CN1212094A - Time compressing transponder - Google Patents

Abstract

A method and apparatus for supporting calls between two mobile stations in a satellite communication system are disclosed. First, when a signal transmitted by a first mobile station using a narrowband transmission format is received at a satellite relay station, the received signal is sampled and digitized. Then, the sampled and digitized signal is stored in a buffer at a first rate. The stored data is then read from the buffer at a rate higher than the first rate and modulated onto a downlink frequency to produce a wideband transmission format. The modulated signal is then transmitted to the second mobile station.

Description

time compression repeater

Field of Invention

The present invention relates to mobile wireless communication systems utilizing orbiting satellites or relay stations, and more particularly to methods and apparatus for supporting mobile-to-mobile calls via satellite relay stations.

Background of the Invention

The prior art contains several examples of duplex wireless communications through the use of Frequency Division Multiple Access (FDMA), where each distinct radiotelephone has a unique pair of frequencies for transmission in both the transmit and receive directions, e.g., the US AMPS cellular phone system. The prior art also discloses duplex wireless communication systems through the use of time division multiple access (TDMA), wherein each distinct radiotelephone has a unique time slot on a first common frequency for communicating in one direction, and There is another unique time slot on a second shared frequency for communicating in the other direction, eg the European GSM digital system or the US digital cellular standard IS-54. In these systems, the time slots in each direction are further offset in time from each other so that the portable radiotelephone does not need to transmit and receive simultaneously. This eliminates the need for transmit/receive duplex filters, which are required for radiotelephones operating in FDMA systems. Alternatively, so-called "time-duplex" phones use simpler transceiver switches, as provided in the existing European cellular system GSM, to alternately couple the antenna to either the receiver or the transmitter.

In some applications, neither TDMA nor FDMA provides an optimal solution. TDMA systems require higher peak transmit power to compensate for compressing transmissions into a time slot that is only a fraction of the total time, since it is the average power that limits communication range and quality. It goes without saying that in any case the base station must have sufficient transmitter power to support all mobile stations, while the total power is the same for FDMA and TDMA solutions. TDMA base stations have a high powered transmitter and an antenna, are simpler and less expensive, and are time-shared between all base station/mobile station links through time division multiplexing (TDM). However, TDMA mobiles generate high peak power, which is generally inconvenient. On the other hand, it is inconvenient for FDMA mobile stations to use antenna duplex filters. Therefore, the present invention seeks to provide a method that combines the use of TDM on the base station to mobile station link (downlink) with FDMA on the mobile station to base station link (uplink) while avoiding the need for dual tool method.

The prior art discloses hybrid TDM/FDMA systems, eg the British Army PTARMIGAN Single Channel Radio Access System (SCRA). The SCRA system is actually a military radiotelephone system that uses TDM on the downlink on a first frequency band and on the uplink by assigning separate frequencies in a second frequency band to each mobile station for uplinking. Use FDMA. However, the SCRA system requires separate antennas for uplink and downlink, respectively, or a duplex filter to allow simultaneous transmission and reception through the same antenna.

Figure 1 shows the prior art transmission format described in the US digital cellular standard IS-54. The base station sends information continuously in 20ms long data frames. The data involved consist of digitized speech information generated by a digital speech compression algorithm, interspersed with synchronization, signaling and control symbols. Each 20 ms data frame is divided into three time slots, each time slot containing information intended for one of the three mobile stations. Thus, a particular mobile station only needs to switch on its receiver one-third of the time because data for that particular mobile station is limited to one of the three time slots making up the frame. In the opposite direction, the 20 ms frame is likewise divided into three time slots. Each mobile transmitter uses only one of the two time slots it is not receiving, leaving the other third of the time, which can be used to scan for other base station frequencies to determine if the reception is stronger to another base station. On the uplink channel, these signal strength measurements are reported to the current base station, which makes a decision whether to hand off communications with the mobile station to a stronger base station. The use of signal strength measurements made by mobile stations for handoff decisions is referred to as "mobile assisted handoff" (MAHO).

In this prior art system, it can be seen that the mobile transmits only one-third of the time available, so must use three times the peak power that would be sufficient if continuous transmissions were employed. If serial transmission is used, all three mobile stations will overlap in time and must therefore be given different channels, as in the British Army's PTARMIGANSCRA system. Also, a transmit/receive duplex filter is required in the mobile station in order to allow simultaneous transmission and reception.

There are currently many proposals for launching satellites in orbit capable of supporting communications with mobile stations or handheld phones. Although a large percentage of calls are between fixed (PSTN or cable) subscribers and satellite terminals, a small percentage of calls are between paired satellite terminals. In the latter case, it is desirable to avoid double delays in the propagation of the signal from one terminal to the satellite, relayed by the satellite to the terrestrial network switch, from the switch back to the satellite, and finally from the satellite to the second terminal. In this way, the signal travels through four times the Earth-satellite distance, increasing voice latency. Therefore, it is desirable to provide a method of interconnecting mobile terminals via a single satellite relay path by using mobile-to-mobile transponders. When using the asymmetric TDMA or TDMA/FDMA system of the invention disclosed in the patent application, the prior art transponder on the satellite will repeat a signal received by using a transmission format of a mobile terminal, the signal will be Received in an incorrect format by the second mobile terminal.

For mobile-to-mobile calls, if the satellite transponder is only programmed to transparently reflect signals received from a first mobile to a second mobile, and vice versa, the mobile also needs to be able to receive narrowband TDMA signals, which increases their complexity. Alternatively, if the satellite receives narrowband TDMA signals, decodes them, and then re-encodes them into wideband TDMA format for retransmission, then the very large capacity required on the satellite to achieve mobile-to-mobile The amount of processing may become excessive. Thus, there is a need for a method and apparatus for supporting mobile-to-mobile calls in a satellite communication system that do not significantly increase the complexity of the satellite relay station or the mobile station.

Public Summary

It is an object of the present invention to overcome the above-mentioned disadvantages of the prior art by providing an innovative method and arrangement for supporting mobile-to-mobile calls in a satellite communication system.

According to one embodiment of the present invention, a method and apparatus for supporting a call between two mobile stations in a satellite communication system are disclosed. First, when a signal transmitted by a first mobile station using a narrowband transmission format is received at a satellite relay, the received signal is sampled and digitized. The sampled and digitized signal is then stored in a buffer at a first rate. The stored data is then read from the buffer at a rate higher than the first rate and modulated on the downlink frequency to produce a broadband transmission format. The modulated signal is then sent to the second mobile station.

According to another embodiment of the present invention, a satellite transponder for supporting a call between two mobile stations in a satellite communication system is disclosed. The transponder comprises receiving means for receiving signals from the first transport format. The received signal is sampled and digitized in the sampling and digitizing means and stored in the buffer at a first rate. The stored signal is then read from the buffer at a rate higher than the first rate and modulated at the downlink frequency in modulating means to produce a broadband transmission format. The transmitting means then transmits the modulated signal to the second mobile station.

