System and method for modulating signal in cdma cellular telephone system

The present invention relates to a method and system for modulating signals in a communications system using spread spectrum communications signals, in particular a CDMA cellular telephone system.

Code Division Multiple Access (CDMA) code division multiple access modulation technique is one of the methods that facilitate communication in systems with a high number of users. Other multiple access techniques are known. Such as time division multiple access (TDMA) system and frequency division multiple access (FDMA) system, as well as various AM modulation systems (e.g., amplitude equalized single sideband ACSSB). However, CDMA spread spectrum modulation techniques have several advantages over these systems in multiple access communication systems. An example of the use of CDMA technology in multiple access systems is disclosed in U.S. Patent No. 4,901,307. The contents of this patent are hereby incorporated by reference.

The above patent discloses a multiple access method in which a plurality of mobile telephone system users with transceivers communicate with each other via satellite-based amplifiers or terrestrial base stations (base stations) using code division multiple access (CDMA) spread spectrum communication signals. The said ground base stations are also called cell-center stations, cell-centers, or, in short, cells. In CDMA based communications, the available frequency spectrum can be multiplied to increase system capacity. Using CDMA technology can utilize much more efficiently available spectrum than other multiple access solutions.

Satellite channels usually have a Rician fading phenomenon. Accordingly, the received signal consists of a directly received component and a plurality of reflected components following the Rayleigh fade distribution. The power ratio of the direct and reflected components - usually in the order of 6 to 10 dB, depending on the antenna of the mobile unit and its environment.

In the case of terrestrial channels, there is usually a fading phenomenon such that the received signal contains only the reflected components following the Rayleigh fading distribution, there is no directly received component. Thus, the terrestrial channels have a stronger fading phenomenon than the satellite channels where the Rician fading phenomenon is typical.

The Rayleigh-like attenuation of the terrestrial channel signal is due to the reflection of many different objects in the physical environment. As a result, the signal is received from the mobile unit in many directions and the delay values ​​are different. In the UHF bands commonly used in mobile telephone systems, there may be significant phase difference between the signals on the various paths. Signal components can extinguish each other and occasionally result in strong fading phenomena.

The degree of fading experienced in terrestrial channels is highly dependent on the current position of the mobile unit. Even a slight change in the position of the mobile unit can change the delays that occur on each signal propagation path. As a result, the phase position will be different for each route. Thus, quite a few fading phenomena can occur while moving the mobile unit. 850 In the cellular radio frequency band operating in MHz, this fading can reach a unit value per second per km / h vehicle speed. This fading phenomenon has an extremely negative effect on the terrestrial channel signal and results in poor quality communication. While increasing power can reduce the impact of fading, it also increases power consumption and interference (interference).

The CDMA modulation technique described in the aforementioned U.S. Patent No. 4,901,307 has a number of advantages over the narrowband modulation techniques used in satellite or terrestrial communications systems. The terrestrial channel poses special problems for all communication systems, primarily due to signals coming from multiple routes. The application of CDMA technology allows to overcome the specific problems of the terrestrial channel by reducing the adverse effects of multiple paths (such as fading), while allowing the benefits of multiple paths to be exploited.

CDMA cellular telephone systems use the same frequency band for communication in each cell. The property of CDMA waveforms to provide so-called signal processing gains can also be exploited to discriminate between signals in the same frequency band. Furthermore, PN modulation by high-speed pseudonoise can separate the different propagation paths, provided that the difference between the path delays exceeds the so-called PN "chip" duration, i.e., 1 / bandwidth. If approximately 1 MHz chip rate is used in the CDMA system, the gain of the whole spread spectrum signal processing, i.e. the ratio of the spread band width to the system data rate, can be exploited for all paths where the current path delay deviates by more than 1 psec. delay of. One millionth of delay delay corresponds to a distance difference of approximately 300 m. Route latency in urban environments is typically greater than 1 psec, and in some locations it can reach 10-20 psec.

In narrowband modulation systems (such as analog FM modulation used in conventional telephone systems), multiple signal paths cause strong fading effects. Broadband

HU 216 989 Β

In the case of CDMA modulation, however, different signal paths can be distinguished during demodulation. This separation greatly reduces the fading phenomenon resulting from multiple signal paths. Multiple path fading is also not completely eliminated by the use of CDMA separation techniques, as there are sometimes paths where the delay time difference is shorter than the PN chip used in the system. Signals of this magnitude are indistinguishable in the demodulator, so they will cause some degree of fading.

Therefore, it would be desirable to apply a distinction that would make it possible to reduce fading. Diversity is one possible way of reducing fading. There are three basic types of discrimination: time discrimination, frequency discrimination, and spatial discrimination.

Time discrimination is preferably achieved by repetition, interleaving, and error detection and coding (one form of repetition). All three techniques are used in the present invention for time discrimination.

The CDM A technique, because it uses a wide band signal to distribute the signal energy to a high bandwidth, already implements a form of frequency diversity (frequency diversity). Therefore, frequency selective fading affects only a small portion of the bandwidth of the CDMA signal.

Spatial or path differentiation can be achieved by connecting the mobile user to two or more cell centers simultaneously, thereby creating multiple signal paths. Routing discrimination can also be achieved by utilizing the multipath nature of the environment by spread spectrum signal processing, and by separately capturing and processing signals arriving at different latencies.

To a certain extent, the harmful effects of fading in the CDMA system can be further reduced by controlling transmit power.

The CDMA technique described in U.S. Patent No. 4,901,307 employs coherent modulation and demodulation in both directions of satellite mobile communication. Accordingly, it uses a pilot carrier that serves as a coherent phase reference for both the satellite to mobile station and cell-to-mobile communication. However, in terrestrial cellular environments, the strength of multiple signaling fading and the resulting phase distortion in the channel preclude the use of coherent demodulation for a cell-to-cell connection. The present invention eliminates the deleterious effects of multiple paths in a mobile-to-cell-site connection by employing a non-coherent modulation and demodulation technique.

The technique described in U.S. Patent No. 4,901,307 further recommends the use of relatively long PN sequences with different PN sequences assigned to each user channel. The cross-correlation of different PN sequences and the autocorrelation of one PN sequence for each time shift other than zero are zero averaged. This allows different user signals to be distinguished at reception.

However, such PN signals are not orthogonal. Although the average value of cross-correlations is zero, for a short period of time, for example, for the duration of an information bit - the cross-correlation has a binomial distribution. In this way, the signals interfere with each other as if they were broadband Gaussian noise with the same power and spectral density. Therefore, other user signals and noise due to mutual interference ultimately limit the available capacity.

A multipath environment may allow differentiation of signal paths in a broadband PN CDMA system. If there are two or more signal paths with a path delay difference greater than one ps, then these signals can be received separately using two or more PN receivers. Since these signals are generally independent from one another in terms of multiple path fading, i.e., they are usually not equally attenuated, the output signals of the two receivers can be combined. Thus, a power loss occurs only when both receivers are simultaneously weakened. Thus, one feature of the present invention is the use of two or more PN receivers in combination with a diversity signal combiner. In order to take advantage of the features of the multipath environment to eliminate fading, a signal shape is used that allows signal combining operations to be performed in signal path discrimination.

It is therefore an object of the present invention to provide PN sequences that are orthogonal and thereby allow for increased user capacity by reducing mutual interference and further promote path diversity and thereby eliminate fading phenomena.

Thus, the use of spread spectrum communication techniques, in particular CDMA techniques, in mobile cellular telephone systems provides opportunities that greatly increase system reliability and capacity. As mentioned above, CDMA technology offers additional opportunities for overcoming problems such as fading and interference. Accordingly, CDMA technology facilitates multiple utilization of the available frequency band, thereby significantly increasing the number of users simultaneously using the system.

The present invention thus provides a new and improved method, system, and means for generating PN sequences that provide orthogonality.

EN 216 989 Β between users, resulting in reduced interference, increased capacity, and improved connectivity. In fact, when orthogonal PN codes are used, the value of the cross-correlation is zero for a given time interval, so there is no interference between the orthogonal codes if the condition is that the time frames are matched in time.

In a preferred embodiment, the signals are transmitted in the form of direct signal spread spectrum communication signals between the cell-site and the mobile units. A pilot, synchronization, paging, and voice channel is established for communication from the cell-site to the mobile station. The information transmitted on the channels linking the cell-site to the mobile station is generally encoded by interleaving, BPSK, orthogonal masking of the BPSK code signals and quadrature-phase keying (QPSK).

In the case of a connection from a mobile station to a cell-site, an access and voice channel is usually established. The information transmitted from the mobile station to the cell-site channels is generally encoded by interleaving, orthogonal signaling and QPSK broadcasting.

Further features and advantages of the present invention will now be described by way of example only with reference to the accompanying drawings, in which:

First FIG. 4 is a schematic diagram of an exemplary CDMA cellular telephone system; FIG. the

Second Fig. 4A is a block diagram of a cell-center equipment of a CDMA cellular system; the

Third block diagram of a cell-center receiver; the

4a., 4b., 4c. Figures 4A to 5B show a block diagram of a cell-center transmitter modulator; the

5th FIG. 4A is a time diagram illustrating synchronization of a synchronization channel code signal; FIG. the

6th FIG. 4A is a time diagram illustrating timing of a synchronization channel for orthogonal masking; the

7th FIG. 4 is an exemplary timing diagram of a full cell-site mobile station connection; the

8th block diagram of apparatus for switching a mobile telephone exchange; the

9th Fig. 4 is a block diagram of a mobile telephone unit constructed in accordance with a CDMA cellular telephone system; the

10th block diagram of a receiving unit of a mobile unit; the

11th Fig. 4 is a block diagram of a mobile unit transmitter modulator; the

12th FIG. 4A is an exemplary timing diagram of a mobile unit-to-cell connection for variable data rate and burst transmission; and the

13th FIG. 4A is an exemplary timing diagram of a complete mobile station-to-cell connection.

In a CDMA cellular telephone system, there are multiple modulator / demodulator units (spread spectrum modem) in each cellular exchange. Modems consist of a digital spread spectrum transmitter, at least one digital spread spectrum receiver, and a paging receiver. In the cell-site, each modem is assigned to a single mobile unit to facilitate communication with designated mobile units.

In a CDMA cellular telephone system, a "smooth handover technique" is employed in which the cellular modem used previously continues to serve the mobile unit when the receiving new cell-site modem is assigned to that mobile unit. When the mobile unit is in the transition range between the two cell centers, the call can be switched back and forth between the two cell centers depending on how strong the received signal is. Because the mobile unit is always connected to at least one cell-site modem, there is less interference with the mobile unit or service. The mobile unit thus facilitates crossing the cell boundary (handover) by utilizing multiple receivers. It also uses the diversity function to reduce the effects of fading.

In a CDMA cellular telephone system, each cellular exchange transmits a "pilot carrier signal". If the cell is subdivided into sectors, each sector within a cell has its own pilot signal. The pilot signal is used by the mobile unit for initial synchronization of the system and for stable phase, frequency, and time tracking of signals from the cell-site. In addition, each cellular center also provides different spread spectrum modulated information. These include, for example, a cell-site identifier, a system timer, a mobile unit paging signal, and other control signals.

The pilot signals transmitted by each sector of each cell have the same spreading code, but the phase shift of the codes is different. Phase shifting allows the pilot signals to be distinguished from each other, thereby distinguishing the cell or sector that is delivering the signals. Applying the pilot code allows the mobile unit to perform system timing synchronization with a single search for each pilot code phase. By performing a correlation process for each code phase, the strongest pilot signal can be easily selected. The strongest pilot signal thus determined generally corresponds to the pilot signal issued from the nearest cell center. The system always uses the strongest pilot signal, whether it is issued by the nearest cell center or not.

After finding the strongest pilot signal, that is, when the mobile unit is synchronized with the strongest pilot signal, the mobile unit switches to another carrier. This carrier must be received by all users within the cell. This

A carrier, called a synchronization channel, transmits a generic message that contains system information to be used by mobile stations in the system. The system information identifies the cell center and the system, and also contains information that enables long PN codes used by the mobile unit to be synchronized without further search, insertion! (interleaver) timeframes, vocoders, and other system timing information. Another channel is used, called a paging channel. Its purpose is to forward messages to mobile stations indicating that they have received a call and to select a channel when a mobile station makes a call.

