JP2000269933A - Spread spectrum communication device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To conduct synchronization maintenance and synchronization detection for variable data rate transmission adopting a CDMA system at a low Eb/No. SOLUTION: When finger sections 102, 103, 104 apply parallel demodulation of data transmitted by 3 code series with different processing gains, an incoming wave with highest power is assigned, the one finger with the least processing gain applies generation of a reference signal and maintenance of synchronization, and the other fingers uses the synchronization timing and the reference signal to generate a synchronization timing and a reference signal for each finger, on the basis of a frequency error estimate value and a processing gain. When applying parallel demodulation to data sent by 3 code series with the same processing gain by the finger sections 102, 103, 104, one of the finger sections compares the frequency error of a referenced processing gain with others and a processing gain of data assigned to the finger section with other, to generate the reference signal and to maintain the synchronization.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a spread spectrum communication apparatus for receiving a transmitted spread spectrum signal by performing code division multiplexing using orthogonal codes.

[0002]

2. Description of the Related Art Conventionally, when transmitting data for a plurality of channels, data is generally divided and multiplexed. As a method of performing the division multiplexing, there are a frequency division multiplexing (FDM) system, a time division multiplexing (TDM) system, a code division multiplexing (CDM) system, and the like.
In the CDM system, each channel is divided by performing orthogonal transform using orthogonal codes spread in the same time-frequency space, and the data rate and weighting can be easily changed for each channel. Since it can be performed, this method is suitable for hierarchical transmission. In the broadcasting field, it is possible to perform graceful degradation by using a plurality of channels according to the CDM method and transmitting with different weights given between the channels, and switching the number of channels to be synthesized on the receiving side according to the quality of the received signal. Practical use of a digital video signal transmission system that can be performed is being studied. In the field of mobile communications, DS (Direct Seque
The IS-95 system standardized as a CDMA cellular telephone system using the spread spectrum of the nce system is known. In the IS-95 system, control channels and communication channels are classified according to the CDM system, and control information and audio information are inserted into a channel that has been orthogonally coded on the transmitting side and transmitted. Communication quality is improved by demodulating one channel containing information according to a communication procedure by RAKE reception using a plurality of fingers.

[0003] Here, the RAKE reception will be briefly described. The RAKE reception is a reception process peculiar to the spread spectrum communication system, and can perform the path diversity reception. In digital communications such as spread spectrum communication systems, the receiving side receives a direct wave from the transmitting side directly arriving at the receiving side and a reflected wave arriving at the receiving side after being reflected by a building or the like. become. In this case, since there are many paths of reflected waves, reflected waves of many paths (multipaths) are received. Therefore, on the receiving side, signals received via many paths are received, but these received signals are received with a propagation delay time due to the paths. As a result, on the receiving side, the received signals interfere with each other to cause a reception failure.

[0004] However, regarding a received signal that has undergone spread spectrum, the PN code used for spread spectrum cannot be correlated if it is offset in time.
Therefore, this is used to avoid a reception failure as follows. In the despreading unit, when a phase offset corresponding to the propagation delay time is given to the PN code and despreading is performed, only the received signal of the propagation delay time corresponding to the phase offset is subjected to despreading processing, and the other received signals are despread. Are not subjected to despreading processing. That is, by giving a phase offset corresponding to the propagation delay time to the PN code, it becomes possible to selectively perform despreading processing on each of the received signals without causing mutual interference. Therefore, a plurality of despreading units are provided in parallel, and in each despreading unit,
By performing despreading processing using a PN code having a phase offset corresponding to the propagation delay time of a received signal, a signal obtained by despreading a plurality of received signals received can be obtained independently.

[0005] A good demodulated signal can be obtained by giving a predetermined weight to the plurality of received signals thus obtained and adding and combining them. The method of receiving a spread spectrum signal in this way is the RAKE reception method, which is capable of selectively despreading and combining received signals from a plurality of paths, so that path diversity reception can be performed. Further, a W that shares a wide frequency band, which is promising as a third generation wireless access system,
There is a CDMA system, and it has been proposed to divide a channel by CDM to realize voice, data, and image communication.

[0006]

In the above-mentioned digital video signal transmission system and CDMA cellular telephone system, the number of channels assigned to one user is generally fixed, and the demodulator on the receiving side is
Demodulation is always performed for the number of channels allocated in advance. However, in the field of mobile communications, in addition to services mainly for voice and low-speed data transmission, services for high-speed data transmission are demanded, but conventional services such as high-speed data transmission by increasing the data rate per channel are required. In the spread spectrum communication, there is a problem that it is difficult to meet such a demand because an occupied band increases.

[0007] Also, without expanding the transmission band, the data is de- coded by CDMA.
As a method of performing data transmission, one
There is a way to assign to the user, but in this case the amount of interference
Will also increase. Also, the assigned code channel
The number of channels and transmission power per channel can be adaptively adjusted.
If you change it or use it with multiple users
The amount of interference is not constant, and synchronization is maintained even if the amount of interference changes
Need to be done. Furthermore, in mobile communications, selectivity and
Non-selective fading and interference from other users
Communication channel conditions change every moment,
In order to maintain stable communication quality, a certain level of E_b
/ N₀There is a problem that it is necessary to hold Was
But E_bIs the energy per bit, N₀Is the noise power
And E _b/ N₀Is the signal power standardized by the noise power
Become.

In wireless communication systems, error control techniques such as convolutional codes and Viterbi decoding are mainly used to improve communication quality. Turbo codes capable of obtaining a large coding gain have been studied. , W-
Adoption is also being considered for the CDMA system. Under static characteristics such as WGN (White Gaussian Noise) environment, the turbo code has a BER (Bit Error Rate) < _Eb / N ₀ <2 dB.
It is reported that a good property of 10 ^-5 can be obtained. However, communication at such low E _b / N ₀ is
There is a problem that it is difficult to stably maintain synchronization and perform synchronous detection. In view of the above-mentioned circumstances, the present invention enables stable synchronization and synchronization detection even when E _b / N ₀ is low, so that services other than voice and low-speed data transmission can be performed. It is an object of the present invention to provide a spread spectrum communication apparatus capable of performing high-speed data transmission with high quality even in a low E _b / N ₀ environment.