Brief description of attached drawings

The invention will now be described in more detail with reference to preferred embodiments of the invention given by way of example only and shown in the accompanying drawings, in which:

Figure 1 shows a prior art TDMA format;

Figure 2 shows a TDM/FDMA format transmitted with overlapping mobile stations according to one embodiment of the present invention;

Figure 3 shows a TDM/FDMA hybrid format according to one embodiment of the present invention;

Figure 4 shows a TDM/FDMA hybrid format according to one embodiment of the present invention;

Figure 5 shows the application of the present invention to satellite communications with a large number of time slots;

Figure 6 shows a frequency reuse plan for a 3-cell;

Figure 7 shows a block diagram of a portable radio according to one embodiment of the present invention;

Figure 8 shows a base station for one embodiment of the invention;

Figure 9 shows satellite/mobile communication in one embodiment of the present invention;

Figure 10 shows a satellite transponder according to one embodiment of the present invention;

Figure 11 shows the satellite transponder from the ground station to the mobile station;

Figure 12 shows the satellite transponder from the mobile station to the central station;

Figures 13a-b show additional components for providing direct mobile-to-mobile forwarding according to one embodiment of the present invention.

Detailed Description of Preferred Embodiments

In a TDMA communication system in which mobile assisted handoff is not required, but simultaneous transmission and reception are avoided in order to eliminate the 3-slot time slot of the transmit/receive duplex filter, the present invention expands the transmit duty factor from one-third to two-thirds, which cuts the peak power requirement in half. In Figure 2 the uplink and downlink formats according to the invention are shown.

As shown in Figure 2, at any one time two of the three mobile station transmissions overlap. In order to allow their overlapping in time, they have to be made orthogonal, ie non-interfering with each other, in some other domain (eg frequency domain). Since the transmission data rate is halved by taking twice as long to transmit, by arranging one transmission to use the upper half of the allocated bandwidth and the other to use the lower half of the allocated bandwidth (or vice versa), it is possible to Two transmissions can be accommodated within one bandwidth. For example, a first mobile station may use the upper half of the channel bandwidth, while in the middle of its two-thirds transmission interval, a second mobile station may start transmitting in the lower half of the channel bandwidth. Then, during another one-third of the frame period, the first mobile station will finish using the upper half of the channel, and the third mobile station can begin transmitting on the upper half of the channel. During yet another one-third period, the second mobile station will finish using the lower half of the channel, and the first mobile station can begin transmitting again. However, the first mobile station will operate on the lower half of the channel instead of the upper half on which it was originally operating. This problem only arises when using an odd number of time slots in combination with the channel bandwidth being divided into even parts, and it can be solved by either of the two methods described below.

To avoid two mobile stations using the upper and lower half of the channel respectively during two thirds of the time, and a third mobile station using the upper half channel during its first third of the time, and then in its solution for switching to the lower half of the channel during the second third of the time, since the goal of any solution should be that all mobiles function equally independently of time slots. Preferably, if the frequency switch is to occur in the middle of a transmission burst in any one mobile station, it also occurs in all mobile stations, so that the system has a consistent design.

Figure 3 shows an embodiment of the invention in which the first mobile station receives the first third of the base station transmissions, and then during the first third of its two-thirds transmission interval by using the uplink The upper half of the channel is sent to the base station. After the first third of the transmission interval, the first mobile station switches frequency in order to use the lower half of the channel during its second third of the transmission. Simultaneously, the second mobile station receives the 40 ms frame of the second third of the base station and starts transmitting on the upper half channel when the first mobile station switches to the lower half channel. Then, when the second mobile station switches to the lower half of the frequency channel in the middle of its transmit burst, the third mobile station begins transmitting on the upper half frequency frequency channel. When the third mobile station switches to the lower half channel, the first mobile station starts transmitting again on the upper half channel. Frequency switching from the upper half channel to the lower half channel in the middle of a burst is preferably accomplished not by rapidly switching frequency synthesizers but by applying system phase rotation to the transmitted signal, thus facilitating the provision of secondary channel Positive or negative frequency offset from center. This can be done within a digital signal processor that is used to generate the modulated waveform, as will be discussed below.

In Fig. 4 is shown a second embodiment of the invention which avoids a frequency shift in the middle of a burst. Here, the first mobile station first transmits on the upper half of the frequency channel, and halfway through its two-thirds transmission frame, the second mobile station starts transmitting by using the lower half of the frequency channel. Then, halfway during the transmission cycle of the second mobile station, the first mobile station ends transmission, and the third mobile station starts transmission by using the upper half channel. Halfway during the transmission period of the third mobile station, the second mobile station ends its transmission on the lower half of the channel. At this point, the first mobile station again starts transmitting on the lower half of the channel, which is the opposite of the channel it was using previously. In this embodiment, each mobile station acts the same, but alternates between transmissions on the upper and lower half channels on successive bursts. In this system, a one-third receive interval of 13.3 ms between successive bursts is available for changing the transmit frequency, which can be accomplished at a moderate rate of frequency change using a frequency synthesizer.

It will be appreciated that the invention is not limited to systems with three time slots. When an even number of time slots (eg four) is used and the mobile stations transmit three-quarters of the time in the frame, the transmissions of the three mobile stations overlap in frequency at some point in time. In such a case, the channel bandwidth is divided into three, and then each mobile station uses the three sub-bands sequentially. Alternatively, one mobile station could transmit at 1/2 the bandwidth 2/4 of the time, while the other three use other combinations of half the frame period and half the channel bandwidth.

The above-mentioned solution is generally characterized in that the uplink channel width is divided into a number of sub-bands, which is at least a number smaller than the number N of downlink time slots. For example, the three-slot case divides the channel into upper and lower halves, while the four-slot case divides the channel into three sub-bands. This is consistent with the reduction of the bit rate by a factor of (N-1) when the transmitter operates in N-1 slots instead of only one. When N is small, it is difficult to divide the number into N subbands while reducing the bit rate not by N but only by N-1. For example, in a 3-slot system, it would be difficult to accommodate half the bit rate transmission by transmitting only 2/3 of the time instead of 1/3 of the time in only 1/3 of the bandwidth. However, this difficulty disappears when N is large.

Figure 5 shows an embodiment of the invention which can be advantageously used for satellite-mobile communications. In this embodiment, a 512-slot TDM downlink and a 512-subband FDMA uplink are combined. To avoid the use of duplex filters in the mobile station, the signal for transmission is compressed into the 511/512 time remaining after the mobile station has received its 1/512 downlink TDM format. However, the 0.2% increase in the information rate does not prevent it from being accommodated in 1/512 of the bandwidth. Without this signaling format, the mobile station must transmit and receive at the same time, so a duplex filter must be used, resulting in unwanted signal loss; otherwise the mobile station uses TDMA on the uplink, including the mobile station at 1/ Using 512 times the peak power to transmit in 512 hours would result in an undesired increase in the peak power and current required from the power supply. In this case, the invention of course also allows sending without serious difficulties in the time of 510/512 or even less, the compression of information is not limited to clearing a time slot for reception, If there are other requirements for timing (such as providing guard time between sending and receiving).