The mobile station continues to search for the received pilot signal for code deviations corresponding to sectors adjacent to the cell-site, i.e., pilot signals issued by neighboring sectors. This search serves to determine whether the pilot signal issued by the neighboring cell or sector is stronger than the pilot signal initially found by the mobile unit to be the strongest. If, in this call deactivated mode, the pilot signal of a neighboring sector or cell center is stronger than the pilot signal emitted by the originally determined sector or cell center, then the mobile unit uses the stronger pilot signal and the synchronization and paging of the new sector or cell center is accordingly used. channel.

When making a call, a pseudorandom noise (PN) code address is assigned to be used during the call. The code address is either assigned by the cell center to the mobile unit or predefined by the mobile unit ID. After the call is initiated, the mobile unit continuously monitors the pilot signal of the cell center through which communication is established, but also monitors the pilot signals of adjacent sectors or cells. The purpose of continuously monitoring the pilot signals is to determine whether a neighboring cell or sector is receiving a stronger pilot signal than that emitted by the cell center with which the mobile unit is connected. If the pilot signal for the adjacent cell or sector is stronger than the pilot signal for the cell or sector in use, this indicates that the mobile unit has moved to a new cell or sector and a handover / takeover has to be initiated.

Figure 1 is a schematic diagram of an embodiment of a telephone system according to the present invention. The system illustrated in Figure 1 implements a spread spectrum modulation technique for communication between the mobile units (mobile phones) of the system and the cellular exchanges. In large cities, cellular systems can have hundreds of cell centers that can serve hundreds of thousands of cellular telephones. The use of dispersive spectral techniques, particularly CDMA techniques, provides significantly increased user capacity for systems of this size compared to conventional FM modulation cellular systems.

The system of Figure 1 has a system control and switching unit 10, also known as a switching center (MTSO) for mobile telephones. This usually includes interface circuits and signal processing circuits that provide system control to the cell center. The system control and switching unit 10 further controls the transfer of telephone calls from the public switched telephone network (PSTN) to the appropriate cell center for transmission to the called mobile unit. The system control and switching unit 10 further controls the transfer of calls from the mobile units to the PSTN via at least one cell-site. The system controller and switching unit 10 may also connect the mobile units to each other via the appropriate cell center, since the mobile units generally cannot communicate directly with each other.

The system control and switching unit 10 may connect to cellular centers by various means, such as dedicated telephone lines, fiber optic or microwave connections. Fig. 1 shows, by way of example, two cellular centers 12, 14 and two mobile units 16, 18 equipped with one cellular telephone. For the purposes of the present disclosure, cell centers 12, 14 are intended to supply whole cells, but, where appropriate, cells may be geographically divided into sectors, each sector being considered as a separate supply area. Accordingly, within a cell, the transfer / receipt between sectors is the same as described for cells, however, it is possible to differentiate between sectors as well as between cells.

In Figure 1, arrows 20a, 20b, and 22a, 22b, respectively, indicate possible communication paths between cell center 12 and mobile units 16, 18. Similarly, arrows 24a, 24b, and 26a, 26b respectively indicate possible communications links between the cellular center 14 and the mobile units 16, 18. The cell centers 12, 14 have the same rated transmit power.

The cell centers' supply areas (i.e., the cells) are geographically shaped so that the mobile unit is generally always only close to one cell center. Each cell can be divided into sectors. When the mobile unit is in standby mode, i.e., no call is in progress, the mobile unit continuously monitors the pilot signals emitted by nearby cell centers. If the cell is subdivided into sectors, then only one cell center pilot signal may be monitored by the mobile unit. As shown in Figure 1, the pilot signals are transmitted by cell centers 12, 14 to the mobile unit 16 via the outbound (forward) communication links 20a, 26a. The mobile unit 16 can determine which cell it is in by comparing the received signal strength of the pilot signals output from the cell centers 12, 14.

In the example illustrated in Figure 1, the bottom mobile unit is closer to the cell center 12. When the mobile unit 16 makes a call, a control message is sent to

EN 216 989 Β to the nearest cell center, that is, to the 12 cell centers. Upon receipt of the call request message, cell center 12 transmits the called number to the system control and switching unit 10. The system controller and switching unit 10 then connects the calling party to the called station via the PSTN (Public Switched Telephone Network).

In the case of a call made within the PSTN, the system controller and switching unit 10 forwards the call information to all cell centers in the area. The cell centers then issue a paging message to the called mobile user within their service area. The called mobile user responds to the paging message by sending a control message to the nearest cell center. This control message indicates to the system control unit 10 that this particular cell center is in communication with the mobile unit. The system control and switching unit 10 then directs the call to this mobile unit via this cellular exchange. If the mobile unit 16 exits the service area of ​​the initially used cell center, i.e. the cell center 12, it will attempt to continue the conversation through another cell center.

For cellular telephone systems, the Federal Communications Commission (FCC) has designated a total of 25 MHz for cell-to-cell communications and 25 MHz for cell-to-cell communications. The FCC divided the designation equally between two providers, one selected by the fixed telecommunications company in the service area and the other selected by lot. In accordance with the designation order, the 12.5 MHz assigned to each carrier in each direction of the connection is divided into two subbands. For wired carriers, the subbands are 10 MHz and 2.5 MHz wide. For non-wired carriers, the sub-bands are 11 MHz and 1.5 MHz wide. Thus, a signal with a bandwidth of less than 1.5 MHz will fit in any subband, while a signal with a bandwidth of less than 2.5 MHz will fit in any subband.

In order to achieve maximum flexibility in the use of the available frequency spectrum when using the CDMA technique, the bandwidth of the waveform used in the cellular telephone system must be less than 1.5 MHz. Alternatively, 2.5 MHz bandwidth would be a good option, providing full flexibility for wired and nearly total flexibility for non-wired carriers. Although the use of a higher bandwidth has the advantage of better multipath discrimination, it has the disadvantage of increasing the cost of the equipment required and of limiting the frequency assignment within a given bandwidth.

For the spread spectrum cellular telephone system of Figure 1, it is desirable to use a direct signal sequence comprising a pseudorandom (PN) spread spectrum carrier. The "chip" frequency of the PN signal sequence in the preferred embodiment

1.2288 MHz. We choose this chip frequency because the resulting bandwidth, which after filtering is about 1.25 MHz, is about one tenth of the total bandwidth allocated to the carrier of the single cell service.

Another consideration when selecting a given chip frequency is that the chip rate should be as many times as many times the baseband data rate that you want to use in the system. It is also desirable that the multiplication factor be a power of two. In a preferred embodiment, the baseband data rate is 9600 bps. Accordingly, the PN chip selected is 128x9600 or 1.2288 MHz.

In the case of a cell-to-cell connection, the binary sequences used for spreading the spectrum are made up of two different types of signals having different characteristics and functions. One is an external code that is common to all signals within a given cell or sector and is intended to discriminate between signals on multiple paths. This external code is also used to distinguish signals transmitted by different cells or sectors to the mobile unit. The other is an internal code intended to discriminate between user signals transmitted by a single sector or cell.

In a preferred embodiment of the invention, the signals transmitted from the cell-center use a four-phase sine carrier modulated by a single sector or cell, which generates an external code and is binary PN. These PN sequences are produced by two different PN generators with the same sequence length. One signal sequence modulates the in-phase channel (channel I) of the carrier phase and the other sequence modulates the carrier channel of the carrier by 90 ° (Q channel) biphasically. The resulting signals are summed to form a complex four-phase carrier.

Although logical "zero" and logical "one" values ​​are generally used to represent binary sequences, the + V former signal voltage represents the logical "one" and the -V former signal voltage represents the logical "zero". In biphasic modulation of a sinusoidal signal, the zero-volt average voltage sinusoidal signal is multiplied by a voltage level of + V or -V according to the control given by the binary sequence, using a multiplication circuit. The bandwidth of the resultant signal can then be limited by a bandpass filter. It is also known in the art to reverse the order of operations by passing a binary sequence through a low pass filter before multiplying it by a sinusoidal signal. The four-phase modulator consists of two two-phase modulators, which are driven by different signal sequences, and there is a 90 ° phase shift between the sine signals used in the two-phase modulators.

In the preferred embodiment, the carrier of the transmitted signal corresponds to 32,768 chips. Sequences of this length are set with a modified maximum-length linear sequence generator6

EN 216 989 Β by adding a zero bit to a sequence of 32 767 chips. The resulting signal sequence is characterized by good cross-correlation and autocorrelation properties. Good cross-correlation and autocorrelation properties are needed to avoid interference between pilot carriers issued by different cells.

It is advisable to use such a short sequence of signals in order to minimize the search time of mobile units when they first enter the system, without knowing its timing. If the timing is unknown, the signal sequence must be scanned the entire length to find the correct timing. The longer the sequence, the longer the search time. Although it is possible to use sequences shorter than 32,768, it should be noted that reducing the length of the sequence also reduces the gain in code processing. If, on the other hand, processing gains are reduced, the effects of multiple path interference and interference from neighboring cells and other signal sources will be amplified, where appropriate, to an unacceptable extent. Therefore, it is advisable to use the longest sequence of characters that can be performed within a reasonable time. It is also recommended that the same code polynomial be used in each cell so that the mobile unit, which at the beginning of the synchronization does not yet know which cell it is in, can achieve full synchronization by searching for a single code polynomial.

To simplify the synchronization process, all cells in the system are synchronized to each other. In the exemplary embodiment, cell synchronization is performed by synchronizing all cells to a common time base, the satellite navigation system Navstar Global Positioning System. This system, in turn, is synchronized to UTC (Universal Coordinated Time).

Signals from different cells are distinguished by shifting the base sequences relative to one another over time. The time offset of the base sequence is different for each cell and, of course, different from its neighboring cells. In the preferred embodiment, the repetition period 32,768 is divided into 512 time offsets which are 64 "chips apart". In a cellular system, each sector of cells also has different time offsets that they must apply to each transmission. If there are more than 512 sectors or cells in the system, time offsets can be reused in the same way as frequencies are reused in the current analog FM cell system. For other embodiments, a number other than 512 may be used. If the time offsets of the pilot signals are properly selected, the time offsets applied by the neighboring cells should not be the same.

All signals emitted by the same cell or sector apply the same external PN code to I and Q channels. The signals are then distributed (distributed, spread out within the available band) by an orthogonal internal code generated by using Walsh functions up10. The signal addressed to a particular user is thus multiplied by both the external PN sequences and a single Walsh sequence or a series of Walsh sequences. The latter are designated by the system control unit for the duration of the user telephone conversation. We use the same internal code for both I and Q channels. Thus, a modulation is implemented which is actually biphasic for the internal code.

It is well known in the art that a set of n orthogonal binary series of length n (one power of two) can be formed (see SW GOLOMB et al., Digital Communications with Space Applications, Prentice-Hall, Inc., 1964. 45-64 p.). There are known orthogonal binary sequences whose length is a multiple of four but less than two hundred. An easily obtainable group of such sequences are the so-called Walsh functions, also known as Hadamard matrices.

An n th order Walsh function can be defined recursively as follows:

W (n) = IW (n / 2), W (n / 2) IIW (n / 2), W '(n / 2) I, where W' is the logical complement of W, and W (1) = | 0 |. so

W (2) =

0, 01 0, II and W (4) =

0,

0,

0,

0,

0,

1, o,

1,

0,

0,

1,

1,

And W (8) reads as follows:

W (8) = 0, 0, 0, 0, 0, 0, 0, 0

0, 1, 0, 1, 0, 1, 0, 1 o, 0, 1, 1, 0, 0, 1, 1

0, 1, 1, 0, 0, 1, 1, 0

0, 0, 0, 0, 1, 1, 1, 1

0, 1, 0, 1, 1, 0, 1, 0

0, 0, 1, 1, 1, 1, 0, 0

0, 1, 1, 0, 1, 0, 0, 1

A Walsh sequence is a line of the Walsh function matrix. The nth order Walsh function contains n sequences, each consisting of n bits.

The n-th order Walsh function (and other orthogonal functions) is characterized by the fact that, for a time interval corresponding to n code signals, the cross-correlation value in the set between all different sequences is zero if the sequences are aligned (synchronized) in time. This can be seen by observing that each sequence differs from each additional sequence by exactly half its bits. Note also that there is always a signal sequence that contains only zeros, and all other signal sequences contain one half and one half zeros.

Adjacent cells and sectors can reuse Walsh sequences because the

GB 216 989 külső external PN codes used in cells and sectors vary. Because the propagation time of a signal is different between a particular mobile station and two or more different cells, it is not possible to satisfy the condition of temporal matching necessary for the orthogonality of the Walsh function to two cells at the same time. Therefore, the external PN code should be left to distinguish the signals from different cells to the mobile unit. However, the signals emitted by the same cell are orthogonal to each other, so they do not interfere with each other. This eliminates most of the disruptions in most locations, and allows for capacity increase.

