as previously described. In one example, ACP is optimized for a non-linear power amplifier.

The output of the modulator 304 is input to the DA converter 307. The analog signal is then reconstructed by reconstruction filter 308. Reconstruction filters are usually analog capacitors, such as switched capacitor filters that perform some spectral shaping, and RC filter networks that primarily handle residual spectral shaping.
Includes filters. When the signal is reconstructed, it is input to the power amplifier 309,
Power amplifier 309 amplifies the signal transmitted by antenna 310. Since the power of the signal is carried in two pulses rather than one, it is important that the receiver be able to receive both the first and second pulses in order to reconstruct the message sent by the handset. is there.

FIG. 5 is a block diagram of a spread spectrum receiver with a demodulator. In this embodiment, the receiver complements the CDMA transmitter of FIG. The receiver includes an antenna for receiving a spread signal, a frequency down-conversion circuit 401, and an AD.
It comprises a converter 402, means for storing the receiver code, and a despreader 404. The despreader comprises means 4 for transcoding the code of the receiver according to the invention.
05, a correlator 406 for correlating the received signal with the transformed code, and a comparator 407 for determining the code of the received signal.

The transcoder 405 may only include a transform to detect +1 bits. In this case, the comparator assumes that it is +1 if the value of the correlation signal output by the correlator 406 is greater than a threshold, and -1 if it is less than this threshold. I assume. However, this type of receiver requires more complex demodulation. Simpler demodulation is possible if the converter 405 also has a conversion to detect -1. In this case, the comparator determines the code that produces the maximum value. Its value is well above the noise floor, so whether the actual received signal is -1 destined for that receiver, or whether it is noise due to signals to other receivers. The complex demodulation needed to make the decision should be eliminated.

FIGS. 6 and 7 show a CDMA receiver according to one preferred embodiment, which complements a transmitter transmitting a signal constructed using superposition of multiple amplitude modulated pulses. . For this embodiment, two pulses are used. As can be seen from FIG. 6, the frequency down conversion circuit 401 has at least 1
Two IF stages 501, mixers 502a and 502b, and a low pass filter 503a.
And 503b. The received signal is passed through the IF stage 501, dropping its frequency to the frequency of baseband, then splitting the signal into I and Q components, using mixers 502a and 502b and low pass filters 503a and 503b. The carrier wave is removed from the signal. The signal is then converted from an analog signal to a digital signal by AD converters 504a and 504b, and demodulator stage 4
It is transferred to 04. FIG. 7 shows this demodulation stage in more detail.

The code converter 405 detects the +1 and −1 by converting the Gold code for both the amplitude modulation pulses that form the received signal. An example of conversion is shown below. Transform 1 (T1) 505a (for detecting +1 symbol for the first AM pulse) i = 0, 1, 2 ,. ． ． Let y _i = (− 1) ⁱ code (Gold code) for _{N−1 be} (C ₀ , C ₁ , ... C _N−1 ). here,
N is the number of elements in the sequence. i = 0 ,. ． ． For N-1, if C _i = 1 then a _i = 1 i = 0 ,. ． ． For N−1, if C _i = 0, then a _i = −1 b ₀ = a ₀ ; i = 1,2 ,. ． ． For N-1, b _i = b _i-1 + a _i ; i = 0,1,2 ,. ． ． For N-1, d _i = y _i i ^bi , and

[Equation 8]

There is another transformation that can also be used. Use the same notation for transform 1b (to detect the +1 symbol for the first AM pulse) d _i . i = 0, 1, 2 ,. ． ． D _i = i- ^bi transform 2 (T2) 505b for N−1 (to detect the −1 symbol for the first AM pulse) i = 0, 1, 2 ,. ． ． The y _i = (− 1) ⁱ codes for N−1 are {C ₀ , C ₁ ,. ． ． C _N-1 }. Here, N is the number of elements in the sequence. i = 0 ,. ． ． For N−1, if C _i = 1 then a _i = −1 i = 0 ,. ． ． For N-1, if C _i = 0, then a _i = 1 b ₀ = a ₀ ; i = 1, 2 ,. ． ． For N-1, b _i = b _i-1 + a _i ; i = 0,1,2 ,. ． ． For N-1, d _i = y _i i ^bi , and

[Equation 9]

There is another transformation that can also be used. Transform 2b (for detecting -1 symbol for the first AM pulse) i = 0, 1, 2 ,. ． ． D _i = i- ^bi transform 3 (T3) 505c for N−1 (to detect +1 symbol for the second AM pulse) i = 0, 1, 2 ,. ． ． The y _i = (− 1) ⁱ code for {N−1} is {C ₀ , C ₁ ,. ． ． C _N-1 }. Here, N is the number of elements in the sequence. i = 0 ,. ． ． For N-1, if C _i = 1 then a _i = 1 i = 0 ,. ． ． For N-1, if C _i = 0, then a _i = -1 b ₀ = a ₀ − / + a _N i = 1,2 ,. ． ． For N-1, b _i = b _i-1 + a _i −a _i−1 ; i = 0,1,2 ,. ． ． For N-1, d _i = y _i i ^bi , and