In the case of satellite communications, it would also be advantageous to consider other hybrid schemes of TDMA and FDMA or even CDMA for the uplink. Satellites whose orbital latitudes are lower than those of geostationary satellites exhibit large velocities relative to stationary or moving terminals on the ground. This results in Doppler shifts when receiving from ground terminals at the satellite, which are large compared to the narrow transmission bandwidth of a pure FDMA uplink. Therefore, it is sometimes desirable to increase the bandwidth of the uplink so that the Doppler shift is relatively insignificant without reducing the capacity of the system. An increase by a small factor, eg, 2, 4 or 8, may often be sufficient. One way to achieve a 2:1 increase in bandwidth while accommodating the same number of transmitters would be to compress each uplink transmission on one of the 256 available subbands to 256/512 of the time; two transmissions in each sub-band are then passed by the transmitter of the first mobile station using 1/512 of the time slots numbered 1 to 256 and the transmitter of the second mobile station using the slots numbered 257 to 256 512 time slots are accommodated by TDMA. The first mobile station receives on time slot 257 and the second mobile station receives on time slot 1, for example, thus avoiding the requirement to be able to receive and transmit simultaneously. This principle can be extended to 4 slots in each of the 128 sub-bands, or 8 slots in each of the 64 sub-bands, etc. However, the mobile peak transmitter power must be increased because its duty cycle is reduced by using more TDMA and less FDMA on the uplink.

Bandwidth can instead be extended by using Code Division Multiple Access (CDMA) on the uplink. In CDMA, each original information bit is sent multiple times with or without polarity inversion according to the bits of the access code. For example, by sending the sequence B1B1B1B1 using the access code 1100 and replacing the original bit B1, a four-fold increase in bandwidth is achieved; B2 is also replaced by B2B2B2B2, etc., giving a four-fold increase in bit rate. By using a different access code, preferably an orthogonal code, such as 1001, another mobile station may be allowed to transmit to overlap it. The other mutually orthogonal codes are 1111 and 1010, resulting in four overlapping non-interfering transmissions sharing a four times wider sub-band. This achieves a four-fold increase in uplink signal bandwidth, which is desirable so that higher transmitter power for the mobile station is not required and Doppler shift is relatively insignificant while preserving capacity.

Capacity in cellular telephone systems or mobile-satellite communication systems depends on being able to multiplex a limited number of allocated frequencies for more than one session. The service area covered is usually divided into a number of cells, each of which is served by a base station (or illuminated by a satellite antenna spot beam). Ideally, it should be possible to tightly utilize the entire allocated spectrum in each neighboring cell, however, this is often not possible due to interference from neighboring cells using the same frequency. As a result, frequency reuse planning must be used to control interference levels. For example, a so-called 3-cell frequency reuse plan can be used, as shown in FIG. 6 . 3-Cell frequency reuse planning ensures a certain minimum desired signal-to-interference ratio (C/I), which can be satisfied if appropriate error-correcting coding is applied to the transmitted signal. In general, for a 3-cell reuse plan, C/I is better in the satellite case than in the terrestrial cellular case, because the sidelobes of the satellite's cell illumination pattern are sharper outside the cell, compared to the signal strength for terrestrial propagation The decrease comes faster with distance.

A problem arises when applying a frequency reuse plan to the TDM downlink. The limited allocated spectrum must be divided into thirds to allow 3-cell reuse planning. As a result, the bandwidth of a full TDM solution can no longer be accommodated. According to one aspect of the present invention, this problem is solved by using a time reuse plan instead of a frequency reuse plan in the TDM downlink in combination with a corresponding frequency reuse plan in the FDMA uplink.

In the time reuse plan, the cells marked "1" in Fig. 6 are illuminated using the full available spectrum using the first 1/3 of the time slots of the TDM format either from the satellite or from their respective terrestrial base stations. Then, the cell marked "2" receives illumination during the second 1/3 of the TDM format, and so on. In this way, adjacent cells are not illuminated with the same frequency at the same time, but the full TDM signal bandwidth is still transmitted. For example, in a 512-slot TDM format, cells marked "1" are illuminated during the first 170 slots. After receiving its respective 1/512 time slot, each mobile station may transmit in the remaining 511/512 frames by using a designated one of the first 170 of the 512 uplink FDMA channels. The cells marked "2" are then illuminated in a second set of 170 time slots of 512 time slots, and the corresponding mobile stations in those cells are illuminated by using FDMA and corresponding uplink frequencies 171 to 340 answer. The cells marked "3" then become illuminated in the third set of 170 time slots of 512 time slots and their mobile stations reply on uplink frequencies 341 to 510. The remaining two slots may be reserved to illuminate all cells with specific signals for paging and call setup. Likewise, two corresponding unused uplink channel frequencies can be reserved in case a mobile station wishes to initiate a connection with the system by performing a so-called random access.

By using the above system with a time reuse plan on the downlink combined with a matching frequency reuse plan on the uplink, the described TDM/FDMA hybrid access method can be utilized while controlling adjacent Interference level between cells.

As previously stated, it is desirable to extend the uplink channel bandwidth by employing TDMA or CDMA on the FDMA uplink and appropriately reducing the number of FDMA channels in order to provide increased Doppler shift tolerance. Those of ordinary skill in the art will appreciate that the actual numbers used above are exemplary and are not meant to limit the invention to those examples.

Figure 7 shows a preferred embodiment of a mobile or portable wireless device suitable for use in the present invention. Antenna 10 operating on uplink and downlink frequencies is alternately connected to receiver 30 and transmit power amplifier 120 through a transmit-receive (T/R) switch 20 controlled by a TDM timing generator 50 . In an alternative, transmit/receive duplex filters may be used if the uplink and downlink frequencies are sufficiently separated to allow the use of simple filters with low loss. When the uplink and downlink frequencies are widely separated, a single antenna may not be economical, in which case separate transmit and receive antennas may be necessary. However, this does not change the principle of the invention to avoid transmitter operation at the moment of reception.

The timing generator provides timing and control pulses to switch 20, receiver 30 and digital demodulator and decoder 40 to provide power to them to select signals on assigned time slots on the downlink. Receiver 30 has sufficient bandwidth to receive the entire TDM downlink signal spectrum, but only one time slot per TDM frame period of this bit rate stream is selected for processing in digital demodulator and decoder 40 . During this selected time slot, the signal from the receiver is digitized in the A-D converter 31 and recorded in a buffer included in the demodulator 40 . The digitization technique preferably preserves the complex vector properties of the signal, for example by separating the real (I) and imaginary (Q) parts by means of quadrature mixers and then digitizing each part. An alternative to this so-called I,Q, or Cartesian method is the LOGPOLAR method described in US Patent No. 5,048,059, assigned to the same assignee and incorporated herein by reference. Another alternative is the so-called homodyne or zero-IF receiver, as described, for example, in US Patent Application No. 07/578,251, incorporated herein by reference.