The system further utilizes a variable data rate audio channel in which the data rate can be varied per block of data and the current data rate can be controlled at a minimum operating cost. The use of variable data rates reduces mutual interference because no unnecessary transmissions occur when there is no conversation. In vocoder units, varying numbers of bits are generated for each vocoder block using specific algorithms, following changes in speech activity. For example, during an active conversation, the vocoder generates blocks of data of 20 ms, which contain 20, 40, 80, or 160 bits, depending on the speech activity. It is advisable to transmit the blocks of data at fixed length intervals while changing the transmission rate. It is also advantageous if no signal bits are needed to inform the receiver of the number of bits transmitted.

The blocks are also encoded by a cyclic redundancy check code (CRCC). This adds additional parity bits to the block, which can be used to verify that the data blocks have been encoded correctly. The CRCC verification codes are generated by dividing the data block by a predetermined binary polynomial. The CRCC consists of some or all of the remaining bits of the splitting process. The receiver checks the CRCC by reproducing this residue and examines whether the received bits are the same as the regenerated control bits.

In the present invention, the receiver decoder first decodes the block as if it were 160 bits, then as if it consisted of 80 bits, etc., until it had tried all possible block lengths. It calculates the CRCC value for each decoding attempt. If any attempt at decoding results in a correct CRCC, it accepts the decoded block of data and passes it on to the vocoder for further processing. If no decryption attempts result in a valid CRCC, it transmits the received code signals to the signal processing unit of the system where other signal processing operations may be performed.

The performance of the signal transmitted by the cell-site transmitter varies depending on the value of the data rate of the current data block. The highest carrier power belongs to the highest data rate. If the data rate is less than the maximum value, the modulator, on the one hand, reduces performance and, on the other hand, repeats each encoded data code signal as many times as necessary to achieve the desired transmission rate. For example, at the highest transmission rate, each encoded data code signal is repeated four times.

In the mobile unit transmitter, peak power is kept constant, but the transmitter is turned off for a period corresponding to 1 / 2.1 / 4 or 1/8 of the available time interval, according to the number of bits transmitted within the data block. The timing (timing) of the transmitter activation periods is changed according to a pseudo-legal rule according to the user code assigned to the mobile user.

In the preferred embodiment, the Walsh function n is selected to be sixty-four (64) for cell-to-mobile communication. Thus, each of the maximum sixty-four different signals to be transmitted is assigned a unique orthogonal sequence. The forward error correction coding (FEC) code sequence for each speech signal is multiplied by the selected Walsh sequence. The code sequence subjected to Walsh encoding and FEC encoding is then multiplied by the external PN encoded waveforms for each audio channel. The resulting distributed (scattered) code sequences are added together to form a composite waveform.

The resulting composite waveform is modulated on a sinusoidal carrier and then subjected to bandpass filtering. It is then converted to the desired operating frequency, amplified and finally radiated using the antenna system. In other embodiments of the present invention, the sequence of some of the operations described herein may be interchanged in generating a signal emitted by the cell-site. For example, it may be advantageous to multiply each audio channel prior to multiplying the external PN coded waveform and filtering operation to sum up all channel signals to be transmitted by the antenna. It is well known in the art that the order of linear operations can be interchanged in order to achieve certain advantages in implementation and to provide various design modes.

In an embodiment considered useful for use in cellular systems, pilot-to-cell technique is used to connect from the cell to the mobile unit (see U.S. Patent No. 4,901,307). Each cell emits a pilot carrier using the same sequence of 32,768 lengths, but with different time offsets to avoid mutual interference.

To create the pilot waveforms, use only the Walsh sequence of zeros. Such a sequence is contained in every Walsh function set. The fact that each cell uses the Walsh sequence of zeros only to generate the pilot waveform allows you to ignore the Walsh functions in the initial search for the pilot waveform until the external PN code synchronization is complete. Walsh time frames are locked to the PN code cycle because the PN sequence is an integer multiple of the Walsh time frame

EN 216 989 Β. Therefore, if the offsets of the PN code assigned to each cell are an integer multiple of the length of the Walsh time frame (corresponding to sixty-four "chips"), then the Walsh time frame is implicitly derived from the timing cycle of the external PN code.

All cells in a given supply area are precisely synchronized. In the preferred embodiment, the timing of the local signal in each cell is synchronized by the GPS receiver to UTC (Uniform Coordinated Time). The GPS system has an accuracy of at least 1 ps. Accurate cell synchronization is recommended so that the call can be smoothly transferred / received when the mobile station switches from one cell to another during an ongoing call. If the neighboring cells are synchronized, the mobile unit can easily synchronize itself to the new cell, thus ensuring smooth reception.

The pilot carrier is generally transmitted at a higher power level than the speaker carrier. Thus, the pilot carrier has a better signal-to-noise ratio and less susceptibility to interference. The higher performance carrier allows the initial search (synchronization) to be performed more quickly and the determination of the phase position of the pilot carrier can be performed with a very high accuracy, also with a relatively high bandwidth phase tracking circuit. The pilot phase value obtained from pilot carrier tracking is used as carrier phase reference for demodulating carriers modulated by a user information signal. This technique makes it possible to use the same common pilot signal as a carrier phase reference for the carrier of a series of users. For example, in a system that emits a total of fifteen loudspeakers simultaneously, the pilot carrier power may be four times that of the speaker carrier.

In addition to the pilot carrier, the cellular center transmits another carrier to all system users in the cell area. This carrier, also referred to as a synchronization channel, also uses the 32,768 PN sequence for spreading the spectrum, but with another predefined Walsh sequence. The message transmitted by the synchronization link contains system information for mobile units in the system. The system information identifies the cell center and the system, and carries information that allows the long PN codes used for mobile unit information signals to be further synchronized without search.

An additional channel, the so-called paging channel, can be used to forward various messages to the mobile unit. The paging channel, on the one hand, alerts the mobile units that they are receiving a call and, on the other hand, performs a channel assignment when the mobile unit makes a call.

Speakers transmit conversations in digital form. The analog speech signal is digitized using standard digital telephony techniques and then compressed using a vocoder method at a data rate of approximately 9600 bps. The resulting data signal is then subjected to convolutional coding at r = 1/2 data rate and K = 9 constraint length, using repeat and interleaving to provide error detection and error correction that allows the system to be much smaller work well under signal to noise and interference conditions. Various methods of convolutional coding, repetition, and interleaving are well known in the art.

The encoded signals obtained are multiplied by a designated Walsh sequence and then the external PN code. This process produces a 1.2288 MHz PN signal, or 128x9600 bit / s data rate. The resulting signal is then modulated to an RF carrier and summed together with the pilot and setup carrier, together with the other audio carriers. The summation can be performed at several stages of signal processing, such as the IF frequency or the baseband frequency, both before and after multiplication by the PN sequence.

In addition, each speaker is multiplied by a value that adjusts the performance of that speaker to that of other speakers. This power control can also be used to adjust performance for connections that require increased performance due to the recipient's unfavorable position. Receivers are equipped with appropriate means to provide feedback on the signal-to-noise ratio, so power can be adjusted to ensure proper operation without wasting energy. The orthogonality of Walsh functions is not disturbed by the fact that different power levels are used for each speaker provided that time matching is ensured.

Figure 2 is a block diagram of a possible embodiment of a cell-center device. There are two receiving systems in the cell center. Both have their own antenna and analog receiver for realizing space diversity reception. The signals are processed in the same way in both receiver systems until the start of the diversity combination process. The dashed units represent the elements that provide communication between the cellular center and the mobile unit. The outputs of the analog receivers may be connected to additional elements. These are used when communicating with other mobile units.

The first receiver system of the apparatus of Figure 2 consists of 30 antennas, 32 analog receivers, 34 paging receivers and digital data receivers 36. The first receiver system may optionally include an additional digital data receiver 38. The second receiver system comprises 40 antennas, 42 analog receivers, 44 pagers, and digital data receivers 46.

The cell center has a cell-center control unit 48 (processor). The control unit 48 is connected to the data receivers 36,38,46 and the search receivers 34,44. The control unit 48 performs, among other things, signal processing, timing signal generation, power control, handover / receipt control, diversity, discrimination combining, coupling of mobile telephones to a switching center (MTSO) (see Figure 8). as well). Controller 48

EN 216 989 Β also assigns Walsh sequences and transmitters and receivers.

The two receiving systems are connected via diversity receivers 36, 38, 46 to a diversity combining and decoding circuit 50. The output of diversity combining and decoding circuit 50 is connected to 52 digital links. The digital link 52 is further connected to the control unit 48, the cell-center transmitter 54 and the digital switch of the MTSO. The digital connection 52 transmits signals controlled by the control unit 48 between the MTSO and the cell-center transmitter modulator 54 and the MTSO and diversity combining and decoding circuits (see also FIG. 8).

The signals emitted by the mobile unit are direct-spread, spread-spectrum signals controlled by a predetermined clock frequency, such as 1.2288 MHz. The clock frequency is suitably selected to be an integer multiple of the baseband data rate of 9600 bps in this case.

The signals received by the antenna 30 are transmitted to the analog receiver 32. The analog receiver 32 is shown in detail in FIG. The signals received by the antenna 30 are transmitted to a conversion unit 100 consisting of an RF amplifier 102 and a mixer 104. The received signals are fed to the input of RF amplifier 102. The signals amplified by the RF amplifier 102 are applied to one of the inputs of mixer 104. The other input of mixer 104 is connected to the output of frequency synthesizer 106. Mixer 104 converts the amplified RF signals to mid-frequency IF signals by mixing with the output signal of frequency synthesizer 106.

The IF signals from the mixer 104 output to the bandpass filter 108 (BPF), which is generally a surface acoustic wave filter (SAW). It has a bandwidth of 1.25 MHz. The filtered signals are transmitted from the output of the bandpass filter 108 to the IF amplifier 110 which amplifies them. The amplified IF signals are output from the output of the IF amplifier 110 to a 112 A / D (analog / digital) converter which digitizes the signals at a clock frequency of 9.8304 MHz. This clock frequency is exactly eight times the PN "chip" frequency. Although the A / D converter 112 is depicted as part of the analog receiver 32, this signal converter may be part of the data receivers or the paging receivers. The digitized IF signals are output from the output of the A / D converter 112 to the data receiver 36 and the search receiver 34 (and, if applicable, the data receiver 38). The output signals of the analog receiver 32 form I and Q channel signals, which are described below. Although the 3. In Fig. 1A, the A / D converter 112 is shown as a single unit, where the I and Q channel signals are separated later, it is possible that the I and Q channel signals are separated before the digitization, and two A separate A / D converter digitizes the I and Q channel signals. RF-IF baseband frequency conversion methods and techniques for analog / digital conversion of I- and Q-channel signals are well known in the art.

The cell-site paging receiver 34 examines the time value characteristics of the received signal to ensure that the associated digital data receiver 36 (and the digital data receiver 38, if any) is tuned to and processing the signal corresponding to the strongest available time value. The paging receiver 34 provides a signal to the cell-center control unit 48, which controls the data receivers 36, 38 in the sense that it selects from the received signals suitable for further processing.

The signal processing performed in the cell-site data receivers and the paging receiver differs in many respects from the signal processing performed in similar elements of the mobile unit. In the case of an inbound connection (from the mobile unit to the cell center), the mobile unit does not provide a pilot signal that could serve as a coherent reference for cell-center signal processing. The connection from the mobile unit to the cell center is characterized by non-coherent modulation and demodulation using 64-bit orthogonal signaling.

In the 64th order orthogonal signaling technique, the code signals transmitted by the mobile unit are 2⁶ , which is one of sixty-four different binary sequences. The set of chips selected for this purpose is known as Walsh functions. For m-th signal encoding using the Walsh functions, the optimal reception function is fast Haddard Transform (FHT).

Returning to Figure 2, the output signals of the analog receiver 32 are received by the search receiver 34 and the data receivers 36,38. In order to decode the spread spectrum signals that are transmitted to the cell-site receiver through which the mobile unit communicates, it is necessary to generate the appropriate PN sequence. Further details regarding the generation of mobile unit signals will be discussed below.