[Equation 10]

There is another transformation that can also be used. Transform 3b (for detecting -1 symbols for the second AM pulse) i = 0, 1, 2 ,. ． ． D _i = i- ^bi transform 4 for N−1 (to detect +1 symbol for the second AM pulse) i = 0, 1, 2 ,. ． ． The y _i = (− 1) ⁱ code for {N−1} is {C ₀ , C ₁ ,. ． ． C _N-1 }. Here, N is the number of elements in the sequence. i = 0 ,. ． ． For N−1, if C _i = 1 then a _i = −1 i = 0 ,. ． ． For N−1, if C _i = 0, then a _i = 1b ₀ = a ₀ − / + a _N i = 1,2 ,. ． ． For N-1, b _i = b _i-1 + a _i −a _i−1 ; i = 0,1,2 ,. ． ． For N-1, d _i = y _i i ^bi , and

[Equation 11]

There is another transformation that can also be used. Transform 4b (for detecting -1 symbols for the second AM pulse) i = 0, 1, 2 ,. ． ． D _i = i ^−bi for N−1

Similarly, correlator 406 correlates each converted code with each received signal pulse. For example, the transformed Gold code associated with the first AM pulse for detecting +1 is received by the correlator 506a as the received signal (x _i1 , x _q1 ).
Associated with the first AM pulse of. Absolute value _{z 1} of the correlation signal _{y 1} is, the comparator 4
It is transferred to 07. The same stage occurs for the first AM pulse for detecting -1, for detecting +1 and for the second AM pulse for detecting -1.

The comparator 407 determines whether the received signal is +/− 1. This compares the absolute value (z ₁ -z ₄ ) received from the comparator with the expected absolute value of the value (E ₁ -E ₄ ), assuming that the received signal is the one whose code is detected. Is done by doing. The values (E ₁ -E ₄ ) can be calculated in advance and stored in the receiver. In this embodiment, when the received signal is +1, the values of _{z 1} and _{z 3} are closer to the expected value _{E 1} and _{E 3} respectively associated, therefore, the value of _{h 1} and _{h 3} is smaller . In contrast, z ₂ and z ₄ are much smaller in value than E ₂ and E ₄ , and thus larger in h ₂ and h ₄ . Therefore, since h ₁ + h ₃ <h ₂ + h ₄ , the comparator determines that +1 has been received. Also, if -1 is received, the value of _{z 2} and _{z 4} are closer to each expected value _{E 2} and _{E 4, therefore,} h 2 + _{h 4} is reduced, whereas,
z ₁ and z ₃ are much smaller than E ₁ and E ₃ , and therefore h ₁ and h ₃ are large values. Therefore, since h ₂ + h ₄ <h ₁ + h ₃ , the comparator determines that −1 has been received. When there is almost no interference on the channel,
The receiver only needs to perform the correlation on the first pulse.

Similar conversions can be provided for other AM pulses if desired. However, it is generally acceptable to use the first two pulses. Most of the energy is in these pulses. In this case, only the first two pulses are transmitted due to the power inefficiency constraint at the handset.

Co-optimizing the pulses together generally provides a larger area for reducing the cost function. For example, in GSM, PA non-linearities are better corrected (with respect to telephone noise and spectral error) by optimizing two pulses together rather than optimizing each pulse separately. In general, two pulses tend to provide a great improvement. Theoretically N pulses can be used.

In the embodiments of the invention described above, the pulse transmitted is another variable in the system that can be used to provide maximum benefit. The pulse waveform and pulse formation are
It can be adjusted to suit the particular needs of one or more elements in the system. From the above description, it will be apparent to those skilled in the art that various modifications can be made without departing from the scope of the invention. The weighting of the cost function can be specifically changed to suit different priorities.

The invention, whether or not related to the claimed invention, or to the extent that any or all of the objects of the solution goal have been solved to a certain degree, either explicitly or in general. In the form of any novel feature or combination of novel features disclosed herein.

[Brief description of drawings]

[Figure 1]   3 is a GSM transmitter according to one embodiment of the present invention.

[Fig. 2]   It is a figure which shows the frame structure of GSM.

[Figure 3]   3 is a GSM receiver according to one embodiment of the present invention.

[Figure 4]   1 is a CDMA transmitter according to one embodiment of the invention.

[Figure 5]   1 is a CDMA receiver according to one embodiment of the invention.

[Figure 6]   1 is a CDMA receiver according to one preferred embodiment of the present invention.

[Figure 7]   3 is a demodulator stage of a CDMA receiver according to one preferred embodiment of the present invention.