The complex vectors recorded in the buffer are then processed by the digital demodulator and decoder 40 for the remainder of the frame period before the next time slot's complex signal samples are collected. The demodulation processing stage may include channel equalization, or echo cancellation, to mitigate the effects of multipath propagation. Typical algorithms suitable for this are described in US Patent Application Nos. 07/964,848 and 07/894,933, assigned to the same assignee, which patents are incorporated herein by reference.

To help account for bridge fading, the error correction coded data frame can be spread over more than one time slot by means of interleaving, so that multiple time slots must be collected and to interweave. The demodulation of the signal samples in each time slot should preferably be optimized in conjunction with an error correction decoding algorithm for best performance at low signal-to-noise ratios, e.g. by combining The soft decision information is passed to the decoder, or by combining demodulation and decoding in a so-called decoder, as described in U.S. Patent Application No. 08/305,727 , which is incorporated herein by reference.

After demodulation and error correction decoding by using, for example, a convolutional decoder based on soft decisions, the value (worth) of the frame of the speech data of the error correction decoding is transmitted to the speech encoder/decoder 60, where Where it is converted to PCM speech samples at 8 ksamples/second by using a decoder matched to the encoder at the originating transmitter. Speech coding/decoding techniques can be, for example, residual pulse-excited linear predictive coding (RELP) or codebook-excited linear predictive coding (CELP), which compress 8 thousand samples/second PCM speech signals to 4 kbit/s in the transmitter , and conversely, the 4 kbit/s signal from the decoder 40 is expanded again into a PCM voice signal of 8 thousand samples/second for D-A conversion and audio amplification in the D-A converter 130 to drive the earphone 132.

In principle, the receiver only needs to receive a single frequency, ie the frequency on which all signals from all mobile stations are multiplexed and modulated. As a result, the receiver does not have to tune to alternate frequencies, but chooses between all available time slots. The controlling microprocessor 110 receives information on call/paging slots during call setup, allocating time slots to be used for the call. The controlling microprocessor 110 then programs the timing generators in turn to generate all the control pulses necessary to turn the receiver and transmitter on and off in accordance with the inventive TDM/FDMA hybrid format described herein. Control microprocessor 110 also programs transmit combiner 90 to generate, by means of upconverter 80, a narrowband uplink frequency channel associated with the assigned downlink time slot. Upconverter 80 can work in several ways. First, the upconverter 30 can generate a sum frequency or a difference frequency at the desired transmit channel frequency by mixing a fixed modulation frequency (TXIF) with a variable frequency generated by a programmable frequency synthesizer, wherein the The sum or difference frequency is selected by the filter. In an alternative, the upconverter 80 can generate a difference frequency by mixing the signal from the voltage controlled oscillator with the synthesizer frequency, and comparing it to a fixed modulation frequency in the phase error detector, then the phase The error is amplified and applied to the VCO to lock it to the modulated TXIF, so that the VCO phase follows the phase modulation on TXIF. The decision on which method to choose depends on whether the selected transmit modulation method is phase-only modulation (ie, constant amplitude modulation), or whether the selected modulation includes a varying amplitude component.

In the opposite direction, the speech signal from the microphone 131 is first amplified and converted to 8 ksamples/sec PCM by using the A-D converter 13, and then compressed to a reduced rate by using the speech encoder 60. Speech compression techniques such as RELP and CELP, which compress speech down to 4kbit/s, typically run 40ms of speech sample frames at a time. A frame is typically compressed to 160 bits and then error correction encoded in a digital encoder 70 before being modulated to radio frequency. The radio frequency being modulated may be a fixed intermediate frequency locked to a precise reference oscillator 100 . Then, the signal is up-converted to the final uplink frequency signal by mixing with the transmit combiner 90 in the up-converter, and then it is amplified by the transmit power amplifier 120 and transmitted to the antenna 10 by the switch 20 . The digital encoder and modulator 70 includes buffering (and, if used, interleaving) functionality to compress the transmission into the time remaining available after receiving the downlink slot, thus implementing the present invention This aspect, thus avoiding simultaneous sending and receiving. The transmission can also be compressed to 1/4 or 1/8 of the frame period if a frame with 4 or 8 uplink time slots is desired. The timing of this compression, modulation, and power amplifier turn-on and turn-off at appropriate times is also controlled by timing generator 50 to achieve coordination between transmit and receive timing.

In some applications, the receiver may have to be able to tune to alternate channel frequencies and select between TDM time slots on those channels. In this case, receiver 30 also includes a frequency synthesizer programmed by the controlling microprocessor and locked to the precision of the reference frequency oscillator. When a call is set up, the assigned frequency is given on the calling or paging channel.

In terrestrial radio applications using "push-to-talk" operation, the mobile station may be part of a group or network of communicators sharing a relay wireless system with other groups. In a trunked system, all idle wireless devices listen to the call setup channel. When the wireless device transmits by activating the talk switch, a short message is sent on the corresponding call setup uplink channel requesting channel allocation. The receiving base station network immediately answers on the downlink call setup channel, giving the currently free frequency/slot assignment, and the mobile terminal then uses this assigned free frequency/slot for the rest of its transmissions. When the "push-to-talk" switch of the transmitting radio is released, the end of the message signal is transmitted to prompt the network of base stations and other team members to revert to idle mode where they listen to the call setup channel. The process is fast and automatic, within a fraction of a second, so it is completely invisible to the operator.

In cellular or satellite phone applications, an idle mobile terminal listens to a specific time slot/frequency assigned by a calling/paging channel. Also, transmissions on calling/paging channel time slots may be further submultiplexed to form infrequently repeating time slots, each of which is designated, for example, by the last few digits of their respective telephone numbers. specific group of mobile stations. These so-called sleep mode groups are only paged on specific submultiplex time slots, and the control microprocessor 110 can recognize it from the received data, and thus can program the timing generator 50 so that only at these times wakes up the receiver, which results in a considerable savings in supply current consumption during standby.

Furthermore, the digital demodulator/decoder 40, after processing each newly received time slot, can produce an estimate of the receiver's frequency error due to reference oscillator inaccuracies and Doppler shift , this frequency shift can be large in satellite systems. By using broadcast information from the satellites, microprocessor 110 can correct for Doppler shift and determine errors due only to the reference oscillator. The microprocessor 110 can then correct the error by sending a correction signal, such as a tuning voltage, to the oscillator to ensure that the transmit frequency generated by the transmit frequency synthesizer 90 with reference to the reference oscillator is accurately generated. The process of correcting for Doppler shift involves determining position or orientation relative to the satellite orbit by using all or any of the following parameters: rate of change in measured Doppler shift; satellite and antenna beam identification signal; broadcast information on satellite instantaneous three-dimensional coordinates and Doppler precompensation (if used); previous mobile terminal location; elapsed time since last position estimate; and mobile terminal velocity.

In addition to the frequency correction mechanism described above, the demodulator also produces information about the position of the signal samples in the buffer that are considered to correspond to known sync symbols, which in turn produces information about the accuracy of the timing produced by the timing generator 50. information. Microprocessor 110 performs a clean check of this information and, if deemed correct, uses it to require a small timing correction to timing generator 50 to correct any drift.