As shown in Figure 3, the receiver 36 comprises two PN generators 120, 122 which produce two PN sequences encoded according to different short PN codes but of the same length. These two PN sequences are common to all cell-site receivers and all mobile units in terms of external modulation, as will be described in more detail below. The PN generators 120, 122 are thus PN_T and PNq output sequences. A PN_t and PNq output sequences are referred to as the PN-j sequence for the in-phase channel (I) or the 90-offset phase (Q) channel.

The PN1 and PNq sequences are generated with two 15-degree polynomials, each of which is incremented to produce 32,768 lengths instead of the normally generated 32767 length sequences. For example, elongation (expansion) can be done by adding a zero to a fourteen zero sequence that occurs once in every maximum length of a 15 degree linear sequence. In other words, a PN sequence generates one state of a PN generator once. Thus, the modified (expanded) sequence comprises a fifteen single sequence and a fifteen zero sequence.

In the exemplary embodiment, the data receiver 36 comprises a PN generator 124 corresponding to a long PN encoding, which generates a PNu sequence. This is the cell space between the mobile unit and the cell unit10

EN 216 989 Β points corresponds to a PN sequence produced. The PN generator 124 may be a maximum length linear signal generator that generates a very long (e.g., degree 42) user PN code with a time offset corresponding to an additional factor. This additional factor may be, for example, a mobile unit address or a user ID (ID) in order to distinguish between individual users. Thus, the signal received at the cell center is both a long-code PNU and a short-code PNU_r and modulated by a PNQ sequence. In an alternative embodiment, a non-linear encryption generator (e.g., a special user key encryption encryption method for displaying universal time with sixty-four code symbols using the DES data encryption standard) may be used instead of the PN generator 124.

The PNy sequences issued by the PN generator 124 and the PN1 and PNQ sequences are subjected to exclusive ORs 126, 128, respectively, thereby generating the PN1 and / or PNQ 'sequences.

The PN ['- and PNQ' sequences, respectively, along with the I and Q channels from the 32 output of the receiver, are transmitted to 130 PN QPSK correlators. Correlator 130 correlates data for I and Q channels with PN / or PNQ'. The correlated I and Q channel signals output from the correlator 130 are provided to a battery 132, 134 which collect code data for four "chip" periods. The output signals of the accumulators 132, 134 are fed to one of the inputs of the FHT processor 136, which performs rapid Hadith transformation. The FHT 136 generates a set of sixty-four coefficients for each of the six code marks. This sixty-four coefficient is multiplied by the weighting function generated by the control unit 48. The weighting function is related to the strength of the demodulated signal. The weighted data from the output of the FHT processor 136 is fed to the diversity combining and decoding circuit 50 (FIG. 2) for further processing.

The second receiver system is shown in FIGS. 7B, process the received signals in a manner similar to that described with respect to the first receiver system of FIGS. Sixty-four weighted code signals from the data receivers 36, 46 are diversity diversity combining and decoding inputs 50, which includes a summing. This summator adds the sixty-four weighted coefficients from the 36 data receivers to the sixty-four weighted coefficients from the 46 data receivers. The resulting sixty-four coefficients are compared with each other to determine the largest coefficient. Using the value corresponding to the result of the comparison and the identifier of the highest of the sixty-four coefficients, a set of decoding weights and codes is generated for use by the Viterbi algorithmic decoder of the diversity combining and decoding circuit 50.

The Viterbi algorithmic decoder of diversity combining and decoding circuit 50 is configured to decode data encoded by the mobile unit at a K = 9 constraint length and r = 1/3 data rate. The Viterbi decoder determines the most likely information bit sequence. The system performs a signal quality assessment according to a defined cycle, with 1.25 ms intervals where appropriate. The result is transmitted to the mobile unit as a power control instruction of the mobile unit along with the data. Signal quality is the average signal to noise ratio over a time interval of 1.25 ms.

Data receivers monitor the timing of the signals they receive. Monitoring is done by correlating the received signal with a slightly earlier local reference PN or a slightly later local reference PN. This solution is known per se. If there is no timing error, the average of the difference between the two correlations is zero. However, if the timing is incorrect, this difference will show the magnitude and sign of the error, and the receiver timing will be adjusted accordingly.

Returning to Figure 2, the cell center has an antenna 62 connected to a GPS receiver 64. The GPS receiver 64 processes signals received by the antenna 62 from the Navstar Global Positioning System satellite navigation system and produces UTC time signals. The GPS receiver 64 transmits these time signals to the control unit 48 for time synchronization of the cell center, as described above.

The apparatus of Figure 2 optionally includes 38 data receivers to improve system performance. The structure and operation of this data receiver 38 is similar to the data receivers 36, 46. With the help of the data receiver 38, additional diversity modes can be implemented in the cell center. The data receiver 38, either alone or with additional data receivers, can follow the signals emitted by the mobile unit and receive them through another path of delay value. The data receiver 38, as well as any additional digital data receivers used, may provide additional discrimination modes. This is extremely useful for cells located in densely built-up urban areas where there is a high probability of multiple signal paths.

The signals from the MTSO are transmitted to the corresponding transmit modulator via digital link 52 and controlled by the control unit 48. The transmitter modulator 54 controlled by the control unit 48 submits the data to spread spectrum modulation for transmission to the designated (addressed) mobile unit. Further details on the construction and operation of transmitter modulator 54 will be described below with reference to FIG.

The output signal of the transmitter modulator 54 is fed to a transmitter power control circuit 56 which is controlled by the control unit 48 to control the transmitter power. The output signal of the transmit power control circuit 56 is provided to one of the inputs of the summing circuit 57. The summing circuit 57 adds this output signal to the output signals of the transmitter modulator / transmitter power control circuits to the additional mobile units in the cell. The 57 totalizer11

HU 216 989 21 output signal to 58 transmit power amplifier circuits and then to 60 antennas. The antenna 60 radiates the received signals to mobile units in the service area of ​​the cell. The apparatus of Figure 2 further comprises a unit 66 comprising pilot / control link generators and a transmit power control circuit. The unit 66, controlled by the control unit 48, generates a pilot signal, a synchronization channel and a paging channel, and controls its power and transmits them to the transmit power amplifier circuit 58 and to the antenna 60.

4a-4c. Figures 6 to 9 are block diagrams of a possible embodiment of a cell-site transmitter. The transmitter contains two PN sequence generators. These produce the external code. PN generators produce two different PN sequences, PN_r and the PNq sequence (already described in Figure 3). The transmitter circuit comprises two PN generators 196, 198 which are PN_r, and generate a PNq sequence. The PN generators 196, 198 delay the PN sequences in a predetermined amount according to the sector or cell address signal from the control unit 48. These are time delayed PNs_rand PNq-j series correspond to the phase (I) channel and the 90 ° offset (Q) channel, respectively. Although, in the figure, PN is the appropriate channel for the cell center or sector_rand PNQ sequences, only two PN generators are depicted, of course, other PN generator arrangements are also possible. For example, for a cell that is not sectored, each of the pilot, synchronization, paging, and voice channels may have one of the synchronized PNs used in the external code._r, or a pair of PN generators that produce PNq sequences. This solution is advantageous because PN_rand PNq sequences do not need to be distributed to a large number of circuits.

In a preferred embodiment, Walsh coding of channel signals is used as internal code. Using the exemplary numeric values, we have a total of sixty-four different Walsh sequences, three of which are assigned to the pilot, synchronization, and paging channels. In the case of synchronization, paging and voice channels, the input data are subjected to convolutional coding and then interleaving in a manner known per se. In addition, convolutionally encoded data is repeated before insertion, in a manner known per se.

The pilot channel does not contain data modulation and represents a non-modulated spread spectrum signal that is used by all users of a particular cell or sector for search or tracking purposes. Each cell (if the cell is divided into sectors, each sector) has a unique pilot signal. However, it has been discovered that it is not advisable to use unique PN generators to generate unique pilot signals. Instead, it is much easier to apply the unique pilot signals from the same set of base signals by unique time offsets. In this solution, the mobile unit examines the entire sequence of signals sequentially and tunes it to the offset (set) that gives the strongest correlation. The unique time offsets of the base sequence must be such that pilot signals from adjacent cells or sectors do not interfere or quench each other.

Therefore, the pilot sequence must be long enough to produce a large number of different signal sequences using offsets in the base sequence, since it is advisable to use as many unique pilot signals as possible in the system. Furthermore, the amount of offset (discrimination) must be large enough so that the pilot signals do not interfere with each other. Accordingly, in the exemplary embodiment of the present invention, the length of the pilot sequence is 2¹⁵. The generation of the sequences is described in Figure 2¹⁵It starts with a sequence of -1 signals, to which a specific zero is added if a specific condition is detected. In the exemplary embodiment, 512 different pilot signals can be generated by occasionally using offsets corresponding to 64 "chips" in the base sequence. The offset may be an integer multiple of 64 chips, but the number of different pilot signals will be reduced.

To produce a pilot signal, a "zero" (W_She) We use the Walsh sequence because it does not modulate the pilot signal, which essentially corresponds to PN_r and PNq sequences. The "zero" (W_She) The Walsh sequence is thus multiplied by the exclusive OR gates PN_r and PNq sequences. The resulting pilot signal contains only the PNj and PNq sequences. Because each cell center or sector uses the same PN sequence as the pilot edge, the phase position of the sequence will be the distinguishing feature when identifying the transmission location (cell center or sector).

In the pilot channel portion of the transmit modulator and power control circuit 66, the Walsh generator (W_She). This produces a signal that corresponds to the aforementioned zero-only function. In the generation of the Walsh function, the timing is determined by the control processor, as is the case with all Walsh function generators for cell centers and mobile units. The output of the Walsh generator 200 is connected to one of the inputs of two exclusive gates 202, 204. The other input of the exclusive OR gate 202 is PN_rand the other input of the exclusive OR gate 204 is the PNq. The PN is exclusively connected to the output signal of the 200 Walsh generator_r and PNq signals are input to the Finite Impulse Response 206, 208 FIR filters. The signals filtered by the FIR filters 206, 208 are fed to a transmit power control circuit consisting of gain control elements 210,212. The gain of the signals to the gain control elements 210, 212 is controlled according to the input signals (not shown) from the control processor. The output signals of the gain control elements are transmitted to a transmit power amplifier circuit 58. Its structure and operation will be described below.

HU 216 989 Β

After encoding, the synchronization channel information is multiplied by exclusive OR gates by a predetermined sequence of Walsh signals. In the exemplary embodiment, the selected Walsh function is (W₃₂) is a sequence of thirty-two and then thirty-two zeros. The resulting sequence of signals is then multiplied by exclusive OR gates PN_r and PN_Q-jelsorozatokkal. In the exemplary embodiment, the synchronization channel data information is usually transmitted to the transmit modulator at a rate of 1200 bits / s. Optionally, the synchronization channel data is preferably subjected to convolutional coding at r = 1/2 coding rate and K = 9 constraint length, and each code signal is repeated twice. This encryption rate and restriction length applies to all encoded outbound connections, i.e., both the synchronization, paging and voice channels. In one embodiment, a shift register system is used for G [= 753 (octal) and G₂= 561 (octal) for code generators. The synchronization channel has a code rate of 4800 code / s, which is 208 ps, which corresponds to 256 PN chips.

The code signals are inserted using the convolutional interleaver, which may be 40 ms. The parameters of the interleaver are 1 = 16 and J = 48. For further details of the interleaving method, refer to "Data Communication, Networks and Systems". (Howard W. Sams & Co., 1987) 343-352. pages. The convolutional interleaver spreads untrusted channel code signals by separating any two code signals within a coherent sequence of 1-1 code symbols or less with at least J + 1 code signals at the output of the deinterleaver. Likewise, within a contiguous sequence of J1 code numbers or fewer, any two code signals are separated by at least 1 + 1 code signals at the output of the deinterleaver. In other words, if 1 = 16 and J = 48, in a chain of fifteen code symbols, the transmission of each code signal is 885 ps, so that there is a diversity of time.

The code signals belonging to the synchronization channel of a particular cell or sector are linked to the pilot signal of that cell or sector. Figure 5 illustrates the timing of two different pilot channels (N) and (N + 1) when the two channels are offset by sixty-four chips. 5. FIG. 6B is an exemplary timing diagram of the exemplary pilot and synchronization channel and does not indicate the status of the current pilot signal chips and the synchronization channel. FIG. Each of its synchronization links initiates a new interleaver cycle and the (c_x, c '_x) first (c_x) is shifted by the same amount of time as the corresponding pilot due to duplicate code repetition.