[0009]

In order to achieve the above object, a spread spectrum communication apparatus according to the present invention can identify one code sequence from a plurality of code sequences in accordance with the speed and processing gain of input data. A spread spectrum communication apparatus that generates output data by using at least one code channel to which such a code sequence is assigned, and receives the transmitted output data, based on the code sequence, And an inverse code converter for identifying one of the code sequences assigned to the known data and the code sequence assigned to the known data, and perform data demodulation on the code sequence identified by the inverse code converter. A demodulator, a frequency error estimator that calculates a frequency error amount for a transmission signal from known data in the code sequence identified by the inverse code converter, A control unit that controls a reception processing operation of the finger unit in accordance with one or more finger units having a synchronization holding unit that performs synchronization, frequency variation information detected by the frequency error estimation unit, and channel information. When the synchronization holding unit holds the synchronization of the data transmitted at the reference processing gain, the frequency error estimating unit based on the frequency error amount calculated at the reference processing gain cycle. When the synchronization is maintained and the synchronization of data transmitted with a processing gain smaller than the reference processing gain is maintained, the frequency error amounts calculated at the processing gain cycle of the data and the reference processing gain cycle are compared. If the frequency error amount calculated in the processing gain cycle of the data is large, synchronization is performed based on the frequency error amount calculated in the reference processing gain cycle. Performs lifting, when the frequency error amount calculated in processing gain cycle of the data is small, the synchronization hold based on the frequency error amount calculated in processing gain cycle of the data is to be performed.

In the spread spectrum communication apparatus according to the present invention, when the data in a plurality of code sequences are demodulated in parallel by the plurality of finger units, one of the finger units to which data having the smallest processing gain is assigned is demodulated. Performing the synchronization holding in one finger portion;
When the processing gain is the same in a plurality of finger units, the synchronization holding may be performed by any one of the finger units.

Further, in the spread spectrum communication apparatus according to the present invention, when a reference signal for performing synchronous detection of data is generated from the frequency fluctuation information detected by the frequency error estimating unit, the generated reference signal is If the reference signal is a reference signal for performing synchronous detection of data transmitted with a reference processing gain, a reference signal is generated by integrating a code sequence to which known data is assigned with a reference processing gain period. In the case where the generated reference signal is a reference signal for performing synchronous detection of data transmitted with a processing gain smaller than the reference processing gain, the integration is performed with the processing gain period of the data and the reference processing. The frequency error amount is calculated and compared with the gain cycle, and the When the frequency error amount calculated in processing gain cycle is large, the reference signal of the data transmitted by serving as a reference processing gain is less than the processing gain,
Generated based on the integral value of the reference processing gain cycle, and when the frequency error calculated by the processing gain cycle of the data is small, the reference signal is generated by integrating the processing gain cycle of the data. Is also good.

Further, in the spread spectrum communication apparatus according to the present invention, the assignment of the reference signal generation and the synchronization holding of the finger units is changed every time the arriving wave and the data processing gain assigned to each finger unit are changed. By providing the changed information to the synchronization holding unit, the phase error calculation unit in the synchronization holding unit controls an integration cycle for calculating a phase error signal, and also obtains a synchronization timing from the phase error signal. The time constant and the loop gain of the loop filter that generates the signal may be adjusted.

Further, in the spread spectrum communication apparatus according to the present invention, the control section stores a threshold value of an error signal at the time of synchronization acquisition corresponding to a processing gain, and selects the frequency from the stored threshold values. A threshold value corresponding to the error amount and the processing gain is selected, and the selected threshold value is compared with a phase error signal output from the phase error calculation unit to detect whether or not synchronization is maintained. When it is detected that the incoming signal has the highest power that has not been assigned to the finger unit, the incoming wave with the highest power may be assigned to all the finger units that are out of synchronization.

According to such a spread spectrum communication apparatus of the present invention, when synchronizing data transmitted at a reference processing gain, the frequency error is calculated based on a reference processing gain cycle. When the synchronization is maintained and the synchronization of data transmitted with a processing gain smaller than the reference processing gain is maintained, the frequency error amounts calculated at the processing gain cycle of the data and the reference processing gain cycle are compared. If the frequency error calculated in the processing gain cycle of the data is large, synchronization is maintained based on the frequency error calculated in the reference processing gain cycle, and the frequency error calculated in the processing gain cycle of the data is used. Is small, the synchronization is maintained based on the frequency error amount calculated in the processing gain period of the data. Even synchronization Yan'neru of fingers allocated it is possible to stably held.

Further, when demodulating data in a plurality of code sequences in parallel at the finger unit, synchronization is maintained in one of the finger units to which data with the smallest processing gain is assigned, and the plurality of finger units are synchronized. When the processing gains are the same in each unit, the synchronization is maintained by any one of the finger units, so that one incoming wave is not held in synchronization by a plurality of finger units, and one finger unit is used. Synchronization can be reliably maintained.

Further, when a reference signal for performing synchronous detection of data is generated from the detected frequency fluctuation information, the generated reference signal is a reference signal for performing synchronous detection of data transmitted with a reference processing gain. In the case of, a reference signal is generated by integrating a code sequence to which known data is assigned with a reference processing gain period, and a synchronous detection of data transmitted with a processing gain smaller than the reference processing gain is performed. When a signal is generated, the above integration is performed with the processing gain cycle of the data and the reference processing gain cycle, and the respective frequency error amounts are calculated and compared, and the frequency calculated by the processing gain cycle of the data is calculated. When the error amount is large, the reference signal of the data transmitted with the processing gain smaller than the reference processing gain is used. It is generated from the integral value of the reference processing gain period, and when the frequency error amount calculated in the processing gain period of the data is small, the reference signal is generated in the processing gain period of the data. The reference signal of the finger section to which the small code channel is assigned can also be generated with high accuracy.

Furthermore, the generation of the reference signal and the synchronization holding allocation to the finger unit are changed every time the processing gain of the incoming wave or data assigned to the finger unit is changed, and the changed information is given to the synchronization holding unit. Controls the integration period of the phase error detection signal of the synchronization holding unit,
Since the time constant and the loop gain of the loop filter of the synchronization holding unit are adjusted, the synchronization is held even when the number of code channels allocated in parallel by the variable data rate transmission or the processing gain of the code channel is changed. be able to. Further, the control unit stores a threshold value of the error signal at the time of synchronization acquisition corresponding to the processing gain, and holds the synchronization by comparing the frequency error amount and the threshold value selected according to the processing gain with the phase error signal. And if it detects that it is out of sync,
Since the arriving wave with the largest power not assigned to the finger part is assigned to all the finger parts that are out of synchronization, if synchronization is lost,
Immediately, synchronization of the arriving wave having the highest power can be obtained.

[0018]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Before describing the spread spectrum communication apparatus of the present invention, a schematic configuration of a spread spectrum transmission apparatus for transmitting a spread spectrum signal will be described with reference to FIG. As shown in FIG. 6, the spread spectrum transmitting apparatus includes a transmission data selector 601, a transmitting section 650, an adding section 611, and a spreading section 612. The transmission unit 650 includes a plurality of data transmission units of a first data transmission unit 613, a second data transmission unit 606,..., An n-th data transmission unit 607, and includes a first data transmission unit 613 and a second data transmission unit 613.
The transmission data assigned by the transmission data selector 601 is input to the data transmission units 606,..., N-th data transmission unit 607. The transmission data selector 601 assigns a plurality of input transmission data having different data rates and communication qualities to the first data transmission unit 613 to the n-th data transmission unit 607 according to the format information. The first data transmission unit 613, the second data transmission unit 606,..., And the n-th data transmission unit 607 have the same configuration.
It is shown in FIG.