Figure 8 shows a block diagram of an embodiment of a base station suitable for use with the present invention. The common antenna 210 is connected to a receiver low noise amplifier 230 and a transmit power amplifier 260 through a duplex filter 220 . The low noise amplifier delivers the entire uplink frequency band to the FDMA channel receiver bank 240 . After digitizing the signal in each channel by using one of the complex vector digitization techniques described above, the signal is processed in the received signal processing means group 520 to perform demodulation and equalization, error correction decoding, and speech decoding. The resulting 8 kilosamples/second are then time multiplexed using standard digital telephony standards (eg T1 format for convenient connection to a digital switch 280 such as an ERICSSONAXE switch).

An alternative to the analog implementation of the FDMA receiver block is to digitize the entire composite signal and process the signal digitally to separate out the individual FDMA signals. This is true as long as the signal strength difference between the signals is not too large for the A-D converter dynamic range. Another aspect of the invention is the inclusion of power control means to limit the signal level difference between different FDMA signals in order to facilitate the digital implementation of the receiver filter bank, potentially simplifying the base station. The proposed power control means may be based on an assumption by the mobile station of the relationship between the strength of the signal received by the mobile station from the base station and the strength of the signal received by the base station from the mobile station. Thus, the mobile station contributes to the increase in signal strength received from the base station via the satellite by reducing its transmit power, and vice versa. This is done by a slower power control arrangement at the base station which includes up/down power control information in the signaling samples which are interleaved with voice symbols in each mobile's time slot.

Switch 280 according to call setup information, requested routing, or pre-set information, in the following three (i.e.: uplink signal received from satellite phone, call received from public switched telephone network, or used in downlink Choose between signals transmitted from the operator or the control room) on the computer. Switch 280 provides the selected signals multiplexed together in accordance with some known digital telephony relay format (eg T1) and passes the signals to transmit DSP group 270. Transmit DSP group 270 encodes each voice signal in the multiplexed data stream separately by using a voice compression algorithm such as RELP or CELP. The transmit DSP group then encodes the signal for error correction and remultiplexes the signal into downlink TDM format for modulation to downlink radio frequency by using modulator 290 and amplification by use of high power amplifier 260 . Switch 280 also extracts the call setup information for the uplink call channel and inserts the call setup information corresponding to the paging slot into the downlink TDM format. This information is recognized as data rather than speech to the respective DSP device, thus bypassing RELP encoding and instead being subjected to a more powerful form of error correction encoding.

Terrestrial systems may further include separate antennas and associated receive signal processing to implement space diversity reception to improve range and overcome fading. Depending on signal quality, the combination of the signal processed by the remote antenna and the signal processed by antenna 210 can be done with demodulation and equalization algorithms, or by means of simple diversity selection on a speech frame-by-speech frame basis. Likewise, in the transmit direction, a second remote transmitter may receive a signal from transmit DSP group 270 for transmission on the same frequency for improved area coverage. The mobile station receiver shown in Figure 7 by means of its equalized demodulator algorithm is able to perceive the delayed signals received from the second transmitter as the echo of the first transmitter and can use these signals in order to improve the reception . Satellite diversity can also be used to improve communication reliability, as disclosed in US Patent Application No. 08/354,904, which is incorporated herein by reference.

Fig. 9 shows a block diagram of a satellite communication system according to one embodiment of the present invention. The orbiting satellite 410 communicates with at least one ground station or outstation called a central station (HUB) 400 , and with a plurality of portable mobile telephones 420 . Each telephone is served by appropriate antenna beams from multiple spot beam antennas on the satellite, which provide high gain in the direction of each telephone. The HUB communicates with the satellite using eg C-band or Ka-band frequencies and the satellite communicates with the phone using eg L-band (uplink) or S-band (downlink) frequencies. In most cases, most calls will be between satellite phones and regular phones belonging to the PSTN. The HUB station receives calls from the PSTN and relays them to the mobile phones via the satellite, and conversely receives calls from the mobile phones and relays them from the satellite and connects them to the PSTN. A small percentage of calls could be mobile to mobile, and the HUB could order the satellite transponders to relay them directly to each other without involving the PSTN. In some systems, two or more HUBs located in different parts of the earth communicate with the same satellite. In such cases, mobile-to-mobile calls may involve HUB-to-HUB connections, which may be accomplished over international trunks, which may be part of the PSTN system. Alternatively, the satellite-hub link could allocate some capacity to hub-to-hub communication via satellite, thereby avoiding the high price of terrestrial lines due to its presence.

According to one embodiment of the present invention, mobile-to-mobile calls can be directly relayed through satellite relay stations without using ground-based HUB stations, in order to avoid delays and additional charges. A satellite transponder 500 according to this embodiment of the invention is shown in FIG. 10 . Signals transmitted by one of the mobile stations to the other are received in receiver 510 at the satellite. The received signal is then sampled and digitized in A-D converter 520 and stored in buffer 530 at a first rate. The stored signal is then read from the buffer at a faster rate and modulated onto the downlink frequency in modulator 540 to create a wideband transmission format.

Sampling and digitization of the uplink signal can be done onboard the satellite in either case as long as the transponder is of the type known as digital processing payload. Such transponders can digitize the entire uplink bandwidth for subsequent use of digital filters or fast Fourier transform techniques to divide the bandwidth into sub-bands, or down to individual uplink channel frequencies altogether. Transponders can also use digitized signals to perform digital beamforming, as described, for example, in the paper entitled "Efficient Apparatus for simultaneous Modulation and digital Beamforming for an Antenna Array" filed on December 7, 1995 effective apparatus for arrays)” in U.S. Patent Application No. 08/568,664, which is hereby incorporated by reference. To form a mobile-to-mobile repeater according to the present invention, one or more narrowband uplink channels are segmented using digital band segmentation and optionally digital beamforming to receive signals to be relayed directly to other mobile stations. mobile signal. However, such directly relayed signals are preferably also relayed to the HUB station, which remains responsible for instructing the mobile station transmitters to adjust their power or timing and for accruing charges to be paid by the user for using the system.

When such a digital processing payload is used as described above, the signal received from the mobile station is digitized by using any of the aforementioned techniques described in the cited references, and processed in the processor. The HUB station provides synchronization signals on the C- or Ka-band feeder link to determine (when the satellite MS to MS processing unit) the uplink frame period and slot timing (if TDMA is used on the uplink ), during the uplink frame period, samples from the mobile station will be digitized, processed, and collected in the memory of the processor. The number of samples per frame to be gathered in processor memory will correspond to an uplink time slot of the signal duration. The HUB station will also monitor the signal received from the mobile station via the C- or Ka-band feeder link, and if necessary, issue a timing adjustment command to control the transmission timing of the mobile station, so that the signal is correctly aligned and timed on the satellite. gap is received.