As shown in Figure 5, the N pilot pilot initiates a new interleaver cycle (pilot signal synchronization)_x time. Similarly, the N + 1 pilot channel initiates a new interleaver cycle (pilot signal synchronization)_Y at a time that is offset by sixty-four chips_x time. In the exemplary embodiment, the pilot cycle length is 26.67 ms. This corresponds to one hundred twenty-eight synchronization links code symbols and thirty-two synchronization links information bits. The coding signals of the synchronization channel are inserted by a convolutional interleaver of 26.67 ms. This way, if the mobile unit finds the pilot signal, synchronization channel interleaver is also ensured

The code symbols of your synchronization links are covered with a predetermined Walsh sequence to ensure the signal's orthogonality. In the case of a synchronization link, a code symbol is written to four overlapping sequences. Thus, in the case of Fig. 6, a code symbol extends four repetitions of a sequence of thirty-two single and thirty-two zeros. As shown in Figure 6. 10A, a single logical "single" thirty-two "single" Walshchips and a single logical "zero" thirty-two "zero" Walsh chips. Even though the synchronization channel code signals are offset relative to the corresponding pilot channel in absolute time, the synchronization channel offsets are orthogonal, since the offsets of the synchronization channel are an integer multiple of the Walsh time frame.

In the exemplary embodiment, the messages in the synchronization channel are of variable length. The message length is an integer multiple of 80 ms corresponding to the three pilot cycles. The synchronization channel information bits include cyclic redundancy (CRC) bits for error detection.

Figure 7 is a timing diagram illustrating the total timing of the exemplary system. Each two-second period contains seventy-five pilot cycles. In Fig. 7, the pilot and synchronization link N represents a cell or sector that uses a non-offset pilot signal, i.e., in which the pilot and synchronization signals are exactly matched to UTC time. Thus, the synchronization of the pilot signal, i.e. the initial state, is exactly the same as a standard signal of 1 pulse / s (1 pps).

In each case where an offset pilot signal is used, a PN phase offset corresponding to the offset of the pilot signal is performed. In other words, the pilot signal synchronization (initial state) and synchronization channel messages are different from the 1 pps signal. The synchronization messages contain this phase shift information, so the mobile unit can adjust its own timing accordingly.

The mobile unit can synchronize itself to a paging channel or voice channel as soon as it receives the message correctly transmitted by its synchronization channel. In the pilot state, which corresponds to the end of each synchronization message, a new 40 ms interleaver cycle always starts. In this case, the mobile unit starts either a code repetition or a (C_x, C_{x + 1}) deinterleaving the first few code signals when accessing the decoder synchronization. The deinterleaver entry address is set to zero13

HU 216 989, and initialize its read address to J and the deinterleaver is now synchronized.

The messages of the synchronization channel carry information about the state of a forty-two bit PN generator of the audio channel designated for communication with the mobile unit. This information is used by the mobile unit digital data receivers to synchronize the corresponding PN generators. 7. For example, the message of the N + 1 synchronization channel comprises a field of forty-two bits denoting the X state represented by a PN generator having a long code corresponding to the sector or cell audio channel at a predetermined time (e.g., 160 ms). After successfully decoding a message from its synchronization channel, the mobile unit puts the long-coded PN generator into the X state at the correctly selected time. In this way, the PN generator of the mobile unit with its long code is synchronized and can decode the message to the user.

Returning to FIG. 4a .... c. 3A, the synchronization link message of the transmit modulator and power control circuit 66 associated with the synchronization channel is transmitted from the control processor to the encoder 214. In the exemplary embodiment, the data of the synchronization channel is subjected to convolutional encoding (as described above) by the encoder 214. The encoder 214 also provides for repetition of the code signals if the synchronization channel repeats the code signals. The code signals from the output of the encoder unit 214

215 they get into an interleaver, which performs the convolutional insertion of code signals (interleaving). After insertion, the code signals are output from an interleaver 215 to an input of an exclusive OR gate 216.

218 Walsh generator a (W₃₂) Produces a signal corresponding to a Walsh sequence which is output to the other input of the exclusive OR gate 216. The 216 exclusive OR gates are the code sequence of your synchronization channel and the (W₃₂) Walsh string is subjected to an exclusive OR operation. The resulting signal is applied to one input of two exclusive gates 220, 222.

The other input of exclusive OR gate 220 receives the PNj signal, while the other input of exclusive OR gate 222 receives the PNQ signal. A PN_r and PNq marks a

216 After an exclusive OR operation with an exclusive OR gate output signal, a finite impulse response FIR filter 224, 226 is input. The signals filtered by the FIR filters 224, 226 are fed to a transmitting power control circuit comprising digitally variable gain control elements 228, 230. The gain of the input signals of the gain control elements 228, 230 is digitally controlled according to the digital input signals (not shown) received from the control processor. The output signals of the gain control elements 228,230 are fed to the transmit power amplifier circuit 58.

The paging channel information is also encoded by repetition and insertion and then multiplied by a predetermined Walsh sequence. The resulting signal sequence is then multiplied by PN_r and PNq sequences. The paging channel data rate for a particular sector or cell is included in a portion of the message on your synchronization link. Although the paging channel data rate is variable, in the exemplary embodiment, it is recorded for each system according to one of the following exemplary values: 9.6; 4.8; 2.4 and 1.2 Kbit / s.

In the paging channel transmit modulator and power control circuitry, paging channel information from the control processor is transmitted to 232 encoders. In the exemplary embodiment, the encoder unit 232 is a convolutional encoder which also provides for repetition of the code signals according to the data rate assigned to the channel. The output signal of the encoder 232 passes to the interleaver 233, where the convolutional insertion of the code signals takes place. The output signal of the interleaver 233 goes to one of the inputs of the exclusive OR gate 234. Although the paging channel data rate varies, the rate of code signals is constant through code repetition,

It is maintained at 19.2 Kbit / s.

236 With the help of a Walsh generator, a signal corresponding to a pre-assigned Walsh sequence is generated and applied to the other input of the exclusive OR gate 234. The code signal data and the Walsh sequence is subjected to an exclusive OR operation 234 by an exclusive OR gate, resulting in 238, 240 exclusive OR gates for each input.

The other input of the exclusive OR gate 238 is assigned the PNp signal and the other input of the exclusive OR gate 240 is the PNq signal. The PN, and PNq signals, together with the output of the exclusive OR gate 234, are subjected to an exclusive OR operation. The resulting signals are applied to one of the inputs of FIR filters 242, 244 having a finite impulse response. The output signals of the FIR filters 242, 244 are fed to a transmit power control circuit consisting of gain control elements 246, 248. The gain of the signals provided to the gain control elements 246, 248 is controlled according to the signals (not shown) received from the control processor. The output signals of the gain control elements are fed to the transmit power amplifier circuit 58.

The data for each audio channel is also encoded by repetition and insertion, followed by encryption and multiplication by the assigned (Wj-Wj) sequence, and finally PN, - and PNq multiplication. The Walsh sequence to be used by a given channel is determined by the system control unit when establishing the call connection, just as the channels are assigned to the calls in the analogue FM cellular system. In the exemplary embodiment described herein, up to sixty-one Walsh sequences are available for use by the audio channel.

In the exemplary embodiment of the present invention, variable data rates are used for voice channels. The purpose of using variable data rate is to reduce the data rate to periods when there is no voice traffic14

HU 216 989 Β get. This reduces the amount of interference this particular audio channel causes to other users. A voice encoder (vocoder) provides variable data rates, depending on the voice activity, based on a 20 ms timeframe with four data rates. In the exemplary case, these are: 9.6 Kbit / s; 4.8 Kbit / s; 2.4 Kbit / s and

1.2 Kbit / s. Although the data rate varies on a 20 ms basis, the data rate of the code signals through code repetition remains constant at 19.2 Kbit / s. Correspondingly, the code symbols are repeated twice, four, or eight times at 4.8 Kbit / s; 2.4 Kbit / s and 1.2 Kbit / s respectively.

Because the variable data rate method is designed to reduce interference, code signals are transmitted at lower data rates with less power. The said 9.6 Kbit / s; 4.8 Kbit / s; For example, for code values ​​of 2.4 Kbit / s and 1.2 Kbit / s, the code symbols are E_s energy of E (/ 2, E (/ 4, Ej / 8 and Ε, / 16 (E, respectively))_b the information bit rate at 9.6 Kbit / s).

The codewords are inserted by the convolutional interleaver in such a way that the codewords of different energy levels are mixed (coded) according to a cipher. In order to keep track of the power level of a code signal, each code signal has a tag attached to indicate the data rate of the code signal for scaling. After coverage with orthogonal Walsh functions and PN scattering (distribution), 90 ° offset phase channels are subjected to digital filtering by an FIR filter. The FIR filter receives a signal corresponding to the energy level of the code signal and scales the energy corresponding to the data rate. The energy scaling factors for I and main channels can be as follows: 1; 1/2; 1/2 or 1/2 2. In one embodiment, the vocoder provides the data rate tag to the FIR filter in the form of a two-bit number to control the filter scaling factor.

A 4a .... 4c. Figures 6 to 9 show exemplary circuits of two audio channels i and j. The voice channel i data is transmitted from the dedicated vocoder (not shown) to the transmit modulator 54 (Fig. 2). Transmitter modulator 54 is 250; coder, 251 µ interleaver, 252 µ_{i (} 255j, 256; and 258; exclusive OR gates, 253j PN generator and 254; It consists of (W;) a Walsh generator.

Audio channel i data is 250; encoder which, in the exemplary embodiment, subverts the data to convolutional encoding and repeats the code signals according to the input data rate. The encoded data is then retrieved by the 251; into an interleaver that performs convolutional insertion. A 251; The interleaver also receives a two-bit data tag from the vocoder for the audio channel i, which is inserted between the code signal data to identify the rate of data going to the FIR filter. The data rate tag is not broadcast. The mobile unit decoder checks all possible codes. For example, after completing the paste, the code signal data

19.2 Kbit / s data is delivered to 251; interleaver 255; to one of the inputs of an exclusive OR gate.

In the exemplary embodiment, each voice channel is encrypted for increased security from cell to cell. Although this encryption is not strictly necessary, it is useful to increase the reliability of communications. For example, encryption of voice channel signals may be accomplished by PN encoding the voice channel signals according to the PN code corresponding to the handset address specified by the user ID. To encrypt, refer to section 3. 5A, the PN_CN or encryption scheme described for a particular receiver for connection from the mobile unit to the cell-site may be used. This function is illustrated in Figures 4a ..., 4c. 1 to 4, a separate PN generator is implemented. Although shown here with only one PN sequence, encryption can, of course, be accomplished using other techniques known per se.

Returning to Figs. 4a .... 4c. FIGS. 253 to 253; This can be accomplished by a PN generator, which obtains the address of the assigned mobile unit from the control processor. A 253; The PN generator generates a unique PN code which is 255; to the other OR of the exclusive gate. A 255; the exclusive OR gate output is 252; connected to one of the inputs of an exclusive OR gate.

A 254; (W;) The Walsh generator receives a function select signal and a timing signal from the control processor. Based on these, it produces a signal corresponding to a predetermined Walsh sequence. The value of the function select signal is determined by the address of the mobile unit. A 252; the other input of the exclusive OR gate will receive this signal corresponding to the Walsh sequence. The encrypted code data and the Walsh sequence are described in 252; exclusive OR gate is subject to an exclusive OR operation. The signal obtained is 256; and 258; at one of the inputs of exclusive OR gates. A 253; PN generator (like other PN generators and cell-centered Walsh generators) produces 1.2288 MHz output signal. It should be noted that 253; The PN generator includes a decimator that transmits a 19.2 kHz output signal to 255; exclusive OR gate input.

A 256; for the other OR of the exclusive OR gate, PN_rmark, 258; and the other input of the exclusive OR gate is marked with PNq. A 252; PN along with the exclusive OR gate output_r and PNq signals with finite impulse response 260 ;, 262; FIR filters are input. The input code symbols are 251; are filtered according to the input data rate label (not shown) from the convolutional interleaver. A 260 ,, 262; The signals filtered by the FIR filters are transmitted by transmitter power controller 56, 266; gain control part. A 264 ;, 266; the gain of the signals to the gain control elements is controlled according to the input signals (not shown) received from the control processor. The output signals of the gain control elements are fed to the transmit power amplifier circuit 58.

Outbound Audio Output Channel - Outside Audio Bits - Power Control15

GB 216 989 also carries information. In the exemplary embodiment, the power control bit rate is 800 bit / s. The power control information is generated by the cell-center demodulating receiver, which is transmitted by the mobile unit and received by the mobile unit-cell-center link. This information is inserted into the voice channel from the cell-site to the mobile unit, addressed to that mobile unit.