[Procedure for Amendment] Submission for translation of Article 34 Amendment of Patent Cooperation Treaty

[Submission date] September 20, 2001 (2001.9.20)

[Procedure Amendment 1]

[Document name to be amended] Statement

[Name of item to be amended] Claims

[Correction method] Change

[Contents of correction]

[Claims]

─────────────────────────────────────────────────── ─── Continued front page    (81) Designated countries EP (AT, BE, CH, CY, DE, DK, ES, FI, FR, GB, GR, IE, I T, LU, MC, NL, PT, SE), OA (BF, BJ , CF, CG, CI, CM, GA, GN, GW, ML, MR, NE, SN, TD, TG), AP (GH, GM, K E, LS, MW, MZ, SD, SL, SZ, TZ, UG , ZW), EA (AM, AZ, BY, KG, KZ, MD, RU, TJ, TM), AE, AL, AM, AT, AU, AZ, BA, BB, BG, BR, BY, CA, CH, C N, CR, CU, CZ, DE, DK, DM, EE, ES , FI, GB, GD, GE, GH, GM, HR, HU, ID, IL, IN, IS, JP, KE, KG, KP, K R, KZ, LC, LK, LR, LS, LT, LU, LV , MA, MD, MG, MK, MN, MW, MX, NO, NZ, PL, PT, RO, RU, SD, SE, SG, S I, SK, SL, TJ, TM, TR, TT, TZ, UA , UG, UZ, VN, YU, ZA, ZW (72) Inventor John Low Gray             British Hampshire Earl Gee 24             8 W N Basing Stoke             Neham Fox Fur Long 74 (72) Inventor Simon Farmedge             British Surrey GU 90 Ar             Ruzette Farnham Folly Hill             Ambreside Crescent St 17 F-term (reference) 5K004 AA01 AA04 AA05 BA02 BD01                       EA04 EH02 FA05 FH03

Claims (20)

[Claims] 1. A digital telecommunication system, wherein first and second communication devices communicate by transmitting pulses indicating data according to a predetermined modulation scheme, and said first and second devices. System comprising means for shaping each data pulse prior to transmission, said shaping being applied according to each system standard. 2. The digital telecommunications system of claim 1, wherein the first and second devices are different types of communication devices. 3. The digital telecommunications system of claim 1 or 2, wherein each system standard is a standard designed to improve the performance of each device. 4. A digital telecommunications system according to any of claims 1 to 3, wherein the first device is a fixed station. 5. The digital telecommunications system of any of claims 1 to 4, wherein the second device is a mobile station. 6. A digital telecommunications system according to claim 1, wherein the means for shaping the pulses superimposes one or more pulses to provide the shaping data pulses. A system comprising means for matching. 7. The digital telecommunications system according to claim 1, wherein one of the first and second devices superimposes a first number of pulse components according to the respective system criteria. And shaping the data pulse by means of another one of the first and second devices shaping the pulse by superposition of a second number of pulse components. 8. A digital telecommunications system according to claim 1, wherein the predetermined modulation scheme is Gaussian minimum shift keying. 9. The digital telecommunications system of claim 7 or 8, wherein the first and second numbers are different from each other. 10. The digital telecommunications system of claim 9, wherein the first number is one and the second number is two. 11. The digital telecommunications system of claim 7, wherein   The system wherein the first and second numbers are two. 12. The digital telecommunications system according to claim 1, wherein the predetermined modulation scheme is QPSK. 13. The digital telecommunications system according to claim 1, wherein the system standard is one of a group including power efficiency, spectrum efficiency, bit error rate, AFC, Nyquist, and energy efficiency. Or a system characterized by being selected from more. 14. The digital telecommunication system according to claim 1, wherein one of the system standards for the first device is power consumption in the second device. system. 15. A digital telecommunications system according to claim 1, wherein one of the system criteria is the demodulation complexity of the data pulse. 16. A transceiver for a digital telecommunication system, wherein first and second communication devices communicate by transmitting information according to a predetermined modulation scheme, said first and second communication devices comprising: Transmitting a pulse indicative of the shaped data prior to transmission, the transceiver providing a data bit stream, a processor shaping the transmitted pulse according to a first set of system criteria, and a second predetermined one. A filter operating according to a second set of system standards, the filter receiving a data bit string having a pulse waveform determined according to the second set of system standards different from the first set of system standards. A transceiver characterized in that. 17. A digital communication system substantially as hereinbefore described with reference to FIGS. 1 to 3 of the accompanying drawings. 18. A digital communication system substantially as hereinbefore described with reference to FIGS. 4-6 of the accompanying drawings. 19. A transceiver substantially as hereinbefore described with reference to Figures 1 to 3 of the accompanying drawings. 20. A transceiver substantially as hereinbefore described with reference to figures 4 to 6 of the accompanying drawings.