The collected time slot values for mobile uplink transmissions are then read from memory on a higher frequency clock and converted to satellite-to-mobile downlink through downlink processing, beamforming and satellite transponder devices link frequency such that it is used in accordance with the principles of the parent patent application for associating frequency channels and time slots of one TDMA format (e.g., narrowband TDMA) with frequency channels and time slots of a different (e.g., wideband) TDMA format. The downlink time slots associated with the uplink frequency channels and time slots are transmitted. Thus, time compression of the relay signal is performed. The bit allocation of the signal within the TDMA burst is unaffected, nor is the modulation characteristic affected by this time compression, only the scaling of time and bandwidth is affected, so that the relayed signal will now have the same Match the signal bandwidth and slot length optimized for machine operation. A mobile assigned a direct mobile-to-mobile transponder channel changes, if necessary, within an uplink burst compared to the bit allocation used for communication via the HUB station or with the PSTN. bit allocation. For example, cited patent application Ser. No. 08/305,727 describes the advantages of using distributed pilot symbols along with a specific bit interleaving technique for efficient communication over narrowband channels. However. Mobile station receivers receive wideband channels and process only short time slots in which signal changes due to fading are negligible. In the case of wideband channels, it is more desirable to use non-distributed or "blocked" pilot symbols in the form of known "sync words", which are centrally placed within the downlink time slots. Therefore, whenever a direct mobile-to-mobile channel is allocated, the mobile uplink transmission is changed to suit the pilot symbol and interleaving bit configuration of the downlink format.

The preferred format for satellite communications for dual-mode phones operating on satellite systems or GSM terrestrial cellular systems uses a satellite downlink with the same 200 kHz channelization as GSM and a burst width TDMA format at the same symbol rate (but the frame period of this TDMA format is twice or four times longer than the GSM "full rate" frame period, so it has 16 or 32 slots compared to GSM's 8). In the uplink direction, the preferred format for mobile station transmissions is a narrower-band format using 50 kHz channelization and correspondingly fewer time slots (4 or 8) in order to reduce the peak power versus average power transmitted by the mobile station ratio. In the uplink direction, the burst width is chosen to be four times the GSM burst width, and with an exact 1/4 symbol rate, the format is easily adopted by the mobile station transmitter by using the built from off-the-shelf components. The format of the present invention is described in detail in U.S. Patent Application No. 08/501,575 entitled "Dual Mode Satellite/Cellular Phone," which is incorporated herein by reference.

The preferred format includes a superframe structure comprising 12 repeating 16-slot TDMA frames plus a 13th frame containing Slow Associated Control Channel information (SACCH). The SACCH slots contain messages from the HUB to the mobile station to enable the HUB station to command the mobile station transmitters to adjust their power or transmission timing in particular. When a mobile terminal is directly connected to another mobile terminal by using the time-compressed transponder of the present invention disclosed herein, it is appropriate that the SACCH order be transmitted from the HUB to the mobile station in question instead of requiring a mobile A mobile station has the ability to generate commands to another mobile station. Moreover, if two directly connected mobiles try to control each other's transmission timing, their absolute timing will not be controlled to any system reference and will drift in an uncontrolled manner, which can occur in the same manner as using There is a risk of timing collisions with other mobiles in other time slots. This risk can be avoided by using the time compressed transponder of the present invention to replace the samples in every 13th time slot relayed to the mobile station with the SACCH signal samples received from the HUB station via the C- or Ka-band feeder link. The uplink SACCH bursts received at the satellite from the mobile terminals continue to be relayed to the HUB station, so that the two-way SACCH message channel between each mobile station and the HUB station it controls even in the direct mobile station It is still held in the communication mode of the mobile station. In this way, the SACCH time slot initiated by the HUB provides an absolute timing reference to the mobile station, and the mobile station timing its transmission relative to the reference, avoiding the aforementioned drift problem. Timing is also maintained during voice silence, when voice traffic slots are not transmitted in order to save satellite or mobile terminal battery power. Transmission of periodic signal bursts to maintain synchronization during discontinuous transmission (DTX) intervals is described in issued US Patent No. 5,239,557, which is incorporated herein by reference.

Therefore, the mobile-to-mobile repeater preferably includes the steps of: receiving narrowband signals from the mobile station and relaying them to the HUB station; digitizing some of the received narrowband signals and time compressing them for sending back in wideband format Other mobile stations; multiplex the time-compressed signal together with other signals received from the HUB station that are already in broadband format (such as SACCH messages for the mobile station or voice or data for PSTN calls to the mobile station) into broadband Downlink TDMA format. The multiplexed signal forms the downlink TDMA signal structure, which is then modulated onto a downlink carrier frequency. The modulated signal is then sent by the transmitter to the second mobile station. Thus, by buffering the received signal and reading the signal from the buffer at a faster rate, it is possible to support mobile-to-mobile calls through the satellite relay without significantly increasing the complexity of the satellite relay or the mobile. For mobile-to-mobile calls, the downlink power level is increased to compensate for dual-path radio noise, since the uplink noise is not removed by error correction decoding on the satellite.

Figures 11 and 12 show a satellite communication payload suitable for one embodiment of the present invention. Figure 11 shows the downlink to the mobile phone, while Figure 12 shows the uplink from the mobile phone. Referring now to FIG. 10, the antenna 360 receives a plurality of signals from the HUB and, using the receiver bank 340, causes them to be demodulated or down-converted. The receiver output signal is then coherently upconverted in an upconverter bank 320 by mixing with a common local oscillator 330 . The upconverted signal is now at the downlink frequency and is amplified by a bank of power amplifiers 310, where each amplifier is coupled to an element, group of elements or feeds of a multibeam antenna or phased array. In one embodiment of the invention, the amplifier is a class C transmit power amplifier operating at maximum efficiency. In one embodiment of the invention, the satellite transmitter includes a saturated traveling wave tube. In this way, by sending appropriate signals to the satellite antenna 360, the HUB can decide what signal will be transmitted by the multi-beam antenna and in what direction at what time. In this way, it can be determined, for example, that in any particular time slot in a downlink TDM format, only a subset of regions on the ground receiving the signal are sufficiently separated along the boresight angle so that they are not affected by each other. interference. In this way, independent signals can be sent to a phone at each time slot in each region without interference. In the next time slot, different zone groups (ie those sub-zones between the first zone group) are illuminated so that all zones receive signals from certain time slots of the frame. Pending U.S. Patent Application Publication No. 08/179,953, entitled "ACellular/Satellite Communication System With Improved Frequency Re-use," filed January 11, 1994 How one-to-one multiplexing is used for this embodiment, where each time slot is used for all multiple sub-regions, is hereby incorporated by reference.

When the system is operating at less than full capacity, not all time slots in the frame are active. Also, at any one time, half of the conversation between the two parties is usually silent, which would benefit from momentarily switching off the signal in the corresponding time slot. When the number of slots is large (ie, 512), it is assumed that only about 50% are active simultaneously, which is statistically accurate. The power amplifier 310 is designed to draw little or no current during periods of inactivity so that the average consumption of the satellite's main power supply, even at full load, is only half of the peak power consumption of the power amplifier. For a solar array of a given size, the power amplifier peak power can thus be rated at twice the value that the solar array would otherwise support.