The bits carrying the power control information are inserted at the output of the convolutional interleaver by a method known as code break. In other words, when transmitting power control bits, two code signals are replaced by two identical code signals whose polarity is given by power control information. The power control bits are transmitted at an energy level corresponding to 9600 bit / s data rate.

An additional constraint on the power control information stream is that the location of the bits is randomly distributed between the channels from the mobile units to the cell center. Otherwise, the full power power control bits would produce interference peaks at regular intervals, reducing the visibility of these bits.

4a ..., 4c. The function and structure of the audio channel j shown in Figs. In the illustrated embodiment, a total of up to sixty-one audio channels are possible, but only two of these are shown.

A 4a .... 4c. In the case of Walsh generators of the system of Figures 1 to 4, the Walsh functions consist of a set of orthogonal binary sequences which can be easily generated by means well known in the art. An important feature of the Walsh function in this case is that all sixty-four chips are perfectly orthogonal to all other chips. So any two sets of signals differ in exactly as many bit positions as they do, that is, out of sixty-four codes in thirty-two. Thus, when encoding information for transmission with the Walsh sequence, the receiver can select any Walsh sequence as the desired "carrier" signal. All signals encoded with additional Walsh sequences will be rejected and will not interfere with the selected single Walsh sequence.

In the exemplary embodiment, convolutional encoding is used for the connection from the cell-site to the mobile unit, as mentioned above, with convolutional coding at a constraint length K = 9 and a code rate r = 1/2 we create and transmit two encoded code signals per bit of information). In addition to convolutional coding, convolutional interleaving of code signal data is also used. Optionally, repetition is used in addition to convolutional insertion. When using this type of code, a Viterbi algorithmic decoder should be used. The standard basic design is suitable for decoding. The resulting decoded information bits are transmitted to the baseband digital device of the mobile unit.

Returning to FIGS. 4a ..., 4c. 5A, the transmit power amplifier circuit 58 comprises a plurality of D / A (digital / analog) transducers that are PN, pilot, synchronization, paging, and voice channels._r and to convert digital information carried by its PNq scattered (distributed) data into analogue form. The PNj broadcast data of the pilot channel is transmitted from the gain control unit 210 to the D / A converter 268. Digitized data is fed from 268 D / A converters to 284 totalizers. Similarly, PNP broadcast data for synchronization, paging, and voice channels 228, 246, 264; and output signals of gain control elements 264j and 264j, 272,276, 280; and 280j to D / A converters, which digitize the signals and transmit to the summing 284. PNq broadcast data for pilot, synchronization, paging, and voice channels are shown in 212, 230, 248, 266; and outputs 270, 274, 278, 282 of gain control elements 266j; and 282j to D / A converters, which digitize the signals and transmit them to a 286 summator.

Summarizer 284 aggregates the ΡΝ, distributed signals of pilot, synchronization, paging, and voice channels, while aggregator 286 aggregates PNq data for the same channels. The summed I- and Q-channel data are provided to the mixers 288, 290 along with sin (2nft) and cos (2nft) signals corresponding to the local oscillator frequency of LO. These signals are mixed and transmitted to 292 summers. The sin (2nft) and cos (2nft) signals are provided by a frequency source (not shown). The mixed IF signals are summed by 292 and transmitted to mixer 294.

The mixer 294 mixes the summed signals with the radio frequency (RF) signals generated by the frequency synthesizer 296, i.e., converts them into the RF frequency band. The RF signal exiting the mixer 294 is filtered by the bandpass filter 298 and transmitted to the RF amplifier 299. The RF amplifier 299 amplifies the bandwidth signal according to the gain control signal received from the transmit power control circuit 56 (FIG. 2). It should be noted that the transmit power amplifier circuit 58 shown is an example only. Signal aggregation, mixing, filtering, and amplification can be accomplished in many other variants known in the art.

The cell-center control unit 48 (Figure 3) assigns digital data receivers and transmit modulators to each call. The control unit 48 also monitors the conversation process, signal strength and initiates disconnection in case of signal loss. The cellular center communicates with the MTSO via connection 52 to which it is connected by conventional telephone line, fiber optic or microwave connection.

Figure 8 is a block diagram of the equipment used in the MTSO. The MTSO generally comprises 300 system control units (control processors), a digital switch 302, a diversity combiner 304, a digital vocoder 306, and a digital switch 308. Additional diversity combiner units and digital vocoders are provided between digital switches 302 and 308

EN 216 989 Β which for the sake of simplicity are not shown.

In cell discrimination mode, the call is processed by two cell centers. Accordingly, the MTSO center receives signals from more than one cell center with nominally the same information content. However, due to fading and interference from the mobile unit to the cell-site and in the reverse direction, the signal received from one cell-site may be of better quality than the signal from the other cell-site.

Digital switch 302 transmits the information stream received from one or more cell centers to the diversity combiner 304 or another diversity combiner 304 according to the signal received from the system control unit 300. If the system is not in cell discrimination mode, the diversity combiner 304 can be bypassed or the same information can be transmitted to each input port.

In the system of Fig. 8, a plurality of members, each in diversity connected to each other and a vocoder, are connected in parallel, usually for each conversation to be processed. The diversity combination unit 304 compares the signal quality indicators accompanying the information bits from two or more cell centers. The diversity combiner 304 selects the bits of the cell center providing the highest quality signal per information time frame and forwards the result to the vocoder 306.

The vocoder 306 converts the digitized audio signal to a standard 64 Kbit / s PCM telephone format or other standard format and transmits the received signals to the digital switch 308. The call is directed to the public switched telephone network (PSTN), controlled by the system control unit 300.

The audio signals from the PSTN network addressed to a mobile unit are transmitted to digital switch 308, which transmits them to a designated voice coding unit, such as a vocoder 306, according to instructions from the system control unit 300. The vocoder 306 encodes the digitized input audio signals and transmits the resulting information bit stream directly to the digital switch 302. Digital switch 302 directs the encoded data to the cell center or cell centers to which the mobile unit is communicating in accordance with the instructions of the system control unit. Although the information transmitted to the MTSO center is discussed above as an analog audio signal, it is of course also possible to transmit digital information. Care must be taken to ensure proper framing of the data to ensure system compatibility.

When the mobile unit is in handover / receive or cell discrimination mode and communicates with multiple cell centers, digital switch 302 transmits the call to the appropriate cell centers for transmission to the called mobile receiver by the appropriate cell center transceiver. However, if the mobile unit is in communication with only one cell center or is not in cell discrimination mode, the signal is transmitted to only one cell center.

The system control unit 300 controls the digital switches 302,308 which transmit data to and from the MTSO center. The system control unit 300 further determines which cell centers and which vocoders are to be assigned to the conversations in the MTSO center. The system control unit 300 also communicates with each cell-site control unit for assigning each call between the MTSO exchange and the cell-sites, and for assigning PN codes to each call (call). Note that although digital switches 302, 308 are shown in Figure 8 as separate units, this task can be physically solved by a single switch unit.

In cell differentiation mode, the mobile unit uses the paging receiver to select the strongest signal from the two cell centers through multiple signal paths and to "stand on it". The digital data receivers are controlled by the paging receiver and the control unit in such a way that they demodulate the strongest signal. If the number of receivers is less than the number of cell centers transmitting the information in parallel, a switching discrimination mode may be used. If, for example, there is only one data receiver and two cell centers are transmitted, the paging receiver monitors the pilot signal of both cell centers and selects the stronger signal for the receiver to demodulate. In this embodiment, selection is possible once per vocoder time frame, i.e., about 20 ms.

The system control unit selects appropriate cell-centered digital data receivers and modulators to handle each call. Thus, in the case of communication from the cell-site to the mobile-unit, the system control unit (processor) controls the assignment of the Walsh sequence used by the cell-site for a specific call transmitted to the mobile-unit. The system control unit also controls the receiver's Walsh sequences and PN codes. In the case of communication from the mobile unit to the cell-site, the system control unit also controls the user PN codes used in the mobile unit call. For this purpose, assignment information is transmitted from the MTSO center to the cell center and thence to the mobile unit in that cell. The system control unit also monitors the conversation, signal quality, and initiates disconnection in case of signal loss.

In the case of a connection from the mobile unit to the cell-site, the characteristics of the particular channel determine whether modulation techniques should be modified. The pilot carrier used for communication from the cell-site to the mobile unit cannot be used here. Namely, the pilot carrier must be more powerful than the speaker carrier to provide good phase reference for demodulating the data. In cases where the cell center is transmitting a large number of speakers at the same time, each speaker may share a single pilot signal.

HU 216 989 Β

Therefore, the performance of the pilot signal per speaker is relatively low.

However, in the case of communication from the mobile unit to the cell-site, there is usually only one voice carrier per mobile unit. If a pilot signal were used, it would have to be much more powerful than the speaker. Clearly, such a situation is undesirable as it would greatly reduce the overall capacity of the system due to the interference caused by the presence of a large number of high power pilot signals. Therefore, modulation should be used that allows efficient demodulation without a pilot signal.

For channels from the mobile unit to the cell center, Rayleigh-fading occurs. As a result, the channel phase changes rapidly. Therefore, coherent demodulation techniques that derive the phase from the received signal (such as the Costas loop) are not applicable. Other methods, such as differentially coherent PSK (Phase Shift Keying) may be used, but may not provide the desired signal-to-noise ratio.

Consequently, some form of orthogonal signaling, such as second, quarter, or m order signaling, is preferred. In the exemplary embodiment, 64-bit orthogonal signaling is used using Walsh functions. In order to demodulate the mth order orthogonal signals, the channel coherence need only exist for the duration of the mth order signal transmission. In the exemplary embodiment, this is only two bits.

The coding and modulation of the message begins with a convolutional encoder characterized by a constraint length K = 9 and a code rate r = 1/3. At a nominal data rate of 9,600 bits / sec, the encoder generates 28,800 bits per second. These are grouped into six characters each, at 4800 characters per second. The number of possible characters is sixty-four. Each character is encoded in a 64-bit Walsh sequence. These contain sixty-four binary bits ("chips"). The 64th order Walsh chip in the exemplary embodiment has a speed of 307,200 chips / s.

The Walsh chips are then "overlaid" (multiplied) by a 1.2288 MHz PN signal sequence. Each mobile unit is assigned a unique PN sequence for this purpose. This PN sequence is either permanently assigned or only assigned to the mobile unit for the duration of the call. This assigned PN sequence is called a user PN sequence. The PN generator, which generates a user PN sequence, operates at a clock frequency of 1.2288 MHz, thus generating four PN chips per Walsh chip.

Finally, a short PN sequence pair of 32,678 lengths is produced. In the exemplary embodiment, the same sequence of signals is used as for the connection from the cell-site to the mobile unit. The Walsh chip sequence "covered" by the user's PN sequence is then overlaid (multiplied) by two short PN sequences. The two sets of signals thus obtained are biphasic modulating a pair of sine squares and summing them into a single signal. The resulting signal is then subjected to bandpass filtering, converted to the final RF frequency, and finally amplified and filtered by the mobile unit's antenna. As noted in the case of communication from the cell-site to the mobile unit, the order of filtering, amplification, frequency conversion and modulation operations can be reversed.

In another embodiment, two different phases of the user PN code can be generated and used to modulate two carrier phases of the four-phase waveform without the need for 32768 length sequences. In yet another embodiment, in the case of communication from the mobile unit to the cell-site, it is possible that the modulation used is biphasic only. Again, short sequences are not required.

The cell-site receiver generates short PN sequences for each signal, and generates a user PN sequence for each received mobile signal received. The receiver, in separate correlators, correlates the received signal energy with each encoded waveform. It then processes the output of each correlator separately to demodulate 64-bit coding and convolutional coding using a fast Hadamard transform (FHT) processor and a Viterbi algorithmic decoder.

Another possible modulation system for a cell-to-cell connection is the same modulation technique as for a cell-to-cell connection. In this case, each mobile unit uses the 32,768 length code pair as an external code. The internal code is a 64-bit Walsh sequence that is assigned to the mobile unit while it is in the sector. Thus, the same Walsh sequence is assigned to the mobile unit for the connection from the mobile unit to the cell-site as used for the reverse link.