The first data transmitting section 613 will be described.
A code symbol is adaptively generated with the time slot and the processing gain set in 1, and the QPSK modulator 603 to which the code symbol is input performs QPSK modulation of the code symbol and outputs the modulation symbol. I have. Further, multiplication section 604 multiplies the input modulation symbol by the assigned orthogonal code 1, and outputs the orthogonally coded modulation symbol output from multiplication section 604 to the gain assigned to the modulation symbol in amplification section 605. Amplified by The orthogonally coded modulation symbols output from the amplification section 605 are input to the addition section 611.

In the second data transmission section 606 to the n-th data transmission section 607 having the same configuration as the first data transmission section 613, the transmission data allocated and input to each is QPSK-modulated, and the modulation thereof is performed. The symbols are orthogonally coded with orthogonal codes 2 to n different from each other. Next, the signal is amplified and input to the addition unit 611, where the first data transmission unit 61 is added.
3. The orthogonally coded modulation symbols from the second data transmission unit 606,..., N-th data transmission unit 607 are added. As described above, the first data transmission unit 613, the second data transmission unit 606 to the n-th data transmission unit 607 provide n
A code channel of the channel is formed.

The transmitting section 650 is provided with a pilot signal transmitting section 620. In the pilot signal transmitting section 620, known data (for example, all “1”) is modulated by the modulating section 608, The modulation symbol is multiplied by the orthogonal code 0 assigned to the pilot channel in the multiplier 609 and orthogonally encoded. Next, the orthogonally coded modulation symbol is supplied to the amplifying section 610.
Is amplified by the gain assigned to the modulation symbol, and is added to another orthogonally coded modulation symbol in the adding section 611. The addition output from addition section 611 is spread in spectrum by PN code assigned for DS-SS in spreading section 612. This spread spectrum DS-SS signal is transmitted. Note that the pilot signal is transmitted with higher power than other code channels, and by receiving the pilot signal, it is possible to synchronize and detect a situation of multipath and a situation of interference from multiple users.

A spread spectrum communication apparatus according to the present invention comprises:
For example, it can receive a spread spectrum signal transmitted from the above spread spectrum transmitting apparatus. Hereinafter, the spread spectrum receiving apparatus of the present invention according to the embodiment of the spread spectrum communication apparatus of the present invention,
This will be described with reference to the drawings. FIG. 1 shows a schematic configuration of an embodiment of a spread spectrum receiving apparatus according to the present invention. The spread spectrum receiving apparatus of the present invention shown in FIG.
It has two finger parts. In FIG. 1, a searcher 101 has a communication path such as a phase offset of a spread code sequence (hereinafter, referred to as a “PN code”) determined by a relative delay time of an incoming wave having a propagation delay of an input received signal and a received power. In addition to measuring the situation, a phase offset and a code sequence (hereinafter referred to as “orthogonal code”) number allocated for code multiplexing are assigned to the first to third finger units 102 to 104 and the channel combining unit 105.

The first finger unit 102, the second finger unit 103, and the third finger unit 104 constitute a RAKE receiving unit 110 together with a channel synthesizing unit 105.
Arrival waves and code channels are assigned to the first finger section 102, the second finger section 103, and the third finger section 104, respectively. First finger portion 102,
The second finger unit 103 and the third finger unit 104 perform despreading, inverse orthogonal transformation, demodulation, reference signal generation, and synchronization holding of the arriving wave and the code channel allocated thereto.

The channel synthesizing section 105 demodulates finger demodulated symbols output from each of the first finger section 102, the second finger section 103, and the third finger section 104 by RAKE reception, and performs demodulation of a plurality of code channels. In each case, weighting and combining are performed for each finger demodulation symbol in accordance with the timing assigned by the searcher 101. The system control unit 106 searches the searcher 101, RA according to channel conditions detected by the searcher 101 or the RAKE receiving unit 110 and channel information such as the number of code channels, the transmission power amount for each channel, and data weighting.
The reception processing operation of the KE reception unit 110 is controlled.

The first finger portion 102, the second finger portion 103, and the third finger portion 104 have the same configuration, and a schematic configuration thereof is shown in FIG. In FIG. 2, a sampling unit 201 oversamples an input received signal at twice or more the chip rate, and samples with a clock of a chip rate whose synchronization timing matches in accordance with a synchronization timing signal supplied from a synchronization holding unit 209. And outputs the result to the first despreading unit 203. Further, the signal is sampled by a clock having a chip rate whose synchronization timing is shifted by し, and is output to the second despreading unit 208. The PN generation unit 202 is provided for the system control unit 10
6, the respective finger portions 102, 10
P according to the phase offset assigned to 3,104
The N code is generated, and the generated PN code is supplied to the first despreading unit 203 and the second despreading unit 208. The first despreading unit 203 performs despreading of a signal synchronized with the chip rate clock output from the sampling unit 201 by using a PN sequence output from the PN generation unit 202.

The orthogonal code generator 204 includes a system control unit 1
The corresponding orthogonal code is generated according to the phase offset and code channel number assigned to each finger known from 09, and is supplied to the inverse orthogonal transform unit 205. The inverse orthogonal transform unit 205 performs an inverse orthogonal transform on the despread signal output from the despreading unit 203 using an orthogonal code output from the orthogonal code generator 204, and extracts a signal of a corresponding code channel. The reference signal generation section 206 is a single code channel (hereinafter, referred to as a “pilot channel”) to which known data is assigned.
, The carrier error is obtained by obtaining the amplitude error and the phase error between the transmission and reception signals. The reference signal generated by the carrier reproduction is output and the demodulation unit 207 is output.
To supply. Further, it outputs integration information to synchronization control section 210 in accordance with information provided from synchronization control section 210.

The demodulation unit 207 includes a first inverse orthogonal transform unit 205
, And demodulates by performing synchronous detection using the carrier-reproduced reference signal output from the reference signal generation unit 206 and outputs a finger demodulated symbol. The second despreading unit 208 despreads the signal whose synchronization timing is shifted by に 対 し て with respect to the chip rate clock output from the sampling unit 201 by the PN sequence output from the PN generation unit 202,
An E-ch signal advanced by 1/2 chip phase and an L-ch signal delayed by 1/2 chip phase are generated and the synchronization holding unit 209 is generated.
Output to Synchronization holding section 209 includes despreading section 208
The phase error between transmission and reception is detected from the phase amount obtained by integrating the E-ch signal and the L-ch signal input from the control unit 210 with the number of integrations according to the control signal given from the synchronization control unit 210. Further, the detected phase error signal is filtered to generate and output a synchronization timing signal, and a lock detection signal indicating whether or not synchronization is maintained is generated and output based on the magnitude of the phase error. Synchronization control section 210 is based on a system control signal such as data format information and finger assignment information of data transmitted from system control section 106 and information output from reference signal generation section 206 and synchronization holding section 209. , And generates a synchronization control signal necessary for maintaining synchronization.