Also, peak capacity is only reached at certain times of the day, while the solar array is converting solar energy to electricity 24 hours a day. By using a rechargeable battery to average the power consumption over 24 hours, a further factor increase in peak transmitter power relative to the continuous load that the solar array can support can be obtained. The advantage of the TDM downlink used in the present invention is that the reduction in current consumption is directly proportional to the underutilization factor, unlike FDMA or CDMA downlinks which use power amplifiers whose current consumption is only proportional to the underutilization factor decreases by the square root of . Therefore, the use of TDM downlinks allows to obtain the full benefit of the average underutilization factor.

The active slots of any TDM signal can be taken together so as to occupy adjacent slots in a subframe which is part of the TDM frame period. The inactive time slots make up the rest of the TDM frame period. The subframes of any TDM signal transmitted in one beam of the multi-satellite antenna beam do not overlap with the subframes of TDM signals transmitted in adjacent beams.

Referring now to FIG. 12, a multi-beam antenna or multi-element phased array 400 receives signals from multiple mobile stations on the uplink. Mobile stations in the same area on the ground using different FDMA channel frequencies on the uplink do not transmit according to the invention when they are in the TDM downlink receive slots. Mobile stations in different regions use the same set of frequencies as mobile stations in the first region, so antenna 400 receives multiple signals on each FDMA channel coming from different directions. In the case of a multi-beam antenna (such as a paraboloid with spaced apart feeds), different directions correspond to different beams, so that signals of the same frequency appear in different beams and can thus be separated. This may require that adjacent beams do not contain the same frequency, but that an appropriate reuse factor is used, eg a three-to-one frequency reuse pattern as shown in Figure 6. The use of the three-to-one time reuse pattern on the downlink described above will automatically result in a three-to-one frequency reuse pattern on the uplink when the uplink FDMA channel is associated with the corresponding TDMA time slot , so that the signals are separated. On the other hand, for the uplink, by using the configuration of Fig. 11, especially when the antenna 400 is a phased array, a one-to-one frequency reuse pattern can be achieved.

Antenna 400, whether a multi-feed parabola or a multi-element phased array, presents multiple RF ports containing multiple mobile station uplink signals. Low noise amplifier bank 410 and downconverter bank 420 amplify these signals and coherently downconvert them by using a common local oscillator 470 to a suitable intermediate frequency for amplification and filtering. The downconverted filtered and amplified signals are then applied to an upconverter bank or transmitter modulator bank 430 which convert the signals to C or Ka band while preserving their phase relationship before applying them to combiner 440 , and then amplified in the traveling wave tube TWT power amplifier 450 for sending to the HUB station through the antenna 460. It should be noted that the antenna 460 in FIG. 12 can be the same antenna as the antenna 360 in FIG. 11, and then the C/Ka-band receiver is separated from the transmitter by a duplex filter. Furthermore, two polarizations can be used in two directions in order to improve bandwidth utilization. Each polarization is then associated with half of the receiver set 340, and half of the transmitter set 430 connected to a separate traveling wave tube. Moreover, the downlink antenna 300 and the uplink antenna 400 can also be the same in principle, and for each beam, array element, or sub-array, a transmit/receive duplex filter is added, thus achieving the same antenna Oral and facial dual use.

The corresponding A description of the HUB station equipment of , this patent is hereby incorporated by reference.

In order to give the satellite transponder the ability to relay mobile-to-mobile calls directly, some processing on the satellite other than that shown in Figures 11 and 12 is provided. Figures 11 and 12 represent the simplest and most flexible form of a multibeam satellite transponder, which relays all signals received by each satellite receive antenna unit to a HUB station for processing. When the receiving antenna is, for example, a multi-element phased array, the combination of the element signals forming the directional beam is realized at the HUB station, not on the satellite, and the signals do not even need to be digitized on the satellite. In order to select a mobile station signal on a satellite for direct relay to another mobile terminal, however, directionality must be established on the satellite. A narrowband uplink channel must also be selected that contains transmissions from mobile stations that are to be forwarded directly to other mobile stations. Digital beamforming may not be required if the satellite receiving antenna forms its directivity for different receive beams by means of a parabolic reflector with multiple feeds. It is also possible to provide multiple narrowband receiver channels by using analog filter hardware which, under remote commands from the HUB station, can be switched to connect to a selected beam or tuned to select a different channel.

Alternatively, the satellite receiving antenna may be a phased array, and the array element signals must be combined on the satellite in order to form a directional receiving beam. This can be done by digitizing the signals and providing them to a digital beamformer for processing by using multiple beamforming coefficients. The coefficients can also be received via remote commands from the HUB station, wherein the coefficients can be adjusted in order to maximize the signal-to-noise-plus-interference ratio according to the method described in U.S. Patent Application No. 08/179,953, which is published in This quote is for reference. When there is such digitization and digital processing for beamforming, it is also logical to use digital channelization, which uses digital filters to limit the bandwidth in order to select a single uplink channel.

Figures 13a-b show the addition of the mobile-to-mobile channel of the present invention to a repeater such as that of Figures 11 and 12 . The transponders of Figures 11 and 12 are configured to support PSTN to mobile station connections without digital processing on board the satellite. This can be accomplished using the composite analog time complex method disclosed in U.S. Patent Application No. 08/225,389, filed April 8, 1994, entitled "Large Deployable Phased Array Satellite." connected feeder links, and have the ability to provide different beamwidths as disclosed in co-filed U.S. Patent Application No. 08/225,399, entitled "Multiple Beamwidth Array", the two The patent is hereby incorporated by reference. Alternatively, the satellite transponder may have a fully digital processing payload capability that performs digital channelization and/or beamforming on the satellite. With either form of PSTN-to-mobile transponder calibration, the purpose of the addition required to construct the mobile-to-mobile transponder of the present invention is to extract some of the uplink signals received from the mobile for use in Perform filtering and time compression according to the principles of the present invention disclosed herein, and then additionally re-insert the time-compressed signal to multiplex with other voice, data, or SACCH signals received from the HUB via the feeder link receiver 340 , for sending to the mobile station.

Figures 13a-b show signals received from a mobile terminal by using a receiving antenna unit connected to a transponder 1000, which may be in the same embodiment as shown in Figure 12 . In addition to forwarding the receiving unit signal to the HUB station through the feeder link, the receiving unit signal is also fed to the digital signal processing unit 1001 after amplification and frequency conversion, and the latter performs analog-to-digital conversion and digital channel filtering (in order to separate the selected uplink channel frequency), and digital beamforming (if the antenna design does not provide enough spatial selectivity). Digital beamforming and digital channelization may be performed in either order. In the case of constituting different sets of beam directions for different uplink frequencies, digital channelization is performed first, and then beamforming is performed for each frequency by using a different set of beamforming coefficients for each frequency. When the described beam characteristics are the same for all frequencies, digital beamforming may be performed first. It is also possible to partially beamform the subbands before further dividing them into individual uplink channels, and then further beamforming to define sets of offset directions for each frequency channel and time slot.