The above orthogonal PN coding technique limits the available bandwidth allocation that can be utilized by the modulation system according to sixty-fourths of the maximum chip frequency (corresponding to 19200 Hz for numerical examples). This would exclude m-coding for large values ​​used in the exemplary embodiment. However, it is possible to use a convolutional code of r = 1/2 rate and K = 9 constraint length with differential binary phase keypad (DPSK) modulation of the encoded binary code signals. In this case, the cell-center demodulator may generate a phase reference for a short interval (see AJ Viterbi, AM Viterbi, Nonlinear Estimation of PSK Modulated Carrier with Application to Burst Digital Transmission, IEEE Transactions on Information Theory, Vol. IT-29, No. 4). ., 1983). This phase reference is, for example, the length of only one code signal

It may be averaged over a reasonable period of time, which does not require more channel coherence than the 64-bit coding above.

However, the above alternative solution does not provide a quality service like the preferred embodiment for strong Rayleighfading and multiple signal paths. However, where the effects of multiple signal paths and fading phenomena are insignificant (for example, from satellite to mobile stations, and for some terrestrial mobile connections), the above alternative solution may provide a better service than the preferred embodiment. This is possible because the gains from rendering mobile signals orthogonal may outweigh the losses in DPSK's sensing efficiency.

In order to satisfy the requirement for time-matching of the orthogonal Walsh functions in the above alternative solution for communication from the mobile unit to the cell-site, each cell-site receiver determines the time difference of each received signal relative to the nominal timing. If a received signal is delayed relative to the timing, the associated cell-center modulator and transmitter instructs this mobile unit to push forward its transmit timing by a small amount. In contrast, if the timing of a received signal from a mobile unit is in a hurry relative to the nominal timing, the mobile unit will be instructed to reset its timing by a small amount. The timing correction steps are of the order of 1/8 PN chip or 101.7 nanosec. The instruction output frequency is relatively low (10-50 Hz). The instruction consists of a single bit embedded in the digital audio data stream.

In a cell-to-cell transition, the mobile unit carries signals from two or more cells. Since the mobile unit can only correct the timing according to the timing setup instruction of a cell, the mobile unit usually adjusts the timing according to the instruction received from the cell with the strongest signal. The signal provided by the mobile unit is thus in coordination with the cell from which it receives the best path signals. Otherwise, there would be greater interference with other users.

If all cell-site receivers receiving the mobile signals perform the above-mentioned time error measurement operations and transmit correction correction transmissions, the timing of all signals received by the mobile units will be approximately the same. This reduces interference.

FIG. 9 is a block diagram of an exemplary CDMA (Code Distributed, Multiple Access) telephone set for a mobile unit. The mobile unit's CDMA telephone has an antenna 430, which is connected via a diplexer 432 to an analog receiver 434 and a transmit power amplifier 436. The standard design 430 antenna and 432 diplexer allow simultaneous transmission and reception of a single antenna. The antenna 430 collects the signals and transmits them to the analog receiver 434 via the diplexer 432. Analog receiver 434 receives RF signals from diplexer 432.

The frequency of the RF signals falls, where applicable, within the 850 MHz band. The signals are amplified and converted to an IF frequency (center frequency). This frequency conversion process is performed by a standard frequency synthesizer which allows the receiver to be tuned to any frequency within the receiving frequency band of the full cellular telephone system. After filtering and digitizing, the signals are fed to 440, 442 data receivers and 444 search receivers.

Further details of the analog receiver 34 are shown in Figure 10. The signals received from the 430 antennas are transmitted to 500 converters. It consists of an RF amplifier 502 and a mixer 504. The received signals are fed to the input of RF amplifier 502. The RF amplifier 502 amplifies the signals and transmits it to an input of mixer 504. The other input of mixer 504 is coupled to the signal output of frequency synthesizer 506. The amplified RF signals are mixed by the mixer 504 with the output signal of the frequency synthesizer 506 and thus converted to IF signals.

The IF signals are transmitted from mixer 504 to band pass filter 508 (BPF, Bánd Pass Filter), which is generally designed as a surface acoustic wave filter (SAW) having a bandwidth of approximately 1.25 MHz. The characteristics of the SAW filter are selected according to the shape of the signals output from the cell center. The signal output from the cell center is a direct-sequence, clock-controlled, PN-modulated, spread-spectrum signal. In the exemplary embodiment, the clock frequency is 1.2288 MHz, which is an integer multiple of 9.6 Kbit / s baseband data rate.

The signals filtered by the bandpass filter 508 are fed to a variable gain IF amplifier 510 which further amplifies them. The amplified IF signals are transmitted from the IF amplifier 510 to a 512 A / D converter (analog / digital converter) which digitizes them. The IF signal is converted into a digital signal in the exemplary embodiment at a clock frequency of 9.8304 MHz. This clock frequency is exactly eight times the PN chip frequency. Although the A / D converter 512 is indicated as being part of the analog receiver 434, it may otherwise be part of the data receiver or paging receiver. The digitized IF signals are transmitted from the A / D converter 512 to the data receivers 440, 442 and the search receiver 444.

The analog receiver 434 also provides a power control function to adjust the power of the mobile unit. The IF amplifier 510 also has an output of 514 automatic gain control (AGC) circuits. Depending on the amplitude level of the amplified IF signal, the AGC circuit 514 provides a feedback signal to the gain control input of the IF amplifier 510. The analog receiver 434 also uses the AGC circuit 514 to generate an analog power control signal. This signal is applied to the transmit power control circuit 438.

As shown in FIG. 9, the digitalized output signal of the analog receiver 434 is transmitted to the digital data receivers 440, 442. It should be noted that a low-cost, low-performance mobile unit may have only one data receiver, while a high-performance mobile unit may have two or more. These allow discriminatory reception.

HU 216 989 Β

The digitized IF signal may include a plurality of ongoing call signals, along with pilot carriers provided by current and adjacent cell centers. The 440,442 digital data receivers correlate IF samples with the corresponding PN signal sequence. This correlation process results in a so-called "processing gain" which improves the signal-to-noise ratio of the assigned PN signal sequence, while not improving the rest. The output of the correlation process is then detected synchronously by the mobile unit using a pilot carrier of the nearest cell center as a carrier phase reference. As a result of this detection process, coded data code signals are obtained.

One of the features of the PN sequence used in the present invention is that it allows discrimination between signals received over multiple paths. When the signal arrives at the mobile receiver over more than one route, the signal reception time varies. This difference in reception times corresponds to the ratio of the distance difference to the propagation speed. If this time difference is greater than 1 ps (one millionth of a second), then the correlation process distinguishes paths. In this case, the receiver can choose whether to follow or receive the earlier or later route. If two receivers are available, such as digital data receivers 440 and 442, then two independent paths can be tracked and processed in parallel.

Controlled by the control unit 446, the paging receiver 444 continuously checks a time range around the nominal time of the pilot signal received from the cell center to determine if a multipath or other cell center pilot signal from the same cell center can be received. The paging receiver 444 measures the strength of the signals corresponding to the desired waveform at non-nominal times and compares the received signal strength values. The search receiver 444 provides a signal strength signal indicating the strength of the strongest signals to the control unit 446.

The control unit 446 provides control signals to the digital data receivers 440,442 in order to process one of the strongest signals. Optionally, the pilot signal from another cell center is stronger than the received cell center signal. The control unit 446 then generates a control message which is transmitted to the system controller via the current cell center, requesting that the cell be moved to the cell corresponding to the strongest pilot signal. The digital data receivers 440, 442 then handle the calls through two different cell centers.

During a cell-to-cell transition, the mobile unit receives signals from two or more cell centers. Because the mobile unit can only set its own timing according to the timing correction instruction of a cell, it usually "obeys" the command it receives from the strongest cell. Thus, the signal issued by the mobile unit is coordinated in time with the cell with which the mobile unit is in contact with the highest quality signal path. Otherwise, there would be greater interference with other users.

Further details of the exemplary digital data receiver 440 are shown in Figure 10. The digital data receiver 440 includes PN generators 516, 518 which, in a manner similar to and corresponding to the series of signals produced by the cell center, PN_r and PNq sequences are generated. For PN generators 516, 518, the control unit 446 provides timing and sequence control signals. The digital data receiver 440 also includes 520 Walsh generators. This produces the function by which the cell-site communicates with the given mobile unit. The Walsh generator 520 generates a signal corresponding to the selected Walsh sequence based on timing signals (not shown) and a function select signal from the control unit (processor). The cell exchange transmits the function selection signal to the mobile unit as part of the call setup message. PN issued by PN generators 516, 518_rand PNQs 522, 524 are assigned to one input of exclusive OR gates. The output signal of Walsh generator 520 is applied to one of the additional inputs of exclusive OR gates 522, 524. Signals subjected to an exclusive OR operation result in PN] 'and PNq' sequences.

PN] 'and PN_Qsequences in the 440 digital receiver receive 526 PN QPSK correlators. The PN QPSK correlator 526 may have a similar design as the PN correlator of the cell-site digital receivers. The PN QPSK correlator 526 correlates the received I and Q link signals with the PN] 'and PNQ' sequences and then transmits the correlated I and Q link data to a 528, 530 battery (data logger). The batteries 528, 530 collect (store) input information for a code signal, i.e., sixty-four chips. The output signals of the accumulators 528, 530 are supplied to a phase inverter 532 which also receives a pilot phase signal from the control unit 446. The phase of the received code signal data is rotated according to the phase of the pilot signal determined by the paging and control unit. The output signal of the phase inverter 532 is the I-channel data, which is supplied to the deinterleaver and decoder circuit.

The control unit 446 also includes a PN generator 534 which generates a user PN sequence corresponding to an input handset address or user ID (ID). The PN signal sequence issued by the PN generator 534 is fed to a diversity combining and decoding circuit. Since the signal transmitted from the cell to the mobile unit is encrypted with the PN sequence of the mobile user address, the output of the PN generator 534 is to decrypt (decode) the signal transmitted from the cell center to that mobile user. This is the same as in the case of a cell-site receiver. PN generator 534 transmits the PN sequence to the deinterleaver and decoder circuit, which uses it to decrypt the encrypted user data. Although encryption is discussed herein in relation to PN sequences, it is obvious that other encryption techniques, well known in the art, may be used.

The output signals of the digital data receivers 440, 442 are thus transmitted to the diversity combining and decoding circuit 448.

HU 216 989 Β

Its diversity combining unit overlaps the two received code streams in time, and then adds them together. This summing operation may be preceded by multiplying the two streams by a factor corresponding to the relative strength of the two streams. This operation can be considered a discriminatory combination of a maximum ratio. The resulting combined stream is decoded by a forward error correction (FEC) decoder, which is also part of the circuit 448. The standard baseband equipment is a digital vocoder system. The CDMA system is designed to be used with a variety of types.

The baseband circuitry 450 generally includes a digital vocoder (not shown). The baseband circuitry 450 also serves as an interface for a handset or other peripheral. The baseband circuitry 450 can be used for vocoders of various designs. The baseband circuit 450 provides outgoing information signals to the user based on information received from the circuit 448.

In the case of communication from the mobile unit to the cell-site, analog user voice signals are generally provided by a handset as an input to the baseband circuitry 450. The baseband circuit 450 includes an analog / digital A / D converter (not shown) that converts the analog signal to a digital signal. The resulting digital signal is sent to a digital vocoder which encodes it. The vocoder output signal is fed to a forward error correction FEC encoding circuit (not shown) for error correction. In the exemplary embodiment, the error correction coding is performed by a convolutional coding technique. The encoded digitized signal is transmitted from the baseband circuit 450 to the transmit modulator 452.

Transmit modulator 452 first performs Walsh encoding on the data to be transmitted and then modulates the encoded data to a PN carrier having a PN sequence selected according to the address function assigned to the call. The PN sequence is determined by the control unit 446 on the basis of information transmitted by the cell-site and received by the digital data receivers 440, 442 and the control unit 446 when the call is established. Alternatively, the control unit 446 may determine the PN sequence based on prior consultation with the cell center. The PN sequence information is transmitted by the control unit 446 to my transceiver unit 452 and to the digital data receivers 440, 442 for decoding the call.

Transmitter modulator 452 outputs a transmit power control circuit 438. Signal output power is controlled by an analog power control signal received from receiver 434. The control bits issued by the cell centers in the form of a power control command are processed by the digital data receivers 440, 442. The control unit 446 uses the power control command to control the transmit power of the mobile unit. As a result of this command, the control unit 446 generates a digital power control signal which is transmitted to the transmit power control circuit 438.