Next, a schematic configuration of the reference signal generating section 206 is shown in FIG. In FIG. 3, an orthogonal code generator 301 generates an orthogonal code according to the phase offset assigned to the finger units 102, 103 and 104 and the code channel number assigned to the pilot channel, and supplies the orthogonal code to the inverse orthogonal transform unit 302. . The inverse orthogonal transform unit 302 performs an inverse orthogonal transform on the despread signal output from the first despreading unit 203 using an orthogonal code assigned to a pilot channel output from the orthogonal code generator 301. The Walsh number is assigned to the pilot channel.

When the orthogonal code [0] is assigned, the orthogonal code generator 301 and the inverse orthogonal transformer 302 can be omitted. The integral dump unit 303 performs an integral dump in which the inverse orthogonal transformed signal of the pilot channel output from the inverse orthogonal transform unit 302 is integrated according to the number of integrations given from the integral control unit 307. The integral dump value output from the integral dump unit 303 is supplied to and stored in the symbol adding unit 304, and symbol addition for adding the integral dump value is performed according to the symbol addition number output from the integral control unit 307. The sum is output as a symbol sum.

The frequency error estimating unit 305 monitors the phase change between the integrated dump value sent from the integrating dump unit 303 and the symbol added value sent from the symbol adding unit 304 for each integration cycle. , A frequency error is estimated from the amount of phase change for each integration cycle and output. An example of the operation of the frequency error estimating unit 305 will be described later. The reference signal setting unit 306 generates a reference signal based on the integrated dump value output from the integration dump unit 303 and the symbol addition value output from the symbol addition unit 304 in accordance with the frequency fluctuation information from the integration control unit 307. Generate and output. An example of the operation of the reference signal setting unit 306 will also be described later. The integration control unit 307 includes the synchronization control unit 2
A necessary control signal is generated based on the synchronization control signal given from the controller 10 to output integration information.

Next, a schematic configuration of the synchronization holding section 209 is shown in FIG. In FIG. 4, an orthogonal code generator 401 generates an orthogonal code according to a phase offset assigned to the finger units 102, 103, and 104 and a code channel number assigned to a pilot channel, and generates a first inverse orthogonal transform unit 402 and a second inverse orthogonal transform unit. It is supplied to the inverse orthogonal transform unit 403. The first inverse orthogonal transform unit 402 includes a second inverse spread unit 2
The inverse orthogonal transform is performed on the inverse spread signal output from the orthogonal code generator 401 by the orthogonal code output from the orthogonal code generator 401. Also, the second inverse orthogonal transform unit 403 performs an inverse orthogonal transform on the inverse spread signal output from the inverse spread unit 208 and delayed by a half chip phase using the orthogonal code output from the orthogonal code generator 401. I have.

E-ch integral dump section 404 integrates the inverse orthogonally transformed signal of the pilot channel advanced in half chip phase output from inverse orthogonal transform section 402 in accordance with the number of integrations given from timing control section 410. Then, the integrated dump value is output to the phase error calculating unit 406. Further, L-ch integral dump section 405 outputs 1 / ch output from second inverse orthogonal transform section 403.
An integral dump is performed to integrate the inverse orthogonally transformed signal of the pilot channel delayed by two chips in accordance with the number of integrations given from the timing control unit 410, and the integrated dump value is output to the phase error calculation unit 406. The phase error calculation unit 406 calculates E-
The integral dump value from the ch integral dump unit 404 and １ ／
The transmission power is obtained from the integral dump value from the L-ch integral dump unit 405 with a delayed chip phase, and the phase error is calculated from the difference between the former and latter integral dump values.

The threshold setting unit 407 is provided for the timing control unit 4
A threshold for determining whether to be locked or unlocked, which is set based on the data format information and the finger assignment information given from 10 according to the frequency fluctuation width of the received signal. I have. The lock detecting unit 408 compares the phase error signal output from the phase error calculating unit 406 with the threshold set by the threshold setting unit 407, and if the magnitude of the phase error is within the threshold, the lock is held. A lock detection signal indicating that the synchronization is not maintained is output if the magnitude of the phase error exceeds the threshold. Further, the loop filter 409 performs filtering of the phase error signal output from the phase error calculation unit 406 according to the filter gain value set based on the data format information and the finger assignment information given from the timing control unit 410, and is generated. Output the synchronized timing information.

The loop filter 409 is a low-pass filter, and can generate a clock synchronized with the clock on the transmission side by supplying the output synchronization timing information to the voltage-controlled oscillator. This loop filter 4
FIG. 5 shows a schematic configuration of the embodiment 09. In FIG. 5, amplifying sections 501, 503, and 505 include timing control section 410.
Filter gain values k1, k set in gain setting section 506 based on data format information and gain information comprising finger assignment information given from
2, k3, the input signal is amplified. The phase error signal amplified by the amplifying unit 501 is integrated by the integrating dump unit 502, and the integrated dump value is amplified by the amplifying unit 503. The output from the amplification unit 501 and the output from the amplification unit 503 are added in the addition unit 504, and the added output is amplified in the amplification unit 503 and output as a synchronization timing signal. The filter gain values k1, k2, and k3 are also changed according to the data symbol rate.

The operation of maintaining the synchronization of the thus configured spread spectrum receiving apparatus will be described with reference to the data format example of FIG. FIGS. 7A, 7B, and 7C each show an example of the transmission data format, and respectively show a pilot channel (Pilot Ch) and a plurality of data channels (Data Ch.1, Data Ch.2, Data Ch.3). ), And each channel can be identified by an orthogonal code. Further, three types of processing gains (three types of one symbol period lengths) are assigned to the respective data channels, and symbols corresponding to the respective processing gains are a [x], b [x], c [x]. (However, x =
0, 1, 2,...), The finger demodulation symbol periods Ta [x], T corresponding to the respective processing gains
When b [x] and Tc [x] are based on the symbol a [x], the cycle of Ta [x] = a [x] Tb [x] = the cycle of b [x] = 2 × a [ x] period Tc [x] = c [x] period = 4 × a [x] period. That is, since the processing gain of the symbol b [x] is twice as large as that of the symbol a [x], the period Tb
[X] is twice the period Ta [x] and the symbol c
Since the processing gain of [x] is four times that of the symbol a [x], the cycle Tc [x] is four times the cycle Ta [x].

In this case, since the power per chip is assumed to be equal in each case, the power Pa [x], Pb [x], Pc [x] per finger demodulation symbol is based on the symbol a [x]. , Pa [x] = power of a [x] Pb [x] = power of b [x] = power of 2 × a [x] Pc [x] = power of c [x] = 4 × a [x] Power. The reference processing gain cycle is the cycle Tc [x] of the symbol c [x] having the longest cycle length.
Let p be the power of Pp = 2 × c [x]. That is, a pilot channel is assigned to one code channel so that demodulation by synchronism holding and synchronous detection can be performed, and the power of this pilot channel is made larger than the power of other code channels. As a result, synchronization can be maintained even at a low E _b / N ₀ , and even under this condition, each code channel can perform demodulation by synchronous detection.

In the example of the transmission data format shown in FIG. 7A, the number of code channels to be allocated is set to 3 code channels.
To the third finger unit 104, the symbols a [x] having the same processing gain are transmitted in parallel. Therefore, data channel 1 (Data Ch.1) to data channel 3 (Dat
a Ch.3) are assigned to the first finger section 102 to the third finger section 104, respectively, to perform parallel demodulation. In this transmission data format example, since the power per symbol of the pilot channel is 0 dB, the power per symbol of data channel 1 (Data Ch.1) to data channel 3 (Data Ch.3) is − 9 dB. In FIG. 7B, the number of code channels allocated is two code channels, and when demodulating, the second finger unit 103 and the third finger unit 1 are used.
04 is assigned to data channel 2 (Data Ch.2), and RAKE reception of symbol a [x] is performed. Data channel 1 of symbol c [x] having a processing gain different from that of data channel 2 is provided to first finger section 102. Data Ch.
1) Assign. Then, the first finger portions 102 to
The third finger unit 104 performs demodulation in parallel. In this transmission data format example, since the power per symbol of the pilot channel is set to 0 dB, the power per symbol of the data channel 1 (Data Ch.1) is -3 dB, and the data channel 2 (Data Ch. The power per symbol in 2) is -9 dB.

Further, FIG. 7 (c) shows that the number of code channels allocated is three code channels. When demodulating, the first finger 102 to the third finger unit 104 have symbols a [x] having different processing gains respectively. , Symbol b
Data channel 1, data channel 2 and data channel 3 of [x] and symbol c [x] are allocated to perform demodulation in parallel. In this transmission data format example, since the power per symbol of the pilot channel is 0 dB, the power per symbol of the data channel 1 (Data Ch. 1) is -3 dB.
And the power per symbol of the data channel 2 (Data Ch.2) becomes -6 dB, and the data channel 3
The power per symbol of (Data Ch.3) is -9 dB.

A description will be given of reference signal generation and synchronization holding in an example of a data format to which three code channels are assigned as shown in FIG. In this case, the first finger portion 102 to the third finger portion 104
Are performed in parallel on three data channels by the three fingers. From this, the first searcher 101
The phase offset of the incoming wave assigned to the finger units 102 to the third finger unit 104 is the phase offset of the incoming wave with the highest received power. At this time, since the same pilot channel is followed even when the synchronization is maintained independently by three fingers, the reference signal generation and the synchronization maintenance may be performed by one finger of the three fingers, and the other fingers may be any one of the three fingers. In this case, the reference signal generated by one finger and the synchronization timing may be used. For example, the first finger unit 1
02 reference signal generation unit 206 and synchronization holding unit 20
9, the reference signal and the synchronization timing are generated, and the demodulation of the data channel 1 is performed by the synchronous detection. The second finger unit 103 and the third finger unit 104
The demodulation of data channel 2 or data channel 3 may be performed by synchronous detection with reference to the reference signal generated by the finger unit 102 and the synchronization timing. This control is performed by the system control unit 106.

A description will be given of reference signal generation and synchronization holding in a data format example in which two code channels are allocated as shown in FIG. in this case,
The second finger unit 103 and the third finger unit 104 perform RAKE reception of the data channel 2, and the remaining first finger unit 102 demodulates the data channel 1. From this, the phase offset of the arriving wave assigned from the searcher 101 is
The second finger unit 103 is assigned the phase offset of the arriving wave with the highest received power, and the third finger unit 104 is assigned the arriving wave with the second highest received power. At this time, since the first finger unit 102 and the second finger unit 103 follow the same pilot channel with different processing gains, the first finger unit 102 and the second finger unit 103 perform the reference signal generation and the synchronization holding by the second finger unit 103 having the small processing gain. The first finger unit 102 performs demodulation of the data channel 1 by synchronous detection with reference to the reference signal generated by the second finger unit 103 and the synchronization timing. Also, the third finger unit 104 independently generates a reference signal and maintains synchronization since incoming waves having different phase offsets are assigned. These controls are performed by the system control unit 106.

A description will be given of reference signal generation and synchronization holding in a data format example in which three code channels are allocated as shown in FIG. in this case,
1st finger part 102 to 3rd finger part 104-3
Since the three data channels are demodulated in parallel by the fingers, the phase offset of the arriving wave assigned from the searcher 101 to the first finger unit 102 to the third finger unit 104 is the phase offset of the arriving wave having the highest power. At this time, since the same pilot channel follows even if the synchronization is maintained independently by the three fingers, the reference signal generation and the synchronization maintenance are performed by any one finger, and the other fingers perform synchronization with the synchronization timing and the reference signal. Should be used.
For example, the reference signal generation unit 206 and the synchronization holding unit 209 of the third finger unit 104 having a small processing gain generate a reference signal and a synchronization timing, and demodulate the data channel 3 by synchronous detection. The second finger unit 103 refers to the reference signal and the synchronization timing generated by the third finger unit 104 and performs demodulation of the data channel 1 or the data channel 2 by synchronous detection. This control is performed by the system control unit 106.

Next, an example of frequency error estimation performed by the frequency error estimator 305 will be described with reference to FIG. Here, there is no fluctuation due to noise or fading in the communication path,
The timing difference between transmission and reception indicates a case where only the accuracy of the clock is used. At this time, the phase of the integration dump value output from the integration dump unit 303 that integrates the pilot channel signal constantly changes as indicated by the dotted line by the clock error. That is, as shown in FIG. 8, the phase θa [x] of the integral dump value corresponding to the symbol a [x]
Is equal for all symbols a [x]. Here, assuming that the integration period, which is the unit of the integration dump unit 303, is the period of the symbol a [x], the phase of the integration dump value phase θb [x] corresponding to the symbol b [x] is the symbol addition unit 304 Is the phase after adding the integral dump value Sa [x] and the integral dump value Sa [x + 1].
[X] + θa [x + 1]. Further, the phase θc [x] of the integral dump value corresponding to the symbol c [x]
Of the integrated dump value Sa
[X], Sa [x + 1], Sa [x + 2], Sa [x +
3], which is the phase after adding the four values of θa [x] + θa
[X + 1] + θa [x + 2] + θa [x + 3]. Symbol a [x] which is this unit
The number of times the integration dump value is added in the integration cycle of is set as the symbol addition number. Therefore, as shown in FIG. 7B and FIG. 7C, when calculating the frequency error of the data channel through which symbols having different processing gains (symbol periods) are transmitted, integration of the data channel having the smallest processing gain is performed. After the dump value is calculated by the integral dump unit 303, the symbol adding unit 304 only needs to add the number of symbol additions corresponding to the processing gain. The frequency error can be obtained by differentiating the amount of phase change.

The reference signal is generated based on the integral dump value output from the integral dump section 303 or the symbol addition value output from the symbol adding section 304 as shown in FIG. An example of a reference signal set value set by the reference signal setting unit 306 will be described with reference to FIG. FIG. 9A shows an example in which the frequency fluctuation is small. This is a case where the fluctuation due to the noise or fading of the communication channel is small. The phase change almost coincides with the case where the timing shift between transmission and reception indicated by the dotted line is only the accuracy of the clock. Therefore, the symbols a [x], b
The reference signal setting unit 306 sets the integration dump value or the symbol addition value of each integration cycle of [x] and c [x] as a reference signal setting value,
Output as a reference signal.

That is, the phase θb [x] of the integration dump value of the integration cycle corresponding to the symbol b [x] is θa [x]
+ Θa [x + 1], which is almost equal to the phase, and the phase θ of the integration dump value of the integration cycle corresponding to the phase symbol c [x].
c [x] is θa [x] + θa [x + 1] + θa [x +
2] + θa [x + 3]. In addition,
θa [x], θa [x + 1], θa [x + 2], θa
[X + 3] is the phase of the integration dump value of the integration cycle corresponding to each phase symbol a [x]. Further, the number of integrations of the synchronization holding unit 209 also depends on the finger units 102 and 1.
An integration cycle determined by the processing gain of the data channel assigned to each of 03 and 104 is assigned.

On the other hand, FIG. 9B shows a case where the fluctuation due to the noise or fading of the communication channel is large, and the phase change of the integral dump value in the integral dump unit 303 of the pilot channel for each integration cycle is indicated by a dotted line. As shown in the figure, the timing deviation between the transmission and reception is greatly different from the case where only the clock accuracy is used. One of the frequency fluctuations is Doppler effect in mobile communication. Doppler frequency in this wireless communication is 100Hz
Although the data rate of high-speed data transmission is 1 Mb
The data rate is much higher than the Doppler frequency because the transmission rate is equal to or more than ps and the data rate per channel is several tens of kbps or more even in transmission using a plurality of code channels. Therefore, the frequency fluctuation hardly affects within one symbol period. Most of the noise is WGN (White Gaussian Noi
Since the signal including noise is integrated, the average value of the noise component becomes zero.

From these, the phase θa [x] of the integral dump value for the symbol a [x] having a small processing gain is obtained.
Has a smaller number of integrations than the phase θc [x] of the integration dump value for the symbol c [x] having a large processing gain, and therefore contains a large error component. Therefore, the symbol dumping unit 304 adds the integral dump value Sa [x] for four symbols with reference to the integral dump value Sc [x] having a large processing gain, and obtains the integral dump value S obtained by performing the addition.
From c [x], Sa '[x] = Sc [x] / 4 is calculated. Then, instead of the integral dump value Sa [x], the averaged integral dump value Sa ′ [x] is set by the reference signal setting unit 306 as a reference signal setting value. This reference signal set value Sa ′ [x] is
The reference signal is output from reference signal setting section 306 as a reference signal for symbol a [x]. Thus, the error component can be reduced as compared with the case where the integral dump value Sa [x] is used as the reference signal of the symbol a [x]. Similarly, for the symbol b [x], the symbol adder 30
From the integrated dump value Sc [x] obtained by adding the integrated dump value Sa [x] for 4 symbols at 4, Sb ′
By setting [x] = Sc [x] / 2, the reference signal set value Sb ′ having a small averaged error component is set.
[X] can be obtained.

The number of integrations in the E-Ch integration dump unit 404 and the L-Ch integration dump unit 405 of the synchronization holding unit 209 is controlled as follows. Compared with the phase change of the integral dump value of the pilot channel for each integration cycle and the timing shift between transmission and reception indicated by the dotted line in FIG. 9 which is only the clock accuracy, the phase shift between the two is smaller. If the fluctuation due to noise or fading in the communication channel is small, the integration period corresponding to the processing gain assigned to each of the finger units 102, 103, and 104 is set to E−
Ch integration dump unit 404 and L-Ch integration dump unit 4
05 is set and synchronization is maintained. If the phase shift between the two is large and the fluctuation due to noise or fading of the communication channel is large, the finger units 102, 103, 10
Regardless of the processing gain of the data channel assigned to No. 4, the integration cycle (the cycle of the symbol c [x] in the above example) determined by the reference processing gain is set to the E-Ch integration dump unit 404 and the L-Ch It is set in the integration dump unit 405 to maintain synchronization. This control is performed by the system control unit 106.

Next, FIG. 10 shows an example of a lock detection threshold value set in the lock detection unit 408, and the operation of maintaining synchronization will be described. The phase error signal output from the phase error calculation unit 406 is within the range of the threshold 1 or the threshold 2 which is the target value shown in FIGS. 10A and 10B when the synchronization is successfully acquired and the synchronization is maintained. To perform synchronous tracking. That is,
The threshold 1 or the threshold 2 is set in the lock detection unit 408 and is set as the lock detection threshold. The range of the lock detection threshold is different between the threshold 1 and the threshold 2. This is for the following reason. When the frequency variation is small as shown in FIG. 10A and when the frequency variation is large as shown in FIG. Will be different. Also, the error amount differs depending on the processing gain. from these things,
This is because it is necessary to match the lock detection threshold with the frequency error amount that can be tracked and the processing gain. Therefore, a plurality of thresholds are prepared, and the frequency error estimation information and the finger unit 1 are prepared.
The threshold setting unit 407 estimates the optimum threshold value in accordance with the processing gain of the data allocated to 02, 103, and 104, and performs lock detection by setting the optimum lock detection threshold value in the lock detection unit 408. ing.

When lock detection section 408 outputs a lock detection signal indicating that synchronization has been lost, the arriving wave having the highest power among arriving waves not assigned to finger sections 102, 103, and 104 is output. Is assigned to all the finger portions that are out of synchronization.
As a result, when synchronization is lost, synchronization of the arriving wave with the strongest power can be immediately obtained. further,
Amplifying sections 501, 503, in loop filter 409
Filter gain values k1, k2, k3 set to 505
Similarly, a plurality of filter gain values are prepared, and the frequency error estimation information and the finger units 102 and 10 are prepared.
An optimum value is estimated in accordance with the processing gain of the data allocated to 3,104, and an optimum filter gain value is set. These controls are performed by the system control unit 106
Do.

In the above-described spread spectrum communication apparatus according to the embodiment of the present invention, a pilot channel is allocated to one code channel so that synchronization can be maintained and demodulation by synchronous detection can be performed. In the case of such a configuration, the transmission power of the pilot symbol transmitted on the pilot channel is made larger than that of the other data channels, so that synchronization can be maintained even under a low E _b / N _0. Under such conditions, each code channel can perform demodulation by synchronous detection. In the above description, the pilot channel is always transmitted. However, the present invention is not limited to this, and pilot symbols may be inserted and transmitted in a time-division manner.

In the above description, the number of finger units on the receiving side is set to three. However, the present invention is not limited to this, and any number of finger units can be used. Further, the number of modulating units is not limited to three on the transmitting side, and an arbitrary number can be provided in parallel.
Furthermore, even when data transmission is performed using a plurality of orthogonal code channels by inserting pilot symbols in a time-division manner, the pilot symbol insertion section estimates a frequency error according to the processing gain of the data channel,
By setting a reference signal and interpolating a frequency error with a reference signal set for each data channel in other sections, it is possible to maintain synchronization and generate a reference signal.

[0051]

According to the spread spectrum communication apparatus of the present invention, when synchronizing data transmitted with the reference processing gain as described above, based on the frequency error amount calculated in the reference processing gain cycle. When the synchronization is maintained and the data transmitted at a processing gain smaller than the reference processing gain is maintained, the frequency error amount calculated for the data processing gain cycle and the frequency error amount calculated at the reference processing gain cycle are compared. If the frequency error calculated at the processing gain cycle of the data is large, synchronization is maintained based on the frequency error calculated at the reference processing gain cycle, and the frequency error calculated at the processing gain cycle of the data is maintained. When the amount is small, the synchronization is maintained based on the frequency error amount calculated in the processing gain period of the data. Even synchronization tunnel of fingers allocated it is possible to stably held.

Further, when data in a plurality of code sequences are demodulated in parallel by the finger unit, synchronization is maintained in one of the finger units to which the data with the smallest processing gain is assigned, and the plurality of finger units are synchronized. When the processing gains are the same in each unit, the synchronization is maintained by any one of the finger units, so that one incoming wave is not held in synchronization by a plurality of finger units, and one finger unit is used. Synchronization can be reliably maintained.

Further, when a reference signal for performing synchronous detection of data is generated from the detected frequency fluctuation information, the generated reference signal is a reference signal for performing synchronous detection of data transmitted with a reference processing gain. In the case of, a reference signal is generated by integrating a code sequence to which known data is assigned with a reference processing gain period, and a synchronous detection of data transmitted with a processing gain smaller than the reference processing gain is performed. When a signal is generated, the above integration is performed with the processing gain cycle of the data and the reference processing gain cycle, and the respective frequency error amounts are calculated and compared, and the frequency calculated by the processing gain cycle of the data is calculated. When the error amount is large, the reference signal of the data transmitted with the processing gain smaller than the reference processing gain is used. It is generated from the integral value of the reference processing gain period, and when the frequency error amount calculated in the processing gain period of the data is small, the reference signal is generated in the processing gain period of the data. The reference signal of the finger section to which the small code channel is assigned can also be generated with high accuracy.

Further, the generation of the reference signal to the finger unit and the synchronization holding allocation are changed every time the processing gain of the incoming wave or data assigned to the finger unit is changed, and the changed information is given to the synchronization holding unit. Controls the integration period of the phase error detection signal of the synchronization holding unit,
Since the time constant and loop gain of the loop filter of the synchronization holding unit are adjusted, the synchronization holding is stable even when the number of code channels allocated in parallel due to variable data rate transmission or the processing gain of the code channel is changed. You can do it. Further, the control unit stores a threshold value of the error signal at the time of synchronization acquisition corresponding to the processing gain, and holds the synchronization by comparing the frequency error amount and the threshold value selected according to the processing gain with the phase error signal. Is detected, and when it is detected that synchronization has been lost, the arriving wave with the highest power not allocated to the finger part is allocated to all of the finger parts that are out of synchronization. Therefore, when the synchronization is lost, it becomes possible to immediately acquire the synchronization of the arriving wave having the highest power.

[Brief description of the drawings]

FIG. 1 is a diagram showing a schematic configuration of an embodiment of a spread spectrum receiving apparatus according to the present invention.

FIG. 2 is a diagram showing a schematic configuration of a finger unit according to the spread spectrum receiving apparatus of the present invention.

FIG. 3 is a diagram illustrating a schematic configuration of a reference signal generation unit according to the spread spectrum receiving apparatus of the present invention.

FIG. 4 is a diagram illustrating a schematic configuration of a synchronization holding unit according to the spread spectrum receiving apparatus of the present invention.

FIG. 5 is a diagram showing a schematic configuration of a loop filter according to the spread spectrum receiving apparatus of the present invention.

FIG. 6 is a schematic configuration diagram of a spread spectrum transmitting apparatus.

FIG. 7 is a diagram illustrating an example of a transmission data format according to the spread spectrum receiving apparatus of the present invention.

FIG. 8 is a diagram illustrating an example of estimating a frequency error in a frequency error estimating unit according to the spread spectrum receiving apparatus of the present invention.

FIG. 9 is a diagram illustrating a reference signal setting example set in a reference signal setting unit according to the spread spectrum receiving apparatus of the present invention.

FIG. 10 is a diagram showing a setting example of a lock detection threshold value set in a lock detection unit according to the spread spectrum receiving apparatus of the present invention.

[Explanation of symbols]

 Reference Signs List 101 searcher 102, 103, 104 finger unit 105 channel synthesis unit 106 system control unit 201 sampling unit 202 PN generation unit 203, 208 despreading unit 204, 301, 401 orthogonal code generation unit 205, 302, 402, 403 inverse orthogonal transformation unit 206 Reference signal generation unit 207 Demodulation unit 209 Synchronization holding unit 210 Synchronization control unit 303, 502, 503, 505 Integration dump unit 304 Symbol addition unit 305 Frequency error estimation unit 306 Reference signal setting unit 307 Integration control unit 404 E-ch integration dump Unit 405 L-ch integration dump unit 406 Phase error calculation unit 407 Threshold setting unit 408 Lock detection unit 409 Loop filter 410 Timing control unit 501 Amplification unit 503 Addition unit 506 Gain setting unit 601 Transmission Taserekuta 602 frame generation unit 603 and 608 modulating unit 604,609 multiplier section 605, 610 amplifier unit 606 data transmission unit 2 607 data transmission unit n 611 adding unit 612 spreading unit

[Procedure amendment]

[Submission date] January 27, 2000 (2000.1.2
7)

[Procedure amendment 1]

[Document name to be amended] Statement

[Correction target item name] Claim 1

[Correction method] Change

[Correction contents]

[Procedure amendment 2]

[Document name to be amended] Statement

[Correction target item name] Claim 3

[Correction method] Change

[Correction contents]

[Procedure amendment 3]

[Document name to be amended] Statement

[Correction target item name] 0009

[Correction method] Change

[Correction contents]

[0009]

In order to achieve the above object, a spread spectrum communication apparatus according to the present invention can identify one code sequence from a plurality of code sequences in accordance with the speed and processing gain of input data. A spread spectrum communication apparatus that generates output data by using at least one code channel to which such a code sequence is assigned, and receives the transmitted output data, based on the code sequence, And an inverse code converter for identifying one of the code sequences assigned to the known data and the code sequence assigned to the known data, and perform data demodulation on the code sequence identified by the inverse code converter. A demodulator, a frequency error estimator that calculates a frequency error amount for a transmission signal from known data in the code sequence identified by the inverse code converter, A control unit that controls a reception processing operation of the finger unit in accordance with one or more finger units having a synchronization holding unit that performs synchronization, frequency variation information detected by the frequency error estimation unit, and channel information. When the synchronization holding unit holds the synchronization of the data transmitted at the reference processing gain, the frequency error estimating unit based on the frequency error amount calculated at the reference processing gain cycle. performs synchronization holding, when holding the synchronization of data transmitted in serving as a reference processing gain is less than the processing gain, the processing gain period when the frequency error amount calculated in processing gain cycle of the data is large, as a reference Synchronization is maintained based on the frequency error amount calculated in step 2.If the frequency error amount calculated in the processing gain cycle of the data is small, the So that the synchronization holding is performed on the basis of the frequency error amount calculated in processing gain cycle of data.

[Procedure amendment 4]

[Document name to be amended] Statement

[Correction target item name] 0011

[Correction method] Change

[Correction contents]

Further, in the spread spectrum communication apparatus according to the present invention, when a reference signal for performing synchronous detection of data is generated from the frequency fluctuation information detected by the frequency error estimating unit, the generated reference signal is If the reference signal is a reference signal for performing synchronous detection of data transmitted with a reference processing gain, a reference signal is generated by integrating a code sequence to which known data is assigned with a reference processing gain period. If the reference signal produced is a reference signal for coherent detection of data sent by serving as a reference processing gain smaller processing gain odor
When the frequency error amount calculated in the processing gain cycle of the data is large, the reference signal of the data transmitted with the processing gain smaller than the reference processing gain is determined based on the integral value of the reference processing gain cycle. When the frequency error calculated by the processing gain cycle of the data is small, the reference signal may be generated by integrating the processing gain cycle of the data.

Claims (5)

[Claims] 1. According to the speed and processing gain of input data,
Spread spectrum communication apparatus for generating output data using at least one code channel to which a code sequence capable of identifying one code sequence from a plurality of code sequences and receiving the transmitted output data An inverse code conversion unit that identifies one of a code sequence assigned to data and a code sequence assigned to known data, based on the code sequence, A demodulation unit that performs data demodulation on the code sequence identified by the unit, a frequency error estimation unit that calculates a frequency error amount for a transmission signal from known data in the code sequence identified by the inverse code conversion unit, and holds synchronization. One or more finger units having a synchronization maintaining unit to perform, frequency fluctuation information detected by the frequency error estimating unit, and channel information And a control unit that controls a reception processing operation of the finger unit.In the synchronization holding unit, when holding synchronization of data transmitted with a reference processing gain, the frequency error estimation unit In the case where the synchronization is maintained based on the frequency error amount calculated in the reference processing gain cycle and the synchronization of data transmitted with a processing gain smaller than the reference processing gain is maintained, the processing gain cycle of the data and Compare the frequency error amounts calculated respectively with the reference processing gain period,
If the frequency error amount calculated in the processing gain period of the data is large, synchronization is maintained based on the frequency error amount calculated in the reference processing gain period, and the frequency error amount calculated in the processing gain period of the data is A spread spectrum communication apparatus characterized in that when it is smaller, synchronization is maintained based on a frequency error amount calculated in a processing gain cycle of the data. 2. In the case where data in a plurality of code sequences is demodulated in parallel by a plurality of finger units, the synchronization is maintained in one of the finger units to which data with the smallest processing gain is assigned. 2. The spread spectrum communication apparatus according to claim 1, wherein when the processing gain is the same in a plurality of said finger units, said one of the finger units holds said synchronization. 3. When generating a reference signal for performing synchronous detection of data from the frequency fluctuation information detected by the frequency error estimating unit, the generated reference signal is transmitted with a reference processing gain. If the reference signal is used for synchronous detection of data, a reference signal is generated by integrating a reference code sequence to which known data is assigned with a processing gain period serving as a reference. If the signal is a reference signal for performing synchronous detection of data transmitted with a processing gain smaller than the following processing gain, the integration is performed at a processing gain cycle of the data and a reference processing gain cycle, and each frequency is Calculate the amount of error and compare,
When the frequency error amount calculated in the processing gain cycle of the data is large, a reference signal of the data transmitted with a processing gain smaller than the reference processing gain is generated based on an integral value of the reference processing gain cycle. 2. The spread spectrum communication according to claim 1, wherein when the frequency error amount calculated in the processing gain cycle of the data is small, the reference signal is generated by integrating the processing gain cycle of the data. apparatus. 4. The assignment of the reference signal generation and synchronization holding of the finger unit is changed each time the arriving wave and data processing gain assigned to each finger unit are changed. By providing the signal to the synchronization holding unit, the phase error calculation unit in the synchronization holding unit controls an integration cycle for calculating a phase error signal, and a time constant of a loop filter that generates a synchronization timing signal from the phase error signal, and 2. The spread spectrum communication apparatus according to claim 1, wherein a loop gain is adjusted. 5. The control section stores a threshold value of an error signal at the time of synchronization acquisition corresponding to a processing gain, and selects a threshold value according to the frequency error amount and the processing gain from the stored threshold values. Then, it is detected whether or not synchronization is maintained by comparing the selected threshold value with the phase error signal output from the phase error calculation unit. 5. The spread spectrum communication apparatus according to claim 4, wherein the arriving wave having the highest power not allocated to the section is allocated to all of the finger sections that are out of synchronization.