The signals separated by channel frequency (channel K) and direction (beam i) are stored as samples in a buffer 1002 connected to the digital processing unit. The samples corresponding to the desired uplink frequencies and beams are selected by the control and timing unit 1003, which also selects samples in the desired uplink time slots, which are timed by using the feeder link receiver 340 The synchronization signal received from the HUB station takes the HUB as a reference. Selected samples of signals currently corresponding to the first mobile station occupying a particular uplink time slot, a particular uplink carrier, and a particular (beam) direction of arrival, are transmitted to the corresponding mobile-to-mobile transmit Beamformer 1005, where they are processed to form antenna element signals that will produce directed transmit beams in the desired direction, ie towards a second mobile station with which the first mobile station wants to communicate. Samples are read from the transmit digital beamformer 1005 at times determined by the timing control unit 1003 such that they will be transmitted on the allocated downlink time slots. Since the downlink time slots are shorter than the uplink time slots, samples are read from the beamformer 1005 and sent to the D-A converter 1004 at a proportionally faster rate. Furthermore, the complex number of samples can be coincident with a progressive angular rotation corresponding to a desired frequency shift, which also optionally includes an adjustment due to satellite orientation, in order to ensure that transmissions occur on the desired downlink carrier frequency. or pre-compensation for Doppler shift caused by the movement of the mobile station away from the destination. The D-A-converted, time-compressed, and frequency-shifted signal is added to other signals received from the HUB station by the feeder link receiver 340, however, in the time slots allocated for the mobile-to-mobile downlink And the channel will not receive the signal from the HUB. The hub station ensures this by not generating a corresponding feeder link signal at that moment. However, the HUB station may present the specific time slot received from the mobile station via the control and timing unit 1003, which corresponds to its SACCH time slot selected from the buffer, so that the SACCH time slot will not be added in the adder 1006, so There will be no forwarding from mobile station to mobile station. Instead, the HUB station fills the downlink SACCH slot with SACCH information to be sent to the mobile station.

The above describes how a single mobile station signal is forwarded to a second mobile station. The hardware of Figure 13 can perform the same function simultaneously for different beams or frequencies and time slots for the reciprocal direction and for many different such mobile terminals, up to a total equal to the processing provided for direct mobile-to-mobile calls capacity. In the case that the capacity of direct mobile station to mobile station calls may be temporarily insufficient, mobile terminals can still forward their signals to the HUB station through the usual double-hop method, and after passing through the mobile station exchange or PSTN switch them It is forwarded back to other mobile stations through the satellite, thereby being connected to other mobile stations. This would have the disadvantage of a double propagation delay, which the present invention seeks to overcome. However, a mobile initially assigned a double-hop path due to a temporary lack of direct mobile-to-mobile forwarding capacity may be queued and assigned a path with Direct mobile-to-mobile link with low latency. This queuing function is completed by the switching computer of the ground network, and the HUB station uses the SACCH channel to issue the so-called "internal switching" command, and a pair of mobile stations are switched from a double-hop connection to a single-hop connection. Internal handover involves sending a control message to the mobile station informing it to change modes, channels, or time slots, and may be due to other channel management reasons (such as minimizing co-channel interference, or avoiding time slot collisions, as well described in the cited literature) and is performed at any time for the above purposes.

Those skilled in the art or those of ordinary skill in the art will appreciate that the number of time slots, frequency bands and applications described above are mainly for illustrative purposes and do not imply any limitation on the present invention. This patent application embraces any and all modifications that are within the spirit and scope of the invention as disclosed and claimed.

Claims (12)

1. method that is used for being supported in satellite communication system two callings between the travelling carriage may further comprise the steps:

By using the narrow band transmission form to transmit a signal to satellite relay station from first travelling carriage;

The signal that sampling and digitlization receive from described first travelling carriage;

With first rate with described signal storage that sampled and digitized in buffer;

Read the signal of described storage with the speed higher from described buffer, and the signal of described storage is modulated on the down-link frequencies, so that set up the wideband transmit form than first rate;

Described modulation signal is sent to second travelling carriage;

Receive and decipher described modulation signal at described second travelling carriage.

2. satellite repeater that is used for being supported in satellite communication system two callings between the travelling carriage comprises:

Be used to receive receiving system by the signal that uses the transmission of narrow band transmission form;

Be used to sample and the device of the signal of the described reception of digitlization;

Be used for device with described that sampled and digitized signal of first rate storage;

Be used for reading the signal of described storage and the signal of described storage being modulated at down-link frequencies, so that set up the device of wideband transmit form with the speed higher than first rate;

Be used for described modulation signal is sent to the emitter of second travelling carriage.

3. method that is used for being supported in satellite communication system two callings between the travelling carriage may further comprise the steps:

At the narrow band signal of satellite relay station reception from described travelling carriage;

The narrow band signal that is received is relayed at least one ground station;

The narrow band signal of some reception of digitlization;

The described digitized signal of time compression;

Is the digitized signal of described time compression broadband downstream link form together with other signal multiplexing from described at least one ground station;

The signal of described multiple connection is sent in the described travelling carriage one.

4. according to the method for claim 3, it is characterized in that wherein said other signal from least one ground station is the SACCH signal.

5. according to the method for claim 4, it is characterized in that wherein said SACCH signal is according to the narrow band signal of reception that is relayed to described at least one ground station and calculated.

6. according to the method for claim 4, it is characterized in that wherein said SACCH signal controlling is in the emission timing of described travelling carriage.

7. satellite repeater that is used for being supported in satellite communication system two callings between the travelling carriage comprises:

Be used at the device of satellite station reception from the narrow band signal of described travelling carriage;

Be used for the narrow band signal that is received is relayed to the device of at least one ground station;

The device that is used for the narrow band signal of some reception of digitlization;

The device that is used for the described digitized signal of time compression;

The device that to be used for the signal of described time compression be broadband downstream link form together with other signal multiplexing from described at least one ground station;

Be used for the signal of described multiple connection is sent to one device of described travelling carriage.

8. according to the satellite repeater of claim 7, it is characterized in that wherein said other signal from least one ground station is the SACCH signal.

9. according to the satellite repeater of claim 8, it is characterized in that wherein said SACCH signal is according to the narrow band signal of reception that is relayed to described at least one ground station and calculated.

10. according to the satellite repeater of claim 8, it is characterized in that wherein said SACCH signal controlling is in the emission timing of described travelling carriage.

11. method that is used for being supported in two callings between the travelling carriage in satellite communication system, described satellite communication system has at least two mode of operations, promptly be used for first mode of operation that relay station via satellite directly connects travelling carriage and be used for second mode of operation that relay station and ground station via satellite directly are connected travelling carriage, may further comprise the steps:

Receive the setup requests that connects described two travelling carriages;

Determined whether that enough capacity are used for by using first mode of operation to connect two travelling carriages;

When enough capacity, connect two travelling carriages by use first mode of operation;

When determining not have enough capacity, connect two travelling carriages by use second mode of operation;

The described calling of lining up so that in case when enough capacity is arranged by using first mode of operation to connect.

12. the method according to claim 11 is characterized in that, wherein the propagation delay when described first mode of operation is less than the propagation delay when described second mode of operation.

Images