Transmitter power control circuit 438 transmits the controlled power modulated signal to transmitter power amplifier circuit 436. Transmit power amplifier circuit 436 amplifies the IF signal and converts it into a RF signal tuned to the appropriate output frequency by mixing the output of a frequency synthesizer. Transmit power amplifier circuit 436 raises the signal power to the appropriate output power level. The transmit signal is transmitted from circuit 436 to diplexer 432. Diplexer 432 couples the signal to antenna 340 for transmission to the appropriate cell center.

The control unit 446 is also capable of generating various control messages. These include, for example, messages requesting diversity mode and commands to terminate cell-site communications. These commands are transmitted to the transmit modulator 452 for transmission. The control unit 446 also decides on handover and discrimination combination operations.

When the mobile unit is in transmit mode, the mobile user's analog audio signal first passes through a digital vocoder. The vocoder output signal is then subjected to convolutional FEC encoding, 64-bit orthogonal coding, and PN carrier signal modulation, respectively. The 64th order orthogonal signal sequence is generated by a Walsh function encoder. The encoder is controlled by collecting six consecutive binary code signals from the convolutional FEC encoder. The six binary codes together determine which of the sixty-four possible Walsh sequences will be transmitted. The Walsh sequence is sixty-four bit long. Thus, for a 9600 bit / s data rate, the Walsh "chip" frequency should be 9600 χ 3 χ (1/6) χ χ 64 = 307 200 Hz.

For communication from the mobile unit to the cell-site, a common short PN sequence is used for all system speakers, while user addresses are encoded using a PN generator that generates a user PN sequence. The user PN sequence is assigned individually to the mobile units for at least the duration of the call. The user PN sequence is subjected to an exclusive OR operation on common PN sequences (these are 32,768 linear incremental maximum length linear sequence sequences). The resulting binary signals then individually modulate a 90 ° phase shift carrier and, after summing, form a composite signal. This is subjected to bandpass filtering and converted to an IF output signal. In the exemplary embodiment, part of the filtering process is performed by a finite impulse response FIR filter applied to the binary output.

The output of the modulator output signal is controlled by the signals received from the digital processor and the analog receiver, the signal is mixed with a frequency synthesizer signal tuning to the appropriate operating frequency, and converted to the desired output level. The signal to be transmitted is finally transmitted to the antenna via the diplexer.

HU 216 989 Β

FIG. 11 illustrates a preferred (also exemplary) embodiment of the transmitting modulator 452 of the mobile unit. The digital data from the user baseband circuitry is sent to an encoder 600 which undergoes convolutional coding of the signals in the exemplary embodiment. The output signal of the encoder 600 is fed to an interleaver 602, which may be configured as a block interleaver. Interleaving signals are transmitted from the interleaver 602 to the Walsh encoder 604 of the transmit modulator 452. Walsh encoder 604 generates an output code sequence from the input code signals. The resulting signal sequence is fed to one of the inputs of 606 exclusive OR gates.

Transmit modulator 452 further includes a PN generator 608 which uses the mobile unit address as an input signal to determine the outgoing PN sequence. The PN generator 608 produces the generator of FIGS. Figures 1 to 4 illustrate a user-specific forty-two bit sequence. Another unprecedented feature of the PN generator 608 that is common to all user PN generators is the use of a "masking" (masking) technique in the production of a user PN output sequence. For example, it generates a forty-two-bit mask for a given user, each bit of the mask being matched in an exclusive OR operation with the output bit of each of the shift registers that make up the PN generator. The signals obtained by comparing the mask and shift register bits with an exclusive OR operation are then also subjected to an exclusive OR operation. This is the output of the PN generator that is used as a user PN sequence. The output PNu signal sequence of PN generator 608 is output to one of the inputs of exclusive OR gate 606. The Walsh code tokens and the PNu sequence are compared by the exclusive OR gate 606 in an exclusive OR operation. The resulting signal is applied to one of the inputs of two additional exclusive gates 610, 612.

The transmit modulator 452 further includes PN generators 614, 616 generating PNj and PNq sequences. All mobile units have the same PN_r and PNq sequences. These PN sequences correspond to the zero offset used in the exemplary embodiment of the connection from the cell-site to the mobile unit. Exclusive OR gates 610, 612 for another input are PN produced by PN generators 614, 616_r and PN_Qsequences are added. The appropriate exclusive OR gates are PN_r and PNq, and the output signals to the transmit power controller 438 (FIG. 9) are compared in an exclusive OR operation.

In the exemplary embodiment, convolutional encoding at a rate of r = 1/3 at a constraint length K = 9 is used for communication from the mobile unit to the cell center. The code generators are G) = 557 (octal), G₂= 663 (octal) and G3 = 711 (octal). As in the case of a cell-site to mobile unit connection, code repetition is used to match the four data rates produced by the vocoder to the 20 ms timeframe. However, unlike cell-to-mobile communication, the solution here is not to transmit the repetitive code signals at a lower power level, but to transmit only one code signal from the repetition group, but at the nominal power level. Thus, the code repetition in the exemplary embodiment is used only to bring the variable data rate method into line with the interleaving and modulation techniques, as will be seen below.

In the case of communication from the mobile unit to the cell-site, a block interleaver is used that spans 20 ms, i.e. exactly one vocoder time frame. At a data rate of 9600 bit / s and a code rate of r = 1/3, this 20 msec receives 576 code signals. If the number of rows in the interleaver unit is N and the number of columns is B, then the values ​​of N and B are thirty-two, respectively.

18th Enter the code characters line by line into the interleaver storage unit. Readings are performed column by column.

The modulation format corresponds to 64th order orthogonal signaling. In other words, interleaved codewords are grouped into six groups to select one of sixty-four orthogonal waveforms. The sixty-four orthogonal waveforms that are orthogonal to each other at the time are made up of the same Walsh functions that were used for the cell-to-mobile connection as masking sequences.

The time interval for data modulation is 208.33 ps. This is called the time interval of the Walsh code signal. At a data rate of 9600 bits / sec, this 208.33 ps corresponds to two information bits, or six code signals, if the code signal rate is 28,800 code signals / s. The time interval of the Walsh code signal is subdivided into sixty-four time intervals of equal length, called Walshchips. The duration of these is 208.33 / 64 = 3.25 ps. The Walsh chip speed in this case is 1 / 3.25 ps = 307.2 kHz. Since the PN distribution rate is symmetric for the two connections, i.e. 1.2288 MHz, there are exactly four PN chips per Walsh chip.

In the case of communication from the mobile unit to the cell-site, a total of three PN generators are used: a user-specific PN generator producing a forty-two-bit PN signal and two PN generators for generating fifteen-bit I and Q link signals. Following the user-specific spreading (spreading) operation, the signal is subjected to QPSK spreading (spreading), as is the case for a cell-to-mobile connection. Unlike the cell-site to mobile unit (where each sector or cell is 2¹⁵), each mobile unit uses the same I and Q PN sequences. These PN sequences are zero-shift sequences used in communication from the cell-site to the mobile unit, also called pilot sequences.

In the case of communication from the cell-site to the mobile unit, code repetition and power scaling are used to accommodate the variable data rate generated by the vocoder. The mobile is a 22

However, in the case of communication from the cell to the cell center, a burst transmission based solution is used.

The vocoder (as with cell-to-cell communication) generates four different data rates at 96 ms, 9600, 4800, 2400, and 1200 bps using 20 ms timeframes. The information bits are encoded with a convolutional encoder of r = 1/3 speed, and for the three lower data rates the code signals are repeated two, four, and eight times, respectively. Thus, the value of the code rate remains constant (28,800 code / s). After encoding, the code signals are inserted by the block interleaver, which has an exact time frame (20 ms). The convolutional encoder generates a total of 576 code signals for every 20 ms, which may include repeated code signals.

The transmitted code sequence is shown in Figure 12. Notice that the vocoder time frame of 20 ms is divided into sixteen 1.25 ms slots. For the connection from the mobile unit to the cell-site, the numbers are as follows: At a rate of 28,800 codecs / sec, each time slot contains thirty-six codecs. This corresponds to a speed of 4,800 codecs per second and six Walsh signals per time slot. At 1/2 speed (4,800 cps), the time slices form eight groups with two time slots per group. At 1/4 speed (2400 bit / s), the time slices form four groups with four time slots per group. At 1/8 speed (1200 bit / s), the time slices form two groups of eight time slots each.

Figure 12 shows an example of how a burst transmission is performed. At a rate of 1/4, i.e. 2400 bits / sec, in the fourth time slice of the first group, read the fourth and eighth rows of the interleaver memory unit per column and output it for transmission. The position of the time slices is randomly assigned to the transmitted data to reduce the possibility of interference.

Figure 13 illustrates a case of communication from a mobile unit to a cell center. FIG. 13 is an extension of the timing diagram of FIG. 7 and includes channels from the mobile unit to the cell center, i.e., the voice channel and the access channel. To synchronize the connection from the mobile unit to the cell center, follow these steps:

First successful decoding of a synchronization message, i.e., CRC verification;

Second loading the long PN traversal systems according to the state received in the synchronization message; and

Third compensating for the phase shift of the pilot code if the reception is from a sector that uses an offset pilot signal.

By performing these steps, the mobile station is fully synchronized, so that both PN synchronization and real time synchronization have been performed and transmission can be started on both the access channel and the audio channel.

The mobile unit can make a call if it can send a signal to make a call to another system user through a cellular exchange. In the case of communication from the mobile unit to the cell center, access is achieved using time-slotted ALOHA techniques. For example, the reverse link data rate may be 4800 bps. A signal packet of the access channel consists of a preamble and subsequent information.

In the exemplary embodiment, the preamble length is an integer multiple of the 20 ms timeframe and is a sector or cell parameter received by the mobile unit in a paging link message. Because cell-center receivers determine propagation delays from the preamble, this method allows the preamble length to be varied as a function of the cell radius. In the access channel, the user PN codes are either predefined or transmitted through the paging channel to the mobile unit.

For the duration of the preamble, modulation is fixed and constant. The orthogonal W used in the preamble_She waveform is a Walsh function consisting of zeros only. The convolutional encoder then produces the desired W_She signal, if you add a series of zeros to its input.

An access channel signal packet may consist of one or up to two 20 msec time frames. For the access channel, the encoding, interleaving and modulation are the same as for the audio channel at 9600 bit / s. In one embodiment, the sector or cell requests the mobile unit to provide a 40 msec preamble and the access channel message requires a data frame. If N_P the number of preamble time frames and k the number of 20 msec intervals from a predetermined start time, then mobile stations can only initiate transmission on the access channel if (k, N_{P + 2}) = 0 equation.

For other communications applications, it may be desirable to rearrange some elements of the error correction coding (orthogonal sequence encoding and PN encoding) to adapt to the characteristics of the particular application.

For example, in satellite mobile communications, where signals are transmitted by one or more satellites between terrestrial hubs and mobile units, it may be desirable to use coherent modulation and demodulation techniques for both directions of communication, since these channels are more phase-coherent than terrestrial mobile. channels. For such applications, the mobile modulator does not use the m-order encoding described above. Instead, two-phase or four-phase modulation or forward error correction code signals may be used, with conventional coherent demodulation. The carrier phase can be derived from the received signal using Costas loop technique. In addition, the application of the orthogonal Walsh function to the channel signal can be used in the same way as described for the connection from the cell-site to the mobile unit. As long

As long as the phase coherence of the channel is acceptable, this modulation and demodulation technique provides operation with less interference than m-th orthogonal signal transmission, thus increasing the capacity of the system.

In another embodiment, it may be advantageous to encode the speech waveform directly into an RF signal rather than using vocoder and FEC techniques. Although the quality of the connection is very good when using the vocoder and FEC techniques, the implementation is very complicated, which increases costs and consumption. This is especially unfavorable for portable handhelds, where power consumption and cost are important considerations. In conventional digital telephone systems, the speech waveform in digital format consists of eight-bit speech samples of 8 kHz sample rate. The CDMA system can encode eight-bit samples directly into carrier phase angles. Thus, no vocoder or FEC encoder / decoder is required. However, for good quality operation, a slightly higher signal-to-noise ratio must be provided, resulting in a reduction in capacity. In another embodiment, the 8-bit samples are coded directly into carrier amplitudes. In yet another alternative, the speech waveform is encoded into carrier phase angles and amplitudes.

Based on the above description of preferred embodiments, the present invention will be readily apparent to those skilled in the art. These embodiments can be modified in many ways without departing from the general principles defined herein and without the need for further inventive activity. Thus, the present invention is, of course, not limited to the embodiments disclosed herein, but should be interpreted as broadly as possible in accordance with the general principles and novel